Clause Labs

Blog

  • Confidentiality Clauses vs NDAs: When to Use Which (and Why It Matters)

    Confidentiality Clauses vs NDAs: When to Use Which (and Why It Matters)

    Confidentiality Clauses vs NDAs: When to Use Which (and Why It Matters)

    A confidentiality clause buried in Section 14 of your MSA is not the same thing as a standalone NDA signed before the first meeting. Yet lawyers and business professionals use these terms interchangeably every day, and the confusion creates real gaps in protection. According to Clio’s 2025 Legal Trends Report, contract review remains one of the highest-volume tasks for solo and small firm lawyers — and confidentiality provisions appear in virtually every agreement that crosses your desk.

    The distinction matters because choosing the wrong instrument at the wrong time can leave your client’s trade secrets, financial data, or proprietary processes exposed. Here’s how to decide which one fits, when you need both, and the drafting differences that actually affect enforceability.

    Upload your NDA or agreement now and get an AI-powered confidentiality analysis in under 60 seconds — free, no signup required.

    They Solve Different Problems

    A standalone NDA (non-disclosure agreement) is a self-contained contract whose entire purpose is governing the exchange of confidential information between parties. It typically runs 2-5 pages, defines confidential information in detail, lists standard exclusions, specifies duration, addresses remedies, and includes provisions for return or destruction of materials.

    A confidentiality clause is a provision embedded within a larger agreement — an MSA, employment contract, vendor agreement, or partnership agreement. It usually occupies 1-3 paragraphs and relies on the host agreement’s broader framework for remedies, termination, and dispute resolution.

    The core question is timing and context:

    • NDAs govern the sharing of sensitive information before a broader agreement exists
    • Confidentiality clauses govern information exchanged during an existing contractual relationship

    This distinction drives everything else: scope, enforceability, duration, and remedies.

    When a Standalone NDA Is the Right Choice

    Use a standalone NDA when no other agreement governs the relationship between the parties. The most common scenarios:

    Pre-deal discussions. Before an acquisition, investment round, or strategic partnership, parties need to share financial data, customer lists, technology specifications, and business strategies. No MSA or operating agreement exists yet. An NDA is the only protection. According to the ABA’s guidance on confidentiality obligations, lawyers have an independent duty under Model Rule 1.6 to protect client information — but that doesn’t extend to the other party’s obligations. You need a contract.

    Due diligence periods. M&A due diligence involves reviewing financials, litigation history, IP portfolios, and operational data. A standalone NDA typically includes specific provisions for data room access, permitted disclosures to advisors, and post-termination data destruction obligations that wouldn’t fit in a confidentiality clause.

    No existing contract. Two companies exploring a potential vendor relationship need to share technical requirements and pricing models. No purchase agreement or MSA exists yet. A standalone NDA fills the gap.

    Employee onboarding (combined with invention assignment). Many companies use a CIIA — Confidential Information and Inventions Assignment Agreement — which functions as a standalone NDA combined with IP assignment provisions. This is a better approach than embedding confidentiality in the employment agreement alone, because the CIIA can survive employment termination with its own specific terms.

    Detailed terms required. When you need comprehensive definitions, specific exclusion carve-outs, detailed remedies (including injunctive relief), and specific return/destruction procedures, a standalone NDA provides the space and structure to address each element properly.

    When a Confidentiality Clause Is Sufficient

    A confidentiality clause works within an existing agreement that already governs the relationship. Common situations:

    MSAs and vendor agreements. The MSA already includes provisions for term, termination, remedies, governing law, and dispute resolution. A confidentiality clause leverages these existing provisions rather than creating a parallel framework. As noted in Bloomberg Law’s analysis of confidentiality agreements, the choice between standalone and embedded protections often depends on the complexity of the broader relationship.

    Employment agreements. Confidentiality is one of several employment terms alongside compensation, duties, termination, and benefits. A well-drafted confidentiality section within the employment agreement covers the basics. But note: many practitioners recommend a standalone CIIA in addition to (or instead of) the employment agreement’s confidentiality section, because it survives on its own terms.

    Standard commercial transactions. When confidentiality is not the primary concern of the agreement — it’s just one of many provisions — a clause is more proportionate and practical.

    Short-term engagements. For a 30-day consulting project with limited information sharing, a standalone NDA may be overkill. A confidentiality clause within the consulting agreement handles it efficiently.

    The test: if the broader agreement’s termination, remedies, and survival provisions adequately cover confidentiality, a clause is sufficient. If you need confidentiality protections that differ from or extend beyond the host agreement’s framework, use a standalone NDA.

    When You Need Both

    The answer is often “both, in sequence.” Here’s the typical pattern:

    1. Pre-deal NDA governs disclosures during negotiations, due diligence, and deal evaluation
    2. MSA with confidentiality clause replaces or supplements the NDA once the deal closes

    The critical issue: what happens to the NDA when the MSA takes effect? If you don’t address this explicitly, you create ambiguity about which document controls. The MSA’s integration clause (“this agreement constitutes the entire agreement between the parties”) may inadvertently terminate the NDA — including protections that applied to pre-signing disclosures.

    Best practice: address the transition explicitly. Include language in the MSA such as:

    “The Mutual Non-Disclosure Agreement dated [date] between the parties shall survive execution of this Agreement with respect to Confidential Information disclosed prior to the Effective Date. For Confidential Information disclosed on or after the Effective Date, Section [X] of this Agreement shall govern.”

    Without this, you’ll have a dispute about which protections apply to which disclosures — and your client’s pre-deal disclosures may end up with less protection than intended.

    Key Drafting Differences

    The structural differences between NDAs and confidentiality clauses aren’t just about length. They reflect different levels of detail, different default assumptions, and different enforcement mechanisms.

    Element Standalone NDA Confidentiality Clause
    Length 2-5 pages 1-3 paragraphs
    Definition of “Confidential Information” Detailed, often 1-2 pages with categories, markings requirements, and oral disclosure protocols Abbreviated — often a single sentence or brief paragraph
    Standard exclusions All 5 standard exclusions typically listed (publicly available, independently developed, previously known, received from third party, required by law) Sometimes truncated to 2-3 exclusions
    Duration Specified independently (typically 2-5 years, sometimes indefinite for trade secrets) Often tied to the host agreement’s term plus a survival period
    Remedies Detailed — injunctive relief, specific performance, prevailing party attorneys’ fees References the host agreement’s general remedies section
    Return/destruction of materials Detailed procedures, certification requirements, timeline Often absent or addressed in 1 sentence
    Non-solicitation Sometimes included as a companion provision Rarely included within a confidentiality clause
    Survival Explicit survival terms independent of any other agreement May or may not survive host agreement termination — check carefully

    The practical takeaway: a confidentiality clause that omits standard exclusions, lacks an independent survival period, or doesn’t address remedies is significantly weaker than a standalone NDA. If you’re drafting a clause, make sure it covers at least the exclusions, duration, and survival — don’t assume the host agreement fills those gaps automatically.

    The 5 Mistakes That Create Exposure

    Mistake 1: Using only a confidentiality clause when a standalone NDA is needed. If you’re sharing sensitive information before any agreement exists, a clause inside a not-yet-signed contract provides zero protection. The NDA must be signed first.

    Mistake 2: Having both an NDA and a confidentiality clause that conflict. The NDA says confidential information means “all information disclosed in writing.” The MSA clause says “all information, whether oral or written.” Which controls? If the MSA’s integration clause supersedes the NDA, the more restrictive definition may apply to pre-deal disclosures — shrinking your protection.

    Mistake 3: Forgetting to address supersession. The most common mistake. The ABA’s Model Rules on competence (Rule 1.1) require lawyers to apply thorough and adequate preparation to every representation. Failing to address the NDA-to-agreement transition is a preparation gap.

    Mistake 4: Confidentiality clause that’s too short. A single sentence — “Both parties agree to keep the other’s information confidential” — is technically enforceable but practically useless. No definition, no exclusions, no duration, no remedies. If challenged, you’ll spend more litigating what the clause means than the information was worth.

    Mistake 5: NDA that’s too broad. An NDA that defines confidential information as “all information of any kind” with no exclusions is likely unenforceable in many jurisdictions. Courts have struck down overly broad NDAs as unreasonable restraints. Specificity matters.

    For a deeper look at NDA-specific pitfalls, see our analysis of common NDA mistakes across 1,000 agreements.

    How Duration and Survival Differ

    Duration is where standalone NDAs and confidentiality clauses diverge most significantly in practice.

    Standalone NDA duration is typically set independently: 2-3 years for general business information, 5 years for technology or financial data, and indefinite for trade secrets. The duration runs from the date of disclosure, not the date of the agreement.

    Confidentiality clause duration is usually tied to the host agreement: “during the term of this Agreement and for [X] years thereafter.” This creates a problem: if the MSA runs for 3 years and the confidentiality survival is 2 years, information disclosed in Year 1 is only protected for 4 more years after disclosure. Information disclosed in Year 3 gets the full 2-year survival. The protection is inconsistent.

    Best practice for clauses: Specify that the confidentiality obligation survives for a fixed period from the date of each disclosure, not from termination of the host agreement. This provides consistent protection regardless of when the information was shared.

    How Clause Labs Reviews Both

    Whether you’re reviewing a standalone NDA or a contract with an embedded confidentiality clause, Clause Labs’s AI analysis identifies the key provisions and flags gaps:

    • Detects whether confidentiality protections exist (standalone, embedded, or both)
    • Flags missing standard exclusions in either format
    • Identifies when duration is too short or absent
    • Checks for adequate remedies provisions
    • Detects potential conflicts between an existing NDA and a new agreement’s confidentiality terms
    • Flags missing return/destruction obligations

    For lawyers who review NDAs regularly, our guide to reviewing NDAs in 10 minutes provides a structured workflow that pairs well with AI-assisted review.

    Frequently Asked Questions

    Can a confidentiality clause fully replace an NDA?

    Yes — but only if the clause is comprehensive enough and the host agreement is already signed before any confidential information changes hands. If information is being shared before the agreement is executed, you need a standalone NDA for the interim period. A well-drafted confidentiality clause within a signed MSA can provide equivalent protection to a standalone NDA, but most clauses in practice are far less detailed and therefore provide weaker protection.

    Should I sign an NDA before every business meeting?

    No. NDAs should be reserved for situations involving genuinely sensitive information: proprietary technology, financial data, customer lists, strategic plans, or trade secrets. Routine business discussions about potential partnerships, general pricing conversations, or publicly available information don’t warrant NDAs. Over-using NDAs creates “NDA fatigue” — where parties start treating them as formalities and stop reading them carefully. That’s worse than having no NDA at all.

    What happens if my NDA and my MSA have different confidentiality terms?

    The MSA’s integration clause will likely control, meaning the NDA may be superseded entirely — including its protections for pre-signing disclosures. This is why addressing the transition explicitly is critical. Without a carve-out preserving the NDA for pre-deal disclosures, you may lose protections you assumed were still in place.

    Is a verbal NDA enforceable?

    In theory, oral agreements can be binding under general contract law principles. In practice, a verbal NDA is nearly impossible to enforce because you can’t prove what information was designated as confidential, what obligations were agreed to, or what the duration was. Always use a written agreement. The Statute of Frauds may not specifically require NDAs to be in writing in most states, but proving the terms of an oral NDA in litigation is prohibitively difficult.

    How long should confidentiality obligations last?

    It depends on the type of information. General business information: 2-3 years. Technology specifications, financial data, or strategic plans: 3-5 years. Trade secrets: indefinite (as long as the information qualifies as a trade secret under applicable state law, such as the Uniform Trade Secrets Act). The key is matching the duration to the shelf life of the information’s competitive value. A 1-year NDA protecting a 5-year product roadmap is inadequate.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    Wondering whether your NDA or confidentiality clause has gaps? Upload any agreement to Clause Labs’s free analyzer — no signup required — and get an instant risk analysis identifying missing exclusions, weak duration terms, and remedies gaps. Solo practitioners reviewing 25+ agreements monthly can upgrade to the Solo plan at $49/month for full redline suggestions and DOCX export with tracked changes.

  • Auto-Renewal Clauses: How to Protect Your Clients from Silent Lock-Ins

    Auto-Renewal Clauses: How to Protect Your Clients from Silent Lock-Ins

    Auto-Renewal Clauses: How to Protect Your Clients from Silent Lock-Ins

    Your client signed a three-year software contract with a 90-day notice window for non-renewal. Nobody calendared the deadline. The contract auto-renewed for another three years at a 12% price increase. Total cost of missing that window: $216,000 in unwanted fees for software the client stopped using eight months ago.

    This happens constantly. According to a Kaplan Group analysis, an estimated 99% of B2B auto-renewal clauses are not written in compliance with current state and federal regulations — which means both sides of the table are exposed. Auto-renewal clauses are designed to exploit a predictable human behavior: we forget. The vendor counts on it. Your job as the reviewing lawyer is to make sure the clause is fair, the deadline is calendared, and your client has a real exit option.

    This article breaks down the mechanics, the red flags, the rapidly expanding state regulatory landscape, and six specific negotiation strategies you can use in your next contract. If you want to check whether a contract’s auto-renewal provisions create a trap, upload it to Clause Labs’s free analyzer — it flags long notice periods, one-sided renewal terms, and buried price escalation in under 60 seconds.

    How Auto-Renewal Clauses Work

    The mechanics are straightforward. An auto-renewal clause provides that the contract automatically extends for successive periods — usually one year — unless one or both parties provide written notice of non-renewal within a specified window before the current term expires.

    The three moving parts:

    The notice window. The period during which you must send your non-renewal notice. Typically 30-90 days before the renewal date. Miss this window by even one day and the contract auto-renews.

    The renewal term. How long the contract renews for. This can match the initial term (a three-year contract renewing for another three years) or default to a shorter period (a three-year contract renewing for successive one-year terms). The difference is enormous.

    The renewal terms. Does the contract renew on the same terms, or can one party change pricing, SLAs, or other conditions upon renewal? A clause that says “renews at Vendor’s then-current rates” is an open-ended price escalation mechanism.

    Here is what a typical auto-renewal looks like in practice:

    “This Agreement shall automatically renew for successive one (1) year terms unless either party provides written notice of non-renewal at least sixty (60) days prior to the expiration of the then-current term.”

    Looks reasonable. But consider: if your client’s three-year initial term expires on March 31, the non-renewal notice must be sent by January 30. If the lawyer reviewing the contract in April of year one doesn’t calendar that date, no one will remember it 33 months later.

    Auto-Renewal Red Flags

    Flag these immediately during any contract review:

    Notice Period Over 60 Days

    A 90-day notice window means your client must decide whether to renew three full months before the renewal date. For a complex vendor relationship, the client may not have the data to make that decision 90 days out — they’re still evaluating the current year’s performance.

    Some contracts push this to 120 or even 180 days. A six-month notice window on a one-year contract means the client effectively has six months to evaluate before they must decide. That’s barely half the contract term.

    Notice Sent to a Specific Person or Address

    If the clause requires notice to “the attention of [Name], Director of Contract Administration, at [specific address],” your client now has an additional trap: the notice must reach the right person at the right place. If that person has left the company or the address has changed, the notice may be argued as defective.

    Renewal Term Equals Initial Term

    A three-year initial term that auto-renews for another three years creates a six-year commitment from a single missed deadline. Market-standard practice for multi-year contracts is to auto-renew for successive one-year terms, not for the full initial term length.

    Price Increases Without a Cap

    “Services shall renew at Vendor’s then-current pricing” gives the vendor unlimited discretion to raise prices at renewal. Without a cap (e.g., “not to exceed 5% per annum” or “not to exceed CPI increase”), your client’s costs could spike dramatically on renewal.

    No Termination for Convenience During Renewal Term

    If the contract includes termination for convenience during the initial term but removes that right during renewal terms, your client is more locked in after renewal than before.

    Auto-Renewal Buried in Dense Text

    Auto-renewal language should appear in a clearly labeled section — “Term and Renewal” or equivalent. When auto-renewal language is buried in the middle of a paragraph in a miscellaneous section, the drafter may be counting on nobody reading it. Our guide on reviewing contracts for red flags covers this pattern and 24 others.

    No Notification Before Renewal

    The most client-hostile version: the contract auto-renews silently with no obligation for the vendor to remind the client that the renewal date is approaching. Several states now require affirmative notification before auto-renewal takes effect.

    State Auto-Renewal Laws: A Rapidly Expanding Landscape

    Auto-renewal regulation has accelerated significantly in the past three years. What started as consumer protection legislation is increasingly extending to B2B contracts.

    California

    California’s Automatic Renewal Law (CARL), codified in Business and Professions Code §§ 17600-17606, was strengthened by amendments that took effect July 1, 2025. Key requirements:

    • Clear and conspicuous disclosure of auto-renewal terms before the consumer agrees
    • Affirmative consent to the auto-renewal terms (not just the overall agreement)
    • Easy cancellation in the same medium used to enroll
    • Records of consent maintained for at least three years

    CARL applies primarily to B2C relationships, but according to analysis by Upcounsel, the law can affect B2B agreements when one side is deemed a small business or “consumer-like” entity.

    New York

    New York General Obligations Law § 5-903 applies directly to auto-renewal provisions in B2B service contracts. It requires that the auto-renewal provision be “clear and conspicuous” and that the service provider give notice of the renewal terms between 15 and 30 days before the renewal date. Non-compliant auto-renewals are unenforceable.

    Virginia

    Virginia amended its consumer protection law effective July 1, 2024, creating requirements for both B2C and B2B auto-renewal contracts. Notably, Virginia treats “small businesses” as consumers under the auto-renewal statute, extending B2C protections to qualifying B2B relationships. Virginia also allows a private right of action for violations.

    Illinois

    Illinois’s enhanced auto-renewal law, effective January 1, 2022, requires cancellation instructions in renewal notices and an online cancellation option for consumers who subscribed online. However, Illinois explicitly excludes B2B contracts from its auto-renewal regulations.

    Federal (FTC)

    The FTC’s “Click-to-Cancel” amendments to the Negative Option Rule were initially set to take effect in May 2025 but were delayed to July 14, 2025. The Eighth Circuit subsequently vacated the rule in July 2025. However, the FTC retains enforcement authority under ROSCA (Restore Online Shoppers’ Confidence Act) and the FTC Act, and restarted rulemaking in January 2026. The key takeaway for practitioners: even without the specific rule, aggressive enforcement of auto-renewal practices continues.

    Summary Table

    State Applies to B2B? Pre-Renewal Notice Required? Key Requirement
    California Limited (consumer-like entities) Yes Affirmative consent, easy cancellation
    New York Yes Yes (15-30 days before renewal) Clear and conspicuous disclosure
    Virginia Yes (small businesses) Yes Private right of action
    Illinois No (B2B excluded) Yes (B2C only) Online cancellation option
    Federal (FTC) Rule vacated; enforcement continues Per ROSCA standards ANPRM pending (2026)

    For a comprehensive breakdown by state, see Faegre Drinker’s auto-renewal state law charts.

    6 Strategies to Negotiate Auto-Renewal Provisions

    Strategy 1: Remove Auto-Renewal Entirely

    Replace auto-renewal with affirmative renewal — the contract expires at the end of the term unless both parties agree in writing to extend. This eliminates the missed-deadline risk entirely.

    When it works: When your client has leverage (they’re the customer, they have alternatives), or when the contract value is high enough that both sides should consciously decide to continue.

    Sample language:

    This Agreement shall expire at the end of the Initial Term. The parties may
extend this Agreement for successive one (1) year Renewal Terms by mutual
written agreement executed at least thirty (30) days prior to the expiration
of the then-current term.

    Strategy 2: Reduce the Notice Period

    If auto-renewal stays, shorten the notice period. Push 90-day windows down to 30 days. Push 60-day windows to 30.

    Market justification: A 30-day notice period is sufficient for most commercial contracts and is the standard in many SaaS agreements. Anything over 60 days benefits only the party who drafted the clause.

    Strategy 3: Shorten the Renewal Term

    If the initial term is three years, negotiate for one-year renewal terms. This limits the exposure from a missed deadline to one additional year rather than three.

    This Agreement shall automatically renew for successive one (1) year terms
(each a "Renewal Term") unless either party provides written notice of
non-renewal at least thirty (30) days prior to the end of the then-current term.

    Strategy 4: Cap Price Increases

    Add a ceiling on any price increase at renewal.

    Upon each Renewal Term, Vendor may increase the fees by no more than the
greater of (i) 3% or (ii) the percentage increase in the Consumer Price Index
(CPI-U) for the twelve-month period ending on the date sixty (60) days prior
to the renewal date.

    Without this cap, a renewal clause is a blank check for the vendor.

    Strategy 5: Require Vendor Notification Before Renewal

    Add a mandatory reminder obligation:

    Vendor shall provide Customer with written notice at least forty-five (45) days
prior to the renewal date, identifying the upcoming renewal date, the applicable
renewal term, and any changes to fees or terms. Failure to provide such notice
shall extend the non-renewal notice period by thirty (30) days from the date
Vendor provides the required notice.

    This shifts the calendar-management burden to the party that benefits from the auto-renewal.

    Strategy 6: Add Termination for Convenience During Renewal Terms

    If the contract includes auto-renewal, include a termination for convenience right during renewal terms with a reasonable notice period (30-60 days). This functions as a safety valve — even if the client misses the non-renewal window, they can still exit during the renewal term.

    Auto-Renewal Calendar Management

    Even the best-drafted auto-renewal clause is useless if nobody tracks the dates. Practical advice for lawyers managing client contracts:

    Create a centralized renewal calendar. Use your practice management system (Clio, MyCase, PracticePanther), a shared Google Calendar, or a dedicated contract management tool. The specific tool matters less than consistency. For a broader framework on building contract review into your workflow, see our guide on how to set up AI contract review.

    Set three reminders per contract:
    – 90 days before renewal: Initial alert — begin client discussion about whether to renew
    – 60 days before renewal: Decision deadline — get client’s written direction
    – 30 days before renewal: Final notice deadline — send non-renewal notice if client has decided to exit

    Designate a responsible person. For each contract, someone specific — not “the firm” or “the team” — owns the renewal date. If that person leaves, the responsibility must transfer explicitly.

    Annual contract audit. Once per year, review all active contracts with auto-renewal provisions. Verify that every renewal date is calendared and every contract is still meeting client needs. This audit prevents the “we forgot about that contract” scenario.

    Clause Labs’s Team tier ($299/month) includes obligation tracking with due date reminders and daily digests — purpose-built for tracking renewal dates, notice periods, and other critical contract deadlines across your entire portfolio.

    How Clause Labs Flags Auto-Renewal Traps

    Clause Labs’s AI identifies auto-renewal provisions automatically during contract review and flags:

    • Notice periods over 60 days
    • Renewal terms that match or exceed the initial term length
    • One-sided auto-renewal (only one party can prevent renewal)
    • Price escalation language without caps
    • Missing vendor notification obligations
    • Interaction between auto-renewal and termination provisions
    • Auto-renewal language buried in non-obvious sections

    The analysis highlights the exact notice deadline language so you know precisely what dates to calendar.

    Frequently Asked Questions

    Can I get out of an auto-renewed contract?

    It depends on the contract terms and applicable state law. If the auto-renewal provision complies with all applicable legal requirements and you missed the notice window, you’re generally bound for the renewal term — unless the contract includes a termination for convenience provision that allows mid-term exit, the auto-renewal clause violates a state auto-renewal statute (making it unenforceable), or you can show the clause was unconscionable. Prevention is far easier than cure: calendar the dates and don’t miss the window.

    Are auto-renewal clauses enforceable?

    Yes, in most jurisdictions, provided they comply with applicable disclosure and notice requirements. States with auto-renewal laws (California, New York, Virginia, and others) impose specific requirements — failure to comply can render the auto-renewal unenforceable. For B2B contracts in states without specific auto-renewal statutes, courts generally enforce auto-renewal provisions as written. The trend, however, is toward greater regulation, as noted in the National Law Review’s analysis of automatic renewal law updates.

    Do auto-renewal laws apply to B2B contracts?

    Increasingly, yes — though coverage varies significantly by state. New York’s auto-renewal statute (GBL § 5-903) explicitly covers B2B service contracts. Virginia extends consumer protections to small businesses. California’s CARL focuses on B2C but can apply to consumer-like business relationships. Illinois excludes B2B entirely. At the federal level, the FTC’s enforcement actions under ROSCA and the FTC Act can reach B2B practices. Check the specific laws in your jurisdiction.

    What’s a reasonable auto-renewal notice period?

    For most commercial contracts, 30 days is sufficient and reasonable. 60 days is common in enterprise agreements and generally acceptable. Anything over 90 days disproportionately benefits the party that drafted the clause. The notice period should reflect the realistic time needed to evaluate the relationship and make a renewal decision — not the drafter’s desire to create a narrow exit window.

    Should I always negotiate out auto-renewal?

    Not necessarily. Auto-renewal provides convenience for both parties in ongoing relationships — neither side has to remember to affirmatively extend. The goal isn’t to eliminate auto-renewal but to make it fair: reasonable notice periods, capped price increases, vendor notification obligations, and termination for convenience as a safety valve. A well-negotiated auto-renewal clause with a 30-day notice period and annual renewal terms is better than a fixed term with no renewal option at all.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    Auto-renewal traps are among the most common — and most preventable — contract risks. If you’re reviewing a SaaS agreement, vendor contract, or MSA with auto-renewal provisions, upload it to Clause Labs and check whether the renewal terms are fair before your client signs. Free tier: 3 reviews/month, no credit card required.

  • IP Assignment Clauses: The Clause That Destroys Startups (and How to Fix It)

    IP Assignment Clauses: The Clause That Destroys Startups (and How to Fix It)

    IP Assignment Clauses: The Clause That Destroys Startups (and How to Fix It)

    A Series A round dies in due diligence because the lead developer — a contractor who built the MVP — never assigned the IP to the company. The investor’s counsel finds no CIIA, no IP assignment clause, and no work-for-hire agreement. The contractor left eight months ago. Now the company’s entire codebase belongs, legally, to someone who isn’t answering emails.

    This isn’t a hypothetical. According to Jones Day’s analysis of IP due diligence in venture capital, IP chain-of-title failures are among the most common diligence killers in early-stage financings and acquisitions. With 85% of VC-backed exits occurring through acquisitions — where the buyer is literally paying for the company’s IP — a missing assignment clause doesn’t just create legal risk. It destroys the entire value proposition.

    If you advise startups, review contractor agreements, or handle employment contracts, this article covers the five IP assignment mistakes that cause the most damage and how to prevent each one. Upload any contractor or employment agreement to Clause Labs to check whether the IP provisions actually protect the company — the free analyzer flags missing assignments, overly broad language, and absent prior inventions carve-outs in under 60 seconds.

    What Is an IP Assignment Clause?

    An IP assignment clause is a contractual provision that transfers ownership of intellectual property from the creator (employee, contractor, co-founder) to the company. Without one, the default rules of copyright, patent, and trade secret law determine ownership — and those defaults rarely favor the company.

    The key concepts:

    Assignment vs. license. An assignment permanently transfers ownership. A license grants permission to use. If your client needs to own the IP outright — and in most startup contexts, they do — a license isn’t sufficient. Investors and acquirers want clean title, not a license that could be revoked.

    Types of IP covered. A comprehensive assignment covers patents, copyrights, trade secrets, trademarks, and moral rights (relevant in Canada and the EU). Narrow assignments that cover only “inventions” may miss copyrightable code, designs, or written content.

    Work-for-hire vs. assignment. Under 17 U.S.C. § 101, a “work made for hire” is automatically owned by the employer. But the work-for-hire doctrine only applies to employees creating within the scope of employment, or to independent contractors creating works in nine specific statutory categories (and only with a written agreement). For most contractor-created work, work-for-hire doesn’t apply — explicit assignment is required.

    The 5 IP Assignment Mistakes That Destroy Startups

    Mistake 1: No IP Assignment from Founders

    The problem. Two founders build a product for six months before incorporating. They form the company, issue stock, hire employees, and raise a seed round. But neither founder ever assigned the IP they created — the code, the algorithms, the designs, the brand — to the company.

    The legal reality. Each founder personally owns their contributions. The company has, at best, an implied license to use the IP. If one founder leaves, they can argue they own “their” portion of the codebase. If the company is acquired, the buyer’s counsel will find the gap in due diligence.

    The fix. Every founder should sign a Confidential Information and Inventions Assignment Agreement (CIIA) at incorporation — or ideally, before. According to Cooley GO’s guidance on CIIAs, the CIIA should assign all IP created for the company, include a prior inventions schedule, and cover confidentiality obligations. This is standard Silicon Valley practice, and investors expect it.

    What the CIIA should include:
    – Assignment of all IP created for or related to the company’s business
    – Prior inventions schedule (Exhibit A) listing excluded pre-existing IP
    – Confidentiality obligations
    – Cooperation clause (founder will sign additional documents to perfect assignment)
    – Representations that the founder has the right to assign

    Mistake 2: Contractor Work Without IP Assignment

    The problem. A startup hires a freelance developer to build the MVP. The contractor agreement covers scope, timeline, and payment — but says nothing about who owns the resulting code. Six months later, the contractor is building a competing product using “their” code.

    The legal reality. Unlike employees, contractors own what they create by default. As California’s IP assignment framework makes clear, the statutory protections for employee inventions (Labor Code § 2870) don’t extend to independent contractors. There’s no statutory scheme governing contractor IP assignment — the parties must contractually address it. If the contract doesn’t include an explicit IP assignment, the contractor retains all rights.

    Work-for-hire won’t save you in most cases. The work-for-hire doctrine under copyright law only applies to independent contractors in nine specific categories — including contributions to collective works, parts of a motion picture, and compilations — and requires a written agreement. Custom software development generally doesn’t qualify.

    The fix. Every contractor agreement must include:
    – Explicit IP assignment (not just work-for-hire language)
    – Scope definition covering all deliverables and work product
    – Representations that the contractor has the right to assign and hasn’t used third-party IP without authorization
    – A cooperation clause for perfecting the assignment

    Mistake 3: Overly Broad IP Assignment in Employment Agreements

    The problem. An employment agreement states: “Employee assigns to Company all inventions, works of authorship, and intellectual property created during the term of employment.” The employee’s weekend side project — an app completely unrelated to the employer’s business, built on the employee’s personal laptop, on personal time — is now arguably company property.

    The legal reality. Overbroad assignment clauses create three risks. First, they may be unenforceable in states with employee invention protection statutes. California Labor Code § 2870 explicitly protects inventions developed entirely on personal time without employer resources, provided they don’t relate to the employer’s business. Similar statutes exist in Delaware, Illinois, Minnesota, North Carolina, Washington, and several other states. Second, overbroad clauses generate employee resentment and departures. Third, if the employee’s personal invention has third-party encumbrances (like an open-source license), those encumbrances now affect the company.

    The fix. Narrow the assignment to IP that is:
    – Created within the scope of employment, OR
    – Created using company resources, equipment, or information, OR
    – Related to the company’s current or reasonably anticipated business

    And explicitly exclude inventions covered by California Labor Code § 2870 or equivalent state protections. The agreement should include a written notice that the assignment doesn’t cover protected personal inventions.

    Mistake 4: No Prior Inventions Carve-Out

    The problem. A senior engineer joins a startup and signs a standard IP assignment. The engineer previously developed a machine learning framework on their own time — a framework they now use (with modifications) in their work for the startup. Without a prior inventions carve-out, that framework is now arguably assigned to the company.

    The legal reality. If the prior invention has encumbrances — an existing license to another company, open-source obligations, or a co-ownership arrangement — those encumbrances now affect the startup’s IP portfolio. During due diligence, an acquirer or investor will discover the clouded title and demand indemnification, escrow, or a price reduction.

    The fix. Every CIIA and employment agreement should include a prior inventions schedule — typically Exhibit A — where the employee or contractor lists all pre-existing IP that is excluded from the assignment. The schedule should include:
    – Description of the prior invention
    – Date of creation
    – Any existing licenses or encumbrances
    – Whether the employee intends to use it in company work (and if so, a separate license grant)

    If the employee states “none” on the prior inventions schedule, that’s a representation the company can rely on.

    Mistake 5: Missing License-Back After Assignment

    The problem. A contractor builds a web application using their proprietary framework — a toolkit they’ve developed over years and use across multiple client projects. The contractor agreement assigns “all IP created in connection with the Services” to the client. Technically, the contractor can no longer use their own general-purpose framework for other clients.

    The legal reality. This creates an unenforceable or unworkable provision. No contractor will actually stop using their own development tools. But if the assignment is drafted broadly enough, the client could theoretically enforce it — or at minimum, it creates a dispute.

    The fix. Include a license-back clause:

    Contractor assigns to Company all right, title, and interest in Deliverable IP
(defined as intellectual property created specifically for Company under this
Agreement). Company grants Contractor a non-exclusive, perpetual, royalty-free
license to use, modify, and distribute Background IP (defined as pre-existing
tools, frameworks, and methodologies owned by Contractor prior to or independent
of this engagement) for any purpose, including for other clients.

    This structure gives the company clean ownership of the custom work while letting the contractor continue using their general-purpose tools.

    If you’re reviewing a contractor agreement right now, Clause Labs’s Solo tier ($49/month for 25 reviews) catches all five of these IP assignment mistakes — including missing prior inventions carve-outs and work-for-hire misapplication — across every contract type.

    IP Assignment Clause Elements: The Complete Checklist

    A well-drafted IP assignment clause should address each of these elements:

    Scope of assignment. What specific IP is being assigned? “All intellectual property” is common but can be overbroad. Better: “All Deliverable IP, including inventions, works of authorship, designs, and trade secrets created in the performance of Services under this Agreement.”

    Timing of assignment. When does the assignment occur? Options include upon creation (strongest for the company), upon delivery, or upon final payment. Assignment upon creation is the standard approach in CIIAs and employment agreements.

    Territory. Worldwide assignment is standard and necessary for companies operating across jurisdictions.

    Consideration. What does the assignor receive in exchange? For employees, it’s bundled into compensation. For contractors, it’s typically bundled into the service fees. Make sure the consideration is clearly stated — assignment without consideration can be challenged.

    Moral rights waiver. Required in jurisdictions that recognize moral rights (Canada, EU member states, Australia). In the US, moral rights are limited to visual art under VARA (17 U.S.C. § 106A), but international assignments should include a waiver.

    Cooperation clause. The assignor agrees to execute additional documents (patent applications, copyright registrations) necessary to perfect the assignment. Include a power of attorney as a backstop.

    Representations. The assignor represents they have the right to assign, haven’t previously assigned the IP to anyone else, and haven’t incorporated third-party IP without authorization.

    Work-for-hire declaration. For copyrightable works that fall within the nine statutory categories, include a work-for-hire declaration as a belt-and-suspenders approach alongside the assignment.

    IP Assignment by Agreement Type

    Founder and Co-Founder Agreements

    All IP created for the company is assigned to the company at incorporation via CIIA. Prior inventions are excluded. The critical question: what happens to assigned IP if a founder is terminated or leaves? Standard approach: the IP stays with the company (it was assigned, not licensed), but unvested stock is forfeited under the vesting schedule.

    Employment Agreements

    The assignment covers inventions and works created during employment, using company resources, or related to the company’s business. State-specific notice requirements apply — California, Delaware (19 Del. C. § 805), Illinois (765 ILCS 1060/2), Minnesota (Minn. Stat. § 181.78), and Washington (RCW 49.44.140) all have statutes protecting employee personal inventions. Failure to include the required statutory notice can render the assignment provision unenforceable. For guidance on related employment agreement issues, see our article on how to review contracts for red flags.

    Independent Contractor Agreements

    No automatic assignment exists. The contract must explicitly assign IP, and the work-for-hire doctrine has narrow applicability for contractors. Every contractor agreement should include: explicit IP assignment, background IP definition and license, prior inventions schedule, cooperation clause, and representations about third-party IP. For a more comprehensive review framework, see our guide on reviewing contractor agreements for misclassification risk.

    Development and Software Contracts

    The standard structure separates deliverable IP (assigned to the client) from background IP (retained by the developer with a license to the client to use it within the deliverables). Open-source components require special attention — if the developer incorporates GPL-licensed code, the copyleft obligations may affect the client’s proprietary code. The contract should require disclosure of all open-source components and compliance with their respective licenses. For SaaS-specific IP considerations, see our analysis of SaaS agreement review issues.

    IP Assignment Red Flags

    When reviewing any contract, flag these immediately:

    No IP assignment clause in a development contract. The contractor owns everything by default. This is the single most common and most expensive IP mistake in startup law.

    “All IP ever created” scope. Assignments covering all intellectual property created by the assignor, regardless of context, are overbroad and potentially unenforceable under state invention protection statutes.

    No prior inventions schedule. Without a carve-out, pre-existing IP with encumbrances gets swept into the assignment, creating title problems.

    Assignment conditioned on payment. If IP assignment only occurs upon full payment, a payment dispute means the company doesn’t own its own product. Standard practice: assignment upon creation, with a security interest or license-back if payment protection is needed.

    No cooperation clause. Without one, perfecting the assignment (filing patent applications, registering copyrights) requires the assignor’s voluntary cooperation — which may not be forthcoming after the relationship ends.

    Work-for-hire misapplication. Using work-for-hire language for contractor work that doesn’t fall within the nine statutory categories creates a false sense of security. The company thinks it owns the IP. It doesn’t. According to Farella Braun + Martel’s analysis, misapplying work-for-hire in contractor agreements can also create employment classification risk — if the IRS or a court determines that calling the relationship “work-for-hire” implies an employment relationship rather than an independent contractor arrangement.

    No representations about right to assign. The assignor may not own what they’re assigning — they may have used code from a previous employer, incorporated open-source components with restrictive licenses, or already assigned rights to someone else. IP assignment issues often interact with limitation of liability provisions — if the assignor breaches a representation about IP ownership, the damages may exceed the contract’s liability cap.

    How Clause Labs Handles IP Provisions

    Clause Labs’s AI identifies IP assignment, license, and work-for-hire clauses in any contract and flags:

    • Missing IP assignment in contractor and development agreements
    • Overly broad assignment language that may conflict with state employee invention protections
    • Absent prior inventions carve-outs
    • Missing cooperation clauses
    • Interactions between IP provisions and termination clauses (does IP assignment survive termination?)
    • Work-for-hire declarations used outside their statutory scope

    The analysis takes under 60 seconds and provides a clause-by-clause breakdown with risk ratings and suggested edits.

    Frequently Asked Questions

    Who owns IP by default without a contract?

    It depends on the relationship. For employees creating within the scope of employment, the employer generally owns copyrightable works under the work-for-hire doctrine, and many states have “hired to invent” doctrines for patentable inventions. For independent contractors, the contractor owns everything they create unless a written agreement says otherwise. For co-founders without a CIIA, each founder personally owns their individual contributions. These defaults are why written IP assignment agreements are essential.

    Can I assign IP verbally?

    Patent rights can potentially be assigned verbally, but copyright assignments require a written instrument signed by the owner under 17 U.S.C. § 204. In practice, verbal IP assignments are virtually impossible to enforce because they’re impossible to prove. Always get IP assignments in writing.

    What’s the difference between IP assignment and work-for-hire?

    Work-for-hire means the hiring party is considered the author and original owner — the creator never owned the IP at all. Assignment means the creator owned the IP and then transferred it to the company. The practical difference: assignment can theoretically be terminated by the author after 35 years under 17 U.S.C. § 203, while work-for-hire cannot. For most startup purposes, both achieve the same practical result — company ownership — but using both approaches (work-for-hire declaration plus assignment as backup) provides the strongest protection.

    Do I need IP assignment for every contractor?

    Yes, if the contractor is creating anything that could constitute intellectual property — code, designs, written content, inventions, processes. The only exception is contractors performing purely mechanical tasks with no creative component (e.g., data entry using your templates). When in doubt, include the clause. The cost of including an unnecessary IP assignment clause is zero. The cost of omitting a necessary one can be the entire company.

    What happens if an IP assignment is invalid?

    The company doesn’t own the IP it thought it owned. In an acquisition, this can reduce the purchase price, require escrow holdbacks, or kill the deal entirely. In a financing, investors may demand that IP issues be resolved before closing. In operations, the actual IP owner could demand license fees, seek an injunction, or start competing using “your” technology. The fix is to obtain a valid assignment as soon as possible — but if the assignor is uncooperative, it may require litigation.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    IP assignment issues are among the most expensive mistakes in startup law. If you’re reviewing a contractor agreement, employment contract, or founder CIIA, upload it to Clause Labs to check whether the IP provisions actually protect the company. The free tier includes 3 reviews per month — no credit card required.

  • Termination for Convenience Clauses: Why They’re Dangerous and How to Limit Them

    Termination for Convenience Clauses: Why They’re Dangerous and How to Limit Them

    Termination for Convenience Clauses: Why They’re Dangerous and How to Limit Them

    Your client just signed a three-year MSA worth $1.2 million. Six months in, the other side sends a one-paragraph notice: they’re terminating for convenience, effective in 30 days. No breach. No explanation. No compensation for the $180,000 your client invested in onboarding, infrastructure, and staffing up to deliver. The termination for convenience clause on page 19 gave them that right, and your client’s lawyer — who copy-pasted the termination section from the last deal — never flagged it.

    Termination for convenience is one of the most powerful rights in any commercial contract. According to World Commerce & Contracting, poor contract management erodes an average of 9% of annual revenue, and poorly drafted termination provisions are among the top five contract terms that drive disputes and value leakage. Yet most lawyers negotiate the indemnification clause for hours and barely glance at the termination section.

    This article covers the seven specific issues you should negotiate in every termination for convenience clause, the red flags that signal danger, and sample language you can adapt for your next deal. If you want a faster way to catch termination traps, upload any contract to Clause Labs’s free analyzer and get a clause-by-clause risk breakdown in under 60 seconds.

    Termination for Convenience vs. Termination for Cause

    Before negotiating either provision, you need to understand what each one does — because most contracts should include both, and the absence of one creates problems.

    Termination for convenience allows one or both parties to end the contract at any time, for any reason, without the other side committing a breach. The terminating party simply provides notice — typically 30 to 90 days — and the contract winds down. No cure period. No default required. Just a decision.

    Termination for cause requires a material breach. The non-breaching party provides notice identifying the breach, the breaching party gets a cure period (typically 15-30 days), and if the breach isn’t cured, the contract terminates. Cause-based termination often triggers different consequences than convenience-based termination — including indemnification obligations and potentially different payment terms.

    Element For Convenience For Cause
    Trigger Decision — no reason needed Material breach by the other party
    Cure period None Typically 15-30 days
    Notice Required (30-90 days typical) Required (specifying the breach)
    Compensation Varies — often negotiated May include damages for breach
    Fault No fault assigned Fault assigned to breaching party
    Frequency in disputes Common Very common

    The risk of having only termination for cause: if the relationship deteriorates but nobody technically breaches, neither party has a clean exit. The risk of having only termination for convenience: a party can exit a multi-year commitment at will, turning a long-term contract into something closer to a month-to-month arrangement.

    Most well-drafted commercial contracts include both provisions with different consequences for each.

    Where Termination for Convenience Comes From

    Termination for convenience originated in federal government contracting under the Federal Acquisition Regulation (FAR). Under FAR Part 12, the government can terminate virtually any contract for its convenience — a power rooted in the sovereign’s right to change policy priorities. Contractors receive equitable compensation for work performed, but they cannot sue for lost profits on unperformed work.

    This government-contract concept has migrated into commercial contracts over the past two decades, particularly in SaaS agreements, professional services MSAs, and vendor relationships. But there is a critical difference: in government contracting, a detailed regulatory framework governs compensation after termination. In commercial contracts, you get whatever the contract says — nothing more.

    That’s why the negotiation matters so much.

    The 7 Issues to Negotiate in Every Termination for Convenience Clause

    1. Who Has the Right?

    The most fundamental question: is termination for convenience mutual, or does only one party have it?

    Mutual termination means either side can walk away. This is the balanced approach and the starting position for most negotiations.

    One-sided termination gives only one party — usually the buyer, client, or larger company — the right to exit at will. The service provider or vendor is locked in for the full term while the other side can leave whenever it wants.

    If the other side insists on one-sided termination for convenience, the trade-off should be compensation. You’re essentially giving them an option, and options have value. A termination fee, extended notice period, or payment for work in progress offsets the asymmetry.

    2. Notice Period

    How much warning does the terminating party need to provide?

    Market ranges:
    – Short-term contracts (under 1 year): 30 days is standard
    – Multi-year contracts: 60-90 days is common
    – Large-scale engagements with significant staffing: 90-180 days is reasonable

    Red flag: Immediate termination with no notice period. This means the other side can pull the plug tomorrow with zero warning, leaving your client scrambling to reallocate resources and find replacement revenue.

    Negotiation principle: The notice period should reflect the time your client realistically needs to wind down operations, reassign staff, and find alternative arrangements. If your client needs to hire 10 people to perform this contract, a 30-day notice period is inadequate.

    3. Termination Fee (Kill Fee)

    A kill fee compensates the non-terminating party for the early exit. Common structures include:

    • Percentage of remaining value: 25-50% of fees that would have been payable for the remainder of the term
    • Fixed fee: A predetermined amount, often calculated as X months of fees
    • Declining fee: The kill fee decreases over time (e.g., 50% in year one, 25% in year two, 0% in year three)
    • Actual costs plus margin: Reimbursement for actual costs incurred plus a reasonable profit margin

    Kill fees are particularly important in contracts where the service provider makes upfront investments — hiring, equipment purchases, software licenses, facility buildouts — that can’t be easily recouped if the contract ends early.

    4. Wind-Down Obligations

    What happens between the termination notice and the effective date? A well-drafted clause addresses:

    • Transition assistance: Is the terminating party entitled to help migrating to a new provider? For how long? At what cost?
    • Data export and migration: Who handles data transfer? In what format? Within what timeframe?
    • Return of materials and IP: All confidential information, work product, and proprietary materials returned or destroyed
    • Final deliverables: Are partially completed deliverables owed? In what state of completion?

    Without wind-down obligations, the terminating party sends a notice and walks away. The other side is left holding half-finished work product, client data they can’t access, and no transition support.

    5. Payment for Work in Progress

    This is where the real money is. Upon termination for convenience:

    • Completed work: Payment for all work performed through the termination date should be non-negotiable
    • Non-cancellable commitments: If the service provider entered subcontracts or purchased materials that can’t be returned, who bears that cost?
    • Pre-paid amounts: If the terminating party pre-paid for a year of service and terminates at month six, is there a refund?
    • Expenses incurred: Reasonable expenses related to wind-down activities

    According to Holland & Knight’s analysis of convenience terminations, contractors in government and commercial contexts can typically recover direct costs for completed work, settlement expenses, and in some cases a proportional share of profit. But in commercial contracts, that recovery depends entirely on what the termination clause says.

    6. Survival Clauses

    Termination for convenience ends the contract, but it shouldn’t end everything. Certain provisions must survive:

    • Confidentiality: Survives (typically 2-5 years or indefinitely for trade secrets)
    • Indemnification: Survives for claims arising before termination
    • Limitation of liability: Survives — if it doesn’t, there’s no cap on post-termination claims
    • IP ownership: Survives — assignment of work product shouldn’t evaporate upon termination
    • Non-solicitation: Survives for stated period (if applicable)
    • Payment obligations: Survives for amounts accrued before termination

    The interaction between termination provisions and survival clauses is often overlooked. For a deeper analysis of how limitation of liability clauses interact with termination, see our clause-by-clause guide.

    7. Effective Date

    When does termination actually take effect?

    • Upon notice: The contract ends the moment notice is delivered. This is aggressive and leaves no transition time.
    • End of notice period: The contract continues through the notice period, then ends. This is the most common and most practical approach.
    • End of current billing cycle: Aligns termination with payment periods, reducing proration disputes.
    • Phased wind-down: Different obligations terminate at different times (e.g., new work stops immediately, but transition assistance continues for 60 days).

    Termination for Convenience by Contract Type

    SaaS Agreements

    SaaS termination provisions are among the most contentious. From the vendor’s perspective, annual or multi-year commitments fund product development and infrastructure — early termination undermines the business model. From the customer’s perspective, being locked into software that doesn’t work is unacceptable.

    Market standard: The customer can typically terminate at the end of any annual term with 30-60 days’ notice. Mid-term termination for convenience usually requires paying the remaining fees for the current term.

    Red flags to watch:
    – No termination right at all during the initial term (complete lock-in)
    – Only the vendor can terminate for convenience (one-sided)
    – Auto-renewal provisions paired with narrow cancellation windows that effectively eliminate exit rights

    MSAs and Professional Services

    Market standard: Mutual termination for convenience with 30-60 days’ notice. The critical issue is payment for work in progress, particularly for milestone-based engagements where work is partially completed.

    Key considerations:
    – Transition assistance is critical — the client needs to bring in a replacement provider without a gap
    – Work product delivered to date should be clearly owned by the client (check the IP assignment provisions)
    – Outstanding invoices remain payable regardless of termination

    Employment Agreements

    At-will employment is, in practical terms, termination for convenience by either party at any time. The key employment-specific issues are:

    • Severance triggers: Does termination without cause trigger severance payments?
    • Notice periods: Are there contractual notice requirements beyond at-will?
    • Garden leave: Does the employer pay the employee during a post-termination non-compete period?

    Government Contracts

    Government terminations for convenience operate under a different regulatory framework entirely. FAR 52.249-2 provides detailed procedures for settlement proposals, cost recovery, and dispute resolution. According to NCMA’s analysis, convenience terminations have surged in the current regulatory environment, making familiarity with FAR procedures essential for government contractors.

    Commercial contracts don’t have this safety net. Everything depends on what the parties negotiated.

    Termination for Convenience Red Flags

    When reviewing any contract, flag these issues immediately:

    One-sided termination rights. If only the other party can terminate for convenience, your client is locked in while the other side can leave at will. This is the most common and most dangerous red flag.

    No notice period. Immediate termination with zero warning leaves your client with no time to wind down operations, reassign staff, or find alternative arrangements.

    No payment for work in progress. If the clause is silent on payment for partially completed work, your client may have no contractual right to compensation for work already performed.

    No wind-down or transition period. Particularly dangerous in services contracts where the client depends on ongoing support.

    Forfeiture of pre-paid amounts. If the terminating party pre-paid for a year and terminates at month three, do they forfeit nine months of prepayment? Some contracts say yes.

    Disguised termination for cause. Watch for termination-for-convenience clauses that effectively avoid cure-period obligations. If the other side can terminate for convenience the day after a dispute arises, they’ll never need to provide a cure period.

    Auto-renewal paired with narrow termination window. A 30-day termination window on a three-year auto-renewing contract means your client has one chance per three years to exit. Miss it by a day, and they’re locked in again. Our guide on how to review contracts for red flags covers this pattern in detail.

    Sample Termination for Convenience Clauses

    Balanced Mutual Termination

    Either party may terminate this Agreement for convenience upon ninety (90) days'
prior written notice to the other party. Upon such termination, Client shall pay
Provider for (i) all Services performed through the effective date of termination,
(ii) all non-cancellable expenses reasonably incurred by Provider in connection
with the Services, and (iii) a wind-down fee equal to one month's average
monthly fees. Provider shall provide reasonable transition assistance for a
period of thirty (30) days following the effective date of termination.

    When to use: Standard MSAs and professional services agreements where both parties want flexibility. The wind-down fee compensates the provider for the early exit without creating an excessive penalty.

    One-Sided Termination with Kill Fee

    Client may terminate this Agreement for convenience upon sixty (60) days' prior
written notice. Upon such termination, Client shall pay Provider (i) all fees
for Services performed through the termination date, (ii) a termination fee
equal to 25% of the fees that would have been payable for the remainder of
the then-current term, and (iii) all non-cancellable third-party costs incurred
by Provider. Provider may not terminate this Agreement for convenience.

    When to use: When the client insists on one-sided termination rights and the provider needs protection for its upfront investment. The 25% kill fee is a reasonable middle ground.

    Termination with Transition Period

    Either party may terminate this Agreement for convenience upon thirty (30) days'
prior written notice. Following the effective date of termination, Provider shall
continue to perform transition assistance services for up to sixty (60) days at
Provider's then-current hourly rates. Provider shall cooperate in the orderly
transition of Services to Client or Client's designee, including data export in
commercially standard formats within ten (10) business days.

    When to use: Technology and managed services contracts where a smooth transition is critical to the client’s operations.

    How AI Assists with Termination Clause Review

    Clause Labs identifies all termination provisions in a contract — for cause, for convenience, auto-renewal, and expiration — and analyzes them in context. Specifically, the AI flags:

    • One-sided termination rights where only one party can exit
    • Missing notice periods or notice periods under 30 days
    • Absent wind-down and transition obligations
    • No payment provisions for work in progress upon termination
    • Interactions between termination and payment, survival, and limitation of liability clauses

    A manual review of termination provisions in a 30-page MSA typically takes 15-20 minutes to trace all the cross-references. Clause Labs’s analysis highlights every termination-related clause and its interactions in under 60 seconds, giving you a starting point for deeper review and negotiation.

    Frequently Asked Questions

    Can a party terminate for convenience without giving any reason?

    Yes — that’s the entire point of a termination-for-convenience clause. Unlike termination for cause, which requires a material breach, termination for convenience allows a party to exit for any reason or no reason at all. The only constraints are whatever notice period, payment obligations, and wind-down requirements the contract specifies. If the contract doesn’t specify any constraints, the terminating party can simply walk away.

    Is a termination fee enforceable?

    Generally yes, provided the fee is reasonable and not a penalty. Courts apply similar logic to termination fees as they do to liquidated damages provisions — the amount should reflect a reasonable estimate of the non-terminating party’s actual losses, not a punishment for exercising a contractual right. A termination fee equal to 25-50% of remaining contract value is typically defensible. A fee equal to 100% of remaining value looks more like a penalty and may face enforceability challenges.

    What’s the difference between termination for convenience and cancellation?

    In practice, the terms are often used interchangeably, but they can have different legal meanings depending on jurisdiction and context. Under the UCC (Uniform Commercial Code), “cancellation” occurs when one party terminates due to the other party’s breach, while “termination” can occur for any reason authorized by the contract. In non-UCC contracts, the distinction depends on how the agreement defines each term. Always check definitions.

    Can I negotiate termination for convenience out of a contract?

    You can try, and it sometimes works — particularly if you’re the party the other side wants locked in. If the counterparty depends on your long-term commitment (as a customer, partner, or key supplier), you have leverage to remove or significantly limit the termination-for-convenience right. The more common approach is to keep the right but add protections: longer notice periods, kill fees, payment for work in progress, and mandatory transition assistance.

    What happens to warranties after termination?

    It depends on whether the contract’s survival clause covers warranties. If warranties survive termination (they should), the warranting party remains liable for defects in work performed before termination. If the survival clause is silent on warranties, you may lose warranty protection on the moment the contract terminates. Always check — and negotiate — the survival clause alongside the termination provisions.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    Termination provisions are one of the most negotiated — and most overlooked — areas of commercial contracts. If you’re reviewing a contract right now and want to check whether the termination clause protects your client, upload it to Clause Labs for a free AI risk analysis. The free tier includes 3 contract reviews per month with no credit card required.

  • Liquidated Damages vs. Actual Damages: What Your Contract Should Say

    Liquidated Damages vs. Actual Damages: What Your Contract Should Say

    Liquidated Damages vs. Actual Damages: What Your Contract Should Say

    A construction contractor missed a project deadline by 47 days. The contract specified liquidated damages of $2,500 per day — a total exposure of $117,500. The contractor argued the clause was an unenforceable penalty. The court disagreed, finding the per diem rate was a reasonable estimate of the owner’s daily losses from delayed occupancy. The clause held.

    Now consider the opposite scenario. A software vendor’s agreement set liquidated damages at $500,000 for any breach of the SLA — regardless of severity. The customer experienced 45 minutes of downtime. The court struck the clause as a penalty: the amount bore no reasonable relationship to the actual harm.

    The difference between these outcomes comes down to a single legal concept: reasonableness at the time of contracting. Get it right, and liquidated damages give you certainty, efficiency, and enforceability. Get it wrong, and you have an unenforceable penalty clause that leaves your client with nothing.

    This guide covers when to use liquidated damages versus actual damages, how to draft them to survive judicial scrutiny, and the red flags that signal a clause is headed for trouble. If you have a contract on your desk with a damages provision you’re not sure about, upload it to Clause Labs free for an instant risk analysis.

    The Core Distinction: Liquidated vs. Actual Damages

    Actual damages (also called compensatory or general damages) are calculated after a breach occurs. The non-breaching party proves what they lost, and the court awards compensation. The burden of proof is on the claimant, the process is expensive, and the outcome is uncertain.

    Liquidated damages are calculated before breach occurs. The parties agree in advance on a fixed amount (or formula) that will be payable if a specific breach happens. No proof of actual loss is required at the time of claim — the agreed amount controls.

    When to use liquidated damages:
    – Actual damages would be difficult to calculate at the time of breach (lost reputation, operational disruption, opportunity cost)
    – The parties want certainty about exposure
    – Litigation over damages would be disproportionately expensive relative to the contract value
    – The breach type is time-sensitive and delay in calculating damages would compound the harm

    When actual damages are preferable:
    – Losses are easily quantifiable (unpaid invoices, cost of replacement goods)
    – The range of potential damages is too wide to predict at contracting
    – You want full recovery without an artificial cap
    – The contract involves unique or novel arrangements where no reasonable estimate is possible

    When Are Liquidated Damages Enforceable?

    The enforceability test is well-established and remarkably consistent across jurisdictions. Both the Restatement (Second) of Contracts § 356 and UCC § 2-718 apply a two-factor analysis:

    Factor 1: Difficulty of estimation. Actual damages must have been difficult to estimate at the time the contract was formed. If damages are easily calculable, courts ask why the parties needed a liquidated amount at all.

    Factor 2: Reasonableness of the amount. The liquidated amount must be a reasonable forecast of anticipated harm — not a punishment for breach. As the Restatement puts it: “A term fixing unreasonably large liquidated damages is unenforceable on grounds of public policy as a penalty.”

    Here’s what many lawyers miss: the parties’ characterization doesn’t matter. Writing “this amount constitutes liquidated damages and not a penalty” doesn’t make it so. Courts look past the label to the substance. As Restatement § 356 comment c states, “Neither the parties’ actual intention as to its validity nor their characterization of the term as one for liquidated damages or a penalty is significant in determining whether the term is valid.”

    The Penalty Problem

    If a liquidated damages provision is struck down as a penalty, the clause is void. In most jurisdictions, the non-breaching party is then left to prove actual damages from scratch — which may be difficult, time-consuming, and expensive. In some cases, striking the liquidated damages clause eliminates the only practical remedy.

    Courts consider several signals when evaluating penalty risk:

    • Disproportionate amount: Liquidated damages that grossly exceed any reasonable estimate of harm
    • Uniform amount regardless of breach severity: The same dollar amount for a minor delay and a complete failure to perform
    • Amount that grows over time without justification: Escalating damages with no connection to escalating harm
    • One party bears all the risk: Liquidated damages imposed only on one side for every possible breach

    Drafting Liquidated Damages Clauses That Hold Up

    Every enforceable liquidated damages clause includes these elements.

    1. State That Actual Damages Would Be Difficult to Calculate

    This isn’t mere recital — it’s the factual predicate for enforceability. Be specific:

    The parties acknowledge that [Vendor’s/Contractor’s] failure to [perform specific obligation] would cause damages to [Client] that are difficult to ascertain with certainty at the time of contracting, including but not limited to [lost revenue from delayed operations, reputational harm, disruption of dependent business activities].

    2. Set a Reasonable Amount with a Demonstrable Basis

    The amount should relate to an actual estimate of harm, not a round number pulled from thin air. Document the basis:

    • For construction delays: daily carrying costs (interest on construction loan, temporary facilities, lost rental income)
    • For SaaS downtime: hourly revenue impact, customer service costs, contractual penalties to the customer’s own clients
    • For late delivery: warehousing costs, idle labor, lost sales during stockout

    3. Specify the Triggering Event

    Define precisely what constitutes the breach that triggers liquidated damages. Ambiguity here leads to disputes:

    For each calendar day after the Completion Date that Contractor has not achieved Substantial Completion, Contractor shall pay Owner $[amount] as liquidated damages.

    4. Address Whether Liquidated Damages Are the Exclusive Remedy

    This is where many practitioners stumble. You must explicitly state whether liquidated damages are:

    • Exclusive remedy for the specific breach (most common and most enforceable)
    • In addition to other remedies for other types of breach
    • An alternative to actual damages (risky — some courts view this as evidence the parties didn’t intend a genuine pre-estimate)

    The liquidated damages set forth in this Section constitute Owner’s sole and exclusive remedy for Contractor’s delay in achieving Substantial Completion. Nothing in this Section limits Owner’s remedies for any other breach of this Agreement.

    5. Include a Mutual Acknowledgment

    Both parties acknowledge that the liquidated damages amount set forth herein represents a reasonable estimate of the anticipated harm from the specified breach, considering all circumstances known at the time of contracting. The parties agree that this amount is not a penalty.

    While labeling alone doesn’t guarantee enforceability, this acknowledgment reinforces the parties’ intent and may be persuasive to a reviewing court.

    6. Cap the Aggregate Exposure

    Uncapped liquidated damages can accumulate to unreasonable totals. A per-day delay penalty that runs for 18 months, for example, may produce an aggregate amount that far exceeds any reasonable damages. Standard practice: cap liquidated damages at a percentage of the contract value (typically 10-20% for construction, variable for other contract types).

    Liquidated Damages by Contract Type

    Construction Contracts

    Construction is where liquidated damages appear most frequently. Per diem delay damages are standard, and courts have extensive case law on enforceability.

    Typical structure: $X per calendar day of delay beyond the contractual completion date.

    Enforceable range: Courts have upheld daily rates ranging from hundreds to tens of thousands of dollars, depending on project value. Bradley’s analysis of liquidated damages provisions notes that the key is demonstrating a reasonable relationship between the daily rate and actual daily losses (construction loan interest, temporary facilities, lost rental income, management overhead).

    Red flag: Per diem rates that apply identically after substantial completion — when the owner can occupy and use the facility — may be struck as penalties because actual harm diminishes once the owner has beneficial use.

    Bonus/penalty structures: Some construction contracts pair early completion bonuses with delay liquidated damages. These are generally enforceable and demonstrate good-faith risk allocation.

    SaaS and Software Agreements

    SLA credits are, functionally, a form of liquidated damages for service failures.

    Typical structure: If uptime falls below X%, customer receives a credit of Y% of monthly fees.

    Why they’re almost always enforceable: SLA credits are typically modest in amount relative to the contract value, they’re tied to specific, measurable performance failures, and the difficulty of calculating actual business harm from software downtime is well-recognized.

    The exclusive remedy trap: Most SaaS agreements designate SLA credits as the customer’s “sole and exclusive remedy” for downtime. This is favorable for the vendor — it prevents the customer from claiming actual damages (which could far exceed the credit amount). If you represent the customer, negotiate carve-outs from the exclusive remedy designation for extended outages, data loss, or security breaches. For more on SaaS agreement traps, see our guide on reviewing SaaS agreements with AI.

    Employment Agreements

    Liquidated damages in employment contexts most commonly appear in:

    • Training cost reimbursement: Employer pays for employee’s professional development; if the employee leaves within X years, they repay a prorated portion. Courts generally uphold these when the repayment schedule is reasonable and declining.
    • Non-compete breach: Some agreements specify a fixed amount payable if the employee breaches a non-compete. Enforceability is jurisdiction-dependent and highly variable — non-compete law itself varies dramatically by state.

    California note: California’s general prohibition on non-competes under Bus. & Prof. Code § 16600 means liquidated damages for non-compete violations are functionally unenforceable there, regardless of how well the damages clause is drafted. For more on the interaction between damages and indemnification provisions, see our guide on indemnification clauses.

    Real Estate Transactions

    Earnest money deposits are the most common form of liquidated damages in real estate.

    How it works: Buyer deposits 1-3% of purchase price. If buyer defaults, seller retains the deposit as liquidated damages.

    California’s specific framework: California Civil Code §§ 1675-1678 provides detailed rules. If the deposit is 3% or less of the purchase price for residential property, the provision is presumed valid. Above 3%, the seller bears the burden of proving reasonableness. The clause must be separately signed or initialed by both parties.

    Other states don’t have California’s statutory specificity, but the general principle holds: earnest money deposits that represent a reasonable percentage of the purchase price (typically 1-3%) are enforceable as liquidated damages. For a comprehensive look at real estate contract risks beyond damages provisions, see our real estate contract review checklist.

    Supply and Procurement Contracts

    Late delivery damages: $X per day or per week of late delivery. Enforceable when tied to buyer’s documented costs of delay (idle manufacturing line, expedited shipping for alternative supply, lost sales).

    Quality defect damages: A fixed amount per defective unit. Enforceable when the cost of inspecting and replacing defective goods is difficult to calculate per unit at the time of contracting.

    Volume shortfall damages: When a supplier commits to deliver a minimum quantity and falls short, liquidated damages may cover the buyer’s cost of sourcing from alternative (usually more expensive) suppliers.

    Common Liquidated Damages Mistakes

    Mistake 1: Setting the amount too high. The most frequent drafting error. A liquidated damages amount that materially exceeds any reasonable estimate of harm will be struck as a penalty, leaving the non-breaching party with no liquidated recovery and the burden of proving actual damages.

    Mistake 2: Setting the amount too low. Less common but equally problematic. A token amount ($100 per day of delay on a $5 million project) may signal that the parties didn’t take the provision seriously — and courts may decline to enforce it.

    Mistake 3: Failing to state that actual damages are difficult to calculate. This recital is more than boilerplate. Without it, courts may infer that the parties had no reason to use liquidated damages, which undercuts the enforceability rationale.

    Mistake 4: Making liquidated damages cumulative with full actual damages. This is a trap. If a party can recover both liquidated damages AND actual damages for the same breach, the aggregate recovery may be deemed punitive. Specify whether liquidated damages are exclusive or address the interaction explicitly.

    Mistake 5: Not specifying whether LDs are an exclusive remedy. Ambiguity about available remedies invites litigation. State it clearly.

    Mistake 6: Using liquidated damages when actual damages are easily calculable. If the most likely form of harm is an unpaid invoice or a quantifiable cost, liquidated damages may be unnecessary and harder to enforce.

    Mistake 7: Forgetting to update amounts when contract value changes. A liquidated damages amount set for a $100,000 contract may be unreasonable for the same contract after a $2 million change order.

    Liquidated Damages Red Flags in Contract Review

    When reviewing a contract, flag these provisions for closer analysis:

    • LD amount exceeding 15-20% of total contract value — potential penalty
    • No statement about difficulty of calculating actual damages — missing enforceability predicate
    • LDs cumulative with unlimited actual damages — double recovery risk
    • One-sided LDs — only one party faces liquidated damages exposure
    • No cap on aggregate LDs — per-day penalties can accumulate to unreasonable totals
    • LDs triggered by events outside the paying party’s control — unfair risk allocation
    • Escalating LDs without justification — penalty indicator if harm doesn’t escalate similarly
    • Identical LD amount for breaches of wildly different severity — suggests penalty, not pre-estimate

    For a broader framework on identifying contract risks, see our complete contract red flags checklist. And for the related issue of how liability caps interact with liquidated damages, see our guide on limitation of liability clauses.

    How AI Handles Damages Provision Review

    AI contract review tools can flag liquidated damages provisions quickly, but the enforceability analysis requires human judgment. Here’s what AI does well and where you need to step in.

    AI handles effectively:
    – Identifying liquidated damages clauses (even when buried in remedy or general terms sections)
    – Flagging whether the clause specifies exclusive or cumulative remedies
    – Detecting interaction between liquidated damages and limitation of liability provisions
    – Identifying missing elements (no difficulty-of-calculation statement, no cap)

    Requires attorney judgment:
    – Whether the dollar amount is reasonable relative to anticipated harm
    – Jurisdiction-specific enforceability standards
    – Whether the contract’s commercial context supports liquidated versus actual damages
    – Strategic advice on whether to accept, negotiate, or reject the provision

    Clause Labs flags damages provisions, identifies missing elements, and highlights potential enforceability concerns — giving you the starting point so you can focus your time on the judgment-intensive analysis. Try it free with 3 reviews per month.

    Solo lawyers handling 20+ contracts per month use Clause Labs to flag liquidated damages issues, liability cap mismatches, and missing remedy provisions automatically — starting at $0 for 3 reviews. Try Clause Labs free.

    Frequently Asked Questions

    What’s the difference between liquidated damages and a penalty?

    A liquidated damages clause represents a genuine pre-estimate of anticipated harm from a specific breach. A penalty is a punishment for breach — an amount meant to coerce performance rather than compensate for loss. Courts look at whether the amount is reasonable relative to anticipated or actual harm. If it’s grossly disproportionate, it’s a penalty and unenforceable regardless of what the contract calls it.

    Can liquidated damages be challenged in court?

    Yes. The party paying liquidated damages can argue the clause is an unenforceable penalty. The party receiving liquidated damages may argue the clause is valid. The court applies the two-factor test: difficulty of estimation and reasonableness of amount. Burden of proof varies by jurisdiction.

    Should liquidated damages be the exclusive remedy?

    For the specific breach they cover, usually yes. Making liquidated damages the exclusive remedy for a defined breach type (delay, downtime, defects) makes the provision more enforceable because it demonstrates the parties’ intent to pre-allocate risk. But preserve separate remedies for other breach types — liquidated damages for delay shouldn’t preclude claims for, say, defective work or IP infringement.

    How do I calculate a reasonable liquidated damages amount?

    Start with the actual costs that the non-breaching party would likely incur. For construction delay: daily loan interest + temporary facilities + lost rental income + management overhead. For SaaS downtime: hourly revenue impact + customer service costs. Document the calculation and keep it in the contract file — this contemporaneous evidence of reasonableness is your best defense if the clause is challenged.

    Can both parties have liquidated damages obligations?

    Yes, and mutual liquidated damages provisions are often viewed more favorably by courts because they suggest balanced risk allocation rather than one-sided punishment. Example: contractor pays delay LDs; owner pays acceleration costs if owner causes delay.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation. Liquidated damages enforceability varies significantly by jurisdiction — verify your state’s standards before relying on this guide.

  • Non-Compete Clauses After the FTC Ban: What’s Enforceable in 2026?

    Non-Compete Clauses After the FTC Ban: What’s Enforceable in 2026?

    Non-Compete Clauses After the FTC Ban: What’s Enforceable in 2026?

    The FTC’s federal non-compete ban is dead. On September 8, 2025, the Fifth Circuit officially dismissed the FTC’s appeal in Ryan, LLC v. FTC, ending the agency’s attempt to ban non-competes nationwide. The rule that was supposed to void an estimated 30 million non-compete agreements never took effect.

    But here’s what matters for the lawyers reviewing employment agreements right now: the legal landscape shifted anyway. At least four states now ban non-competes almost entirely. More than a dozen impose income thresholds, notice requirements, or durational limits. And the FTC is still enforcing against individual non-competes it considers unfair under Section 5 of the FTC Act — case by case, employer by employer.

    If you review employment agreements for clients, the question is no longer “Are non-competes legal?” It’s “Is this specific non-compete enforceable in this specific state for this specific employee?” This guide maps the current landscape. Try Clause Labs Free to upload any employment agreement and get instant AI analysis of restrictive covenant issues — including jurisdiction-specific flags.

    What Happened: The FTC Rule Timeline

    Understanding the current state requires knowing how we got here.

    April 23, 2024: The FTC issues a final rule banning most non-compete agreements nationwide, with an effective date of September 4, 2024. The rule would have voided existing non-competes for most workers and banned new ones entirely, with narrow exceptions for senior executives and business sale agreements.

    August 20, 2024: A federal judge in the Northern District of Texas (in Ryan, LLC v. FTC) sets aside the rule nationwide, holding that the FTC exceeded its statutory authority. A separate challenge in Florida reached a similar conclusion.

    September 2025: The FTC officially abandons its appeal, accepting the vacatur. The federal non-compete ban is permanently dead.

    January 2026: At a public workshop, the FTC clarifies it will no longer pursue a categorical ban. Instead, it retains authority to challenge specific non-competes under Section 5 of the FTC Act as unfair methods of competition — particularly those targeting low-wage workers or imposing exceptionally broad restrictions.

    The net result: no federal ban, but a significantly heightened enforcement environment and a patchwork of state laws that have only gotten more aggressive.

    State-by-State Non-Compete Enforceability in 2026

    The real action on non-competes is now entirely at the state level. Here’s where things stand, organized by restriction level.

    States That Ban Non-Competes Almost Entirely

    Four states effectively prohibit non-compete agreements for employees in most or all circumstances:

    California: The broadest ban. Cal. Bus. & Prof. Code § 16600 voids non-competes in nearly all employment contexts. Recent legislation strengthened the ban by voiding out-of-state non-competes applied to California workers — meaning a Texas employer can’t enforce a Texas non-compete against an employee who moves to California.

    Minnesota: Non-competes entered into on or after July 1, 2023, are prohibited for most workers, with limited statutory exceptions for the sale of a business or dissolution of a partnership.

    Oklahoma: Title 15, § 219A generally voids non-competes, with exceptions for the sale of a business and dissolution of a partnership. This has been Oklahoma law for decades.

    North Dakota: N.D. Cent. Code § 9-08-06 broadly prohibits non-competes, with the standard business-sale exception.

    States with Income Threshold Restrictions

    A growing number of states tie non-compete enforceability to how much the employee earns. As of 2026, these income thresholds have been updated:

    State 2026 Non-Compete Income Threshold Notes
    Colorado $127,091 (non-compete); $76,254 (non-solicit) Must protect trade secrets; penalties for violations
    Illinois $75,000 (non-compete); $45,000 (non-solicit) Next increase in 2027
    Washington $126,858.83 (employee); $317,147.09 (contractor) Annual adjustment
    Oregon $119,541 Bureau of Labor and Industries sets annual threshold
    Massachusetts Hourly employees exempt; requires garden leave or other consideration 12-month maximum duration
    Virginia Low-wage employees exempt (below median state wage) Applies to non-competes and non-solicits
    Rhode Island Non-exempt employees exempt Additional protections for healthcare workers

    What this means in practice: If your client employs a marketing coordinator in Illinois earning $60,000/year, a non-compete in that employee’s agreement is void as a matter of law. The same non-compete for an Illinois sales director earning $200,000/year could be enforceable — subject to reasonableness analysis on scope, duration, and geography.

    States with Significant Restrictions (But Not Bans)

    These states allow non-competes but impose meaningful procedural or substantive limits:

    Massachusetts: Non-competes cannot exceed 12 months. The employer must provide either garden leave pay (at least 50% of base salary during the restricted period) or “other mutually-agreed upon consideration.” Non-competes are void for hourly workers, undergraduate or graduate students in internships, and employees terminated without cause.

    Washington: Beyond the income threshold, non-competes exceeding 18 months are presumptively unreasonable. Employers must disclose the non-compete terms before or at the time of the job offer.

    Oregon: Non-competes are limited to 12 months, must be provided in writing at the time of offer or at least two weeks before the start date, and the employee must be given a copy. Only employees earning above the threshold are covered.

    Colorado: Beyond the income threshold, employers must notify the worker of the non-compete “in a separate document” and provide the non-compete terms before the worker accepts the offer. Colorado imposes civil penalties of up to $5,000 per violation for non-competes that don’t meet statutory requirements.

    States Where Non-Competes Are Generally Enforceable

    In the remaining states, non-competes are permitted subject to the traditional common-law reasonableness test:

    • Duration: Typically 6-24 months. Courts in most states view anything over 2 years skeptically.
    • Geographic scope: Must be tied to the employer’s actual business footprint or the employee’s territory.
    • Activity scope: Must be limited to competitive activities, not all employment.
    • Consideration: Most states require independent consideration for existing employees (not just continued employment).

    States like Texas, Florida, Georgia, and Ohio generally enforce non-competes that meet these reasonableness standards — but even in these states, judicial scrutiny has increased since the FTC attempted its ban.

    Non-Compete Red Flags to Catch in Every Employment Agreement

    For lawyers reviewing contracts containing non-competes — whether reviewing for red flags generally or specifically evaluating restrictive covenants — here are the issues that should trigger immediate attention.

    Duration Exceeding State Limits

    The most common enforceable window is 12-24 months. A 36-month non-compete is likely unreasonable in most jurisdictions. Several states cap duration by statute: Massachusetts and Oregon at 12 months, Washington at 18 months (presumptive). Any non-compete exceeding these limits is either unenforceable or requires modification.

    Employee Below Income Thresholds

    If the employee earns less than the applicable state threshold — $75,000 in Illinois, roughly $127,000 in Colorado, roughly $127,000 in Washington — the non-compete is likely void. Review compensation details before analyzing the non-compete itself.

    Independent Contractor Non-Competes

    Non-competes for independent contractors are almost always unenforceable, and in states like Washington (which sets the contractor threshold at $317,147), they face even more scrutiny. Worker misclassification compounds the risk: if the “contractor” is actually an employee under state law, the non-compete analysis shifts entirely.

    Geographic Scope That’s Unreasonable

    A nationwide non-compete for a regional sales representative is overbroad. A non-compete covering “anywhere the Company does business” when the company operates globally is almost certainly unenforceable. Geographic scope must be proportional to the employee’s actual territory or the employer’s legitimate business interests.

    Non-Solicitation Disguised as Non-Compete

    Some agreements label a broad restriction as a “non-solicitation” to avoid non-compete scrutiny. If the non-solicitation effectively prevents the employee from working in their field — for example, by prohibiting contact with any current or former customer of a company with thousands of customers — it may be analyzed as a de facto non-compete.

    Forum Selection Clauses That Forum-Shop

    Watch for a Texas employer with California employees specifying Texas law and Texas courts for non-compete enforcement. California’s strong anti-non-compete policy may override the contractual choice of law — but the employee would have to litigate that point, which costs money and time. Flag these provisions for your clients.

    Missing Adequate Consideration

    In many states, an existing employee signing a non-compete must receive independent consideration beyond continued employment. A new bonus, promotion, stock grant, or similar benefit may be required. If the non-compete is presented to an existing employee with no new consideration, it may be unenforceable.

    Alternatives to Non-Competes That Actually Work

    When advising clients who want to protect business interests without the enforceability risk of non-competes, recommend these alternatives:

    Non-Solicitation Agreements

    Prohibit the departing employee from soliciting specific clients, customers, or employees — but don’t prohibit competitive employment itself. Non-solicitation agreements are enforceable in almost every state, including California (with limitations), because they restrict a specific behavior rather than the ability to earn a living.

    Drafting tip: Limit the non-solicitation to clients the employee actually worked with during their last 12-24 months of employment. Restrictions covering all company clients (including those the employee never met) face enforceability challenges.

    Enhanced Confidentiality / NDA Agreements

    Protect specific trade secrets and confidential information through robust NDAs with clearly defined protected information. When properly drafted, confidentiality agreements protect the employer’s most valuable information without restricting where the employee can work.

    Drafting tip: Define confidential information by category (customer lists, pricing algorithms, strategic plans) rather than using catch-all language. Include the five standard exclusions and a reasonable duration.

    Garden Leave Provisions

    Require the employer to continue paying the employee during the restricted period in exchange for the employee not competing. This approach is gaining traction because it provides the employee with compensation during the restriction, making enforceability far more likely.

    Drafting tip: Specify the percentage of base salary (50-100%) and duration. Massachusetts now requires garden leave as consideration for non-competes. Other states are likely to follow.

    Intellectual Property Assignment Agreements

    Protect company innovations by requiring employees to assign work-related IP to the employer. This doesn’t restrict where the employee works — it restricts what they can take with them.

    Drafting tip: Always include a “prior inventions” schedule so employees can document pre-existing IP. Many states — including California, Delaware, Illinois, Minnesota, and Washington — have statutes limiting the scope of invention assignment agreements.

    Clawback Provisions

    Tie bonuses, equity vesting, or deferred compensation to post-employment behavior. If the employee competes within a specified period, they forfeit unvested benefits. This is functionally a non-compete with an economic enforcement mechanism rather than a judicial one.

    Drafting tip: Ensure the clawback amount is proportional and doesn’t function as a penalty. Courts may analyze excessive clawbacks as non-competes subject to the same enforceability standards.

    How to Review Non-Compete Clauses: A 7-Step Process

    When a client brings you an employment agreement with a non-compete, follow this process:

    1. Identify the governing jurisdiction: Which state’s law controls? Is the employee located in a different state from the employer? If the employee works remotely from a ban state (California, Minnesota), the non-compete may be void regardless of the contract’s choice-of-law clause.

    2. Check if non-competes are enforceable in that state: Refer to the state-by-state table above. If the state bans non-competes, your analysis is short.

    3. Verify income threshold compliance: For states with income thresholds (Colorado, Illinois, Washington, Oregon, Virginia), confirm the employee’s compensation exceeds the applicable limit.

    4. Evaluate reasonableness: Duration (over 24 months is suspect), geographic scope (proportional to the employee’s actual territory), and activity scope (limited to competitive activities, not all employment).

    5. Check for adequate consideration: Is the non-compete part of the initial offer, or is it being imposed on an existing employee? If the latter, what independent consideration is being provided?

    6. Assess blue-pencil doctrine applicability: Some states allow courts to modify overbroad non-competes to make them reasonable. Others strike them entirely. Knowing whether your jurisdiction blue-pencils affects your risk assessment and negotiation strategy.

    7. Review interaction with other restrictive covenants: Does the agreement also contain a non-solicitation, NDA, and IP assignment? Together, these may provide the protection the employer needs without the non-compete — which gives you a negotiation argument for striking it.

    For a broader framework on reviewing employment agreements for all types of red flags, see our complete contract red flags checklist.

    Industry-Specific Non-Compete Considerations

    Technology and Software

    Tech non-competes face the most scrutiny. California’s ban covers virtually all tech workers. In other states, courts are increasingly skeptical of non-competes for software engineers because the “confidential information” at issue (coding skills, general technology knowledge) is difficult to distinguish from general professional competence.

    Healthcare

    Multiple states — including Colorado (effective August 2025) — have specific statutes restricting non-competes for physicians, nurses, dentists, and other healthcare providers. The concern is patient access: if a physician can’t practice within 50 miles, patients lose access to care.

    Financial Services

    FINRA-registered representatives are subject to industry-specific rules. FINRA’s Protocol for Broker Recruiting allows registered representatives to take certain client information (names, addresses, phone numbers, account types) when switching firms, which limits the practical impact of non-compete and non-solicitation restrictions.

    Sales and Business Development

    Sales non-competes are among the most commonly litigated. Courts evaluate them based on whether the salesperson had access to proprietary customer relationships, pricing data, or strategic plans — not just whether they “know the clients.” Account-based restrictions (prohibiting solicitation of specific named accounts) are more enforceable than blanket geographic restrictions.

    Executive Employment

    Senior executives face the fewest protections. Most state restrictions and income thresholds don’t apply at executive compensation levels. Courts generally enforce reasonable non-competes for C-suite executives and senior officers, particularly when tied to significant equity or severance packages. That said, even executive non-competes must meet reasonableness standards on duration, scope, and geography.

    How Clause Labs Handles Non-Compete Analysis

    When you upload an employment agreement to Clause Labs, the AI automatically identifies non-compete, non-solicitation, non-disclosure, and other restrictive covenant provisions. It flags potential enforceability issues based on the governing jurisdiction, highlights overbroad restrictions on duration, geography, and activity scope, and identifies missing consideration provisions.

    For employment agreements specifically, Clause Labs’s employment playbook covers all the restrictive covenant issues discussed in this article — plus compensation risks, termination traps, and IP assignment problems. See our free employment agreement review tool for details on the full employment agreement analysis.

    Frequently Asked Questions

    Are non-competes banned federally in 2026?

    No. The FTC’s federal non-compete rule was struck down by a Texas federal court in August 2024 and the FTC abandoned its appeal in September 2025. There is no federal ban. However, the FTC retains authority to challenge specific non-competes it considers unfair under Section 5 of the FTC Act — particularly those targeting low-wage workers or imposing unreasonably broad restrictions.

    Can my client’s employer enforce their non-compete in 2026?

    It depends entirely on the state, the employee’s compensation level, and the specific terms of the non-compete. In California, Minnesota, Oklahoma, and North Dakota, the answer is almost certainly no. In states with income thresholds (Colorado, Illinois, Washington, Oregon), it depends on whether the employee earns above the threshold. In all other states, enforceability requires meeting the traditional reasonableness test on duration, geography, and scope.

    What happens to non-competes already signed before state bans took effect?

    This varies by state. Minnesota’s ban applies only to non-competes entered into on or after July 1, 2023 — pre-existing non-competes remain enforceable if otherwise valid. California’s strengthened law, by contrast, voids non-competes regardless of when they were signed because the state’s underlying prohibition (§ 16600) has existed for over a century.

    Can a non-compete be enforced across state lines?

    Potentially, but it’s complicated. If a Texas employer sues in Texas to enforce a non-compete against a former employee now working in California, the California court may refuse to enforce it under California’s strong public policy against non-competes. Choice-of-law provisions in the agreement don’t always control — courts apply their own state’s public policy when it conflicts with the contractual choice of law.

    How do I advise a client who already signed a non-compete?

    First, determine the governing jurisdiction and whether the non-compete is enforceable under current law. Second, assess whether the specific terms (duration, geography, scope) are reasonable. Third, evaluate the practical enforcement risk — would the employer actually sue, given the cost and uncertainty? Many non-competes that are technically enforceable are never enforced because the economics don’t justify litigation. Finally, if the client wants to compete, consider whether negotiating a release or buyout is possible.

    Have an employment agreement with a non-compete you need to evaluate? Upload it to Clause Labs free — the AI flags jurisdiction-specific enforceability issues, overbroad restrictions, and missing consideration provisions in under 60 seconds. Start with 3 free reviews per month, no credit card required.

    This article is for informational purposes only and does not constitute legal advice. Non-compete enforceability varies significantly by jurisdiction, and state laws change frequently. The information in this article reflects the legal landscape as of February 2026. Consult a qualified attorney in the relevant jurisdiction for advice specific to your situation.

  • Force Majeure Clauses in 2026: What Changed After COVID and What to Include

    Force Majeure Clauses in 2026: What Changed After COVID and What to Include

    Force Majeure Clauses in 2026: What Changed After COVID and What to Include

    Before March 2020, force majeure was the clause nobody read. Buried between the notice provision and the severability section, it was copy-pasted from one contract template to the next without a second thought. Then a pandemic shut down the global economy, and lawyers discovered that a clause written to handle hypothetical earthquakes was useless against a real public health emergency.

    The litigation wave that followed was massive. Courts across the country addressed force majeure claims in hundreds of COVID-related cases, and the results were overwhelmingly consistent: force majeure clauses are interpreted narrowly. If “pandemic” was not listed as a triggering event, many courts held that COVID did not qualify. If the clause required performance to be “prevented” rather than merely “hindered,” many businesses that could still perform — just at greater cost or difficulty — had no defense.

    Six years later, force majeure drafting has permanently changed. If your force majeure clause still looks like it did in 2019, it is dangerously outdated. This guide covers what changed, the 10 elements every modern clause needs, and sample language you can adapt for your next agreement. Upload any contract to Clause Labs to check whether your force majeure clause meets 2026 standards — free, no signup required.

    What Force Majeure Actually Means

    Force majeure — French for “superior force” — is a contractual provision that excuses one or both parties from performing their obligations when extraordinary events occur beyond their control. It is not a common law doctrine in the United States. It is purely contractual: if the contract does not include a force majeure clause, neither party can invoke it.

    This is a critical distinction that many business clients misunderstand. Unlike frustration of purpose (which exists at common law in most U.S. jurisdictions) or impracticability under the Uniform Commercial Code Section 2-615, force majeure only exists if the parties put it in the contract. No clause, no defense.

    Three additional principles shape how courts analyze force majeure:

    Force majeure clauses are interpreted narrowly. The Fifth Circuit confirmed this in Mieco LLC v. Pioneer Natural Resources USA Inc. (2024), holding that force majeure provisions in a natural gas contract required strict adherence to the contract’s specific language. Courts do not stretch force majeure clauses to cover events the parties did not explicitly contemplate.

    The invoking party bears the burden of proof. You must prove that the event qualifies under the clause, that it actually caused your inability to perform, and that you took reasonable steps to mitigate the impact.

    General catch-all language is unreliable. “Events beyond the reasonable control of the affected party” sounds broad, but courts frequently limit catch-all provisions to events similar in kind to those specifically listed (the ejusdem generis rule). If your clause lists “fire, flood, earthquake” and a pandemic hits, the catch-all may not save you.

    What COVID Changed

    Before COVID: Standard Force Majeure (Pre-2020)

    The typical force majeure clause before 2020 looked like this:

    “Neither party shall be liable for any failure or delay in performing its obligations under this Agreement to the extent that such failure or delay results from acts of God, war, terrorism, earthquake, flood, fire, or other natural disasters.”

    Short. Generic. Copy-pasted. No one negotiated it because no one expected to invoke it. The clause was in the contract because the template included it, not because anyone thought about what it needed to cover.

    After COVID: Modern Force Majeure (2026)

    The pandemic exposed every weakness in the old model:

    • No mention of pandemics, epidemics, or public health emergencies. When COVID hit, parties argued whether a virus qualified as an “act of God.” Courts split — some said yes, many said no.
    • No mention of government orders. Lockdowns, travel bans, and business closure orders were not natural disasters. Were they “other events beyond reasonable control”? Courts disagreed.
    • No provision for supply chain disruption. Manufacturers who could not source raw materials had contracts that only excused performance for events at the production facility itself.
    • No notice or mitigation requirements. Parties invoked force majeure months after the event began, with no documentation of what they did to minimize the impact.
    • No termination trigger. Contracts suspended performance indefinitely, creating “zombie contracts” — not active, not terminated, just in limbo.

    Modern force majeure clauses, drafted in light of six years of post-COVID litigation, address every one of these gaps. As WilmerHale’s analysis recommended early in the pandemic, parties should draft force majeure provisions with both specificity and flexibility.

    The 10 Elements of a Well-Drafted Force Majeure Clause (2026)

    1. Triggering Events List — Specific Plus Catch-All

    List specific events for certainty. Add a catch-all for flexibility. Include both.

    Specific events that should be listed explicitly:
    – Pandemic, epidemic, public health emergency
    – Government orders, regulations, sanctions, embargoes
    – War, armed conflict, terrorism, civil unrest
    – Natural disasters (earthquake, flood, hurricane, wildfire, volcanic eruption)
    – Cyberattack, data breach, critical infrastructure failure
    – Supply chain disruption
    – Labor shortages, strikes, lockouts
    – Utility failures (power, telecommunications, internet)

    Catch-all language:

    “…or any other event beyond the reasonable control of the affected party that could not have been reasonably foreseen at the time of entering into this Agreement.”

    The “reasonably foreseen” qualifier matters. Post-COVID, parties cannot claim a pandemic is unforeseeable — it already happened. The catch-all needs to account for genuinely novel events while excluding known risks that should be managed through other contractual provisions.

    Notably, cyberattacks are increasingly included in modern force majeure clauses. The Jaguar Land Rover cyberattack in 2025, estimated at GBP 1.9 billion in losses affecting 5,000+ businesses, demonstrated that cyber events can disrupt supply chains as severely as natural disasters.

    2. Causation Standard — “Prevented” vs. “Hindered” vs. “Delayed”

    The single most litigated word in force majeure clauses is the verb connecting the event to the performance failure.

    Standard Threshold Effect on Invoking Party
    Prevented Performance must be impossible Highest burden — most difficult to invoke
    Hindered Performance is significantly more difficult or burdensome Middle ground — reasonable balance
    Delayed Any delay in performance qualifies Lowest burden — easiest to invoke

    Recommendation: Use “prevented, hindered, or delayed” to give the invoking party reasonable protection. If you represent the party more likely to receive performance (the buyer, the client), you may prefer “prevented” alone — it limits the other side’s ability to invoke force majeure for mere inconvenience.

    The UK Supreme Court’s decision in RTI Ltd v. MUR Shipping BV (2024) clarified that the causation standard interacts with mitigation obligations: the force majeure event, not the party’s own failure to act, must be the cause of the performance failure.

    3. Notice Requirements

    A force majeure clause without notice requirements is a clause that invites abuse. Specify:

    • Timeline: Written notice within 5-10 business days of the force majeure event commencing.
    • Content: Nature of the event, expected duration, specific obligations affected, steps being taken to mitigate.
    • Ongoing updates: Regular updates (every 14-30 days) during the force majeure period.
    • Consequences of failure: Late or missing notice waives the right to invoke force majeure.

    Sample notice provision:

    “The affected party shall provide written notice to the other party within seven (7) business days of becoming aware of the force majeure event, describing the nature of the event, the obligations affected, the expected duration, and the mitigation measures being undertaken. The affected party shall provide updated notices at least every fourteen (14) days during the continuation of the force majeure event.”

    4. Mitigation Obligation

    Force majeure does not mean “stop working and wait.” The affected party must take reasonable steps to minimize the impact.

    What “reasonable mitigation” includes:
    – Sourcing alternative suppliers or materials
    – Reassigning personnel
    – Adjusting timelines or sequencing
    – Using alternative methods of performance
    – Communicating proactively with affected stakeholders

    What it does not include:
    – Spending disproportionate sums to overcome the event
    – Accepting non-contractual performance from the other party (as clarified in RTI Ltd v. MUR Shipping)
    – Taking on unreasonable business risk to maintain performance

    Key principle: Failure to mitigate can void the force majeure defense entirely. Document every mitigation step.

    5. Duration and Termination Right

    How long can a force majeure event suspend performance before one or both parties can terminate? Without a defined trigger, you get zombie contracts — suspended indefinitely, impossible to plan around.

    Standard approach: Either party may terminate if the force majeure event continues for 60-90 consecutive days, or 120 cumulative days in any 12-month period.

    What to specify:
    – Termination notice period (typically 30 days after the duration threshold)
    – Whether termination is automatic or requires written notice
    – Treatment of prepaid fees, deposits, and work-in-progress upon termination
    – Wind-down obligations after termination

    6. Allocation of Risk During Force Majeure

    What happens to money during the suspension period?

    • Payment obligations: Are they suspended too, or does payment continue? In most commercial agreements, payment obligations should be suspended proportionally when performance is excused.
    • Partial performance: If the affected party can perform partially, what are the obligations on both sides?
    • Mitigation costs: Who pays for the costs of mitigation measures?

    This is the most commonly overlooked element in force majeure clauses, and it generates the most post-event disputes.

    7. Exclusions

    Force majeure is not a general excuse clause. Define what it does not cover:

    • Economic hardship. A contract becoming more expensive to perform is not force majeure. Prices fluctuate. Markets shift. That is business risk, not force majeure.
    • Currency fluctuation. Exchange rate changes are foreseeable market risks.
    • Market downturns. A recession, declining demand, or competitive pressure is not a force majeure event.
    • Known risks at time of contracting. If a risk was known or foreseeable when the contract was signed, it should not qualify. Post-2020, pandemics are arguably foreseeable — courts are already weighing this question in international trade disputes.
    • Self-inflicted events. Events caused by the invoking party’s own acts or omissions do not qualify.

    8. Insurance Interaction

    Force majeure clauses and business interruption insurance overlap — but not perfectly. Address:

    • Whether the invoking party must first look to insurance before invoking force majeure
    • Whether insured losses reduce the force majeure protection
    • Requirements for maintaining specific insurance coverage for force majeure-type risks
    • Cooperation obligations for insurance claims processing

    9. Dispute Resolution for Force Majeure Claims

    Force majeure disputes are time-sensitive. By the time a standard arbitration resolves, the underlying event may be over. Consider:

    • Expedited resolution: 30-day expert determination for disputes about whether an event qualifies
    • Interim measures: Provisional orders maintaining the status quo during the dispute
    • Default rule: If no resolution within the expedited period, the force majeure claim stands pending full determination

    10. Post-Force Majeure Obligations

    What happens when the force majeure event ends?

    • Resumption timeline: How quickly must the affected party resume performance?
    • Catch-up provisions: Is the affected party entitled to additional time to make up for the suspension period?
    • Changed circumstances: If the economic or operational landscape has permanently changed, is there a mechanism for renegotiating terms?
    • Documentation: Final report on the force majeure event, its impact, and the mitigation measures taken

    Force Majeure by Contract Type

    Force majeure operates differently depending on the contract type. Here are the unique considerations for each.

    Supply and Procurement Contracts. The most heavily negotiated force majeure clauses post-COVID. Key issues: raw material shortages, logistics disruptions, alternative sourcing obligations. Baker McKenzie’s analysis of 2025 tariff-related supply chain uncertainty confirms that force majeure provisions in supply contracts now routinely include trade sanctions, export controls, and tariff escalation as triggering events.

    SaaS Agreements. Tension between uptime SLAs and force majeure. A force majeure clause that excuses downtime effectively guts the SLA. Best practice: force majeure excuses the SLA credits, but prolonged outages (beyond 72 hours) trigger termination rights regardless. See our SaaS agreement review guide for the full clause-by-clause analysis.

    Commercial Leases. Rent abatement during force majeure events is now standard in many markets. Key negotiation points: whether rent is suspended or deferred (deferred means you still owe it later), and whether tenant improvements and build-out timelines are extended.

    Employment Agreements. Force majeure in employment contracts addresses remote work mandates, furloughs, layoff triggers, and return-to-office requirements. Post-COVID, many employers include pandemic-specific provisions authorizing temporary remote work without modifying the employment relationship. For a broader look at what to watch for when reviewing these agreements, see our limitation of liability clause guide, which covers risk allocation provisions that interact directly with force majeure.

    Construction Contracts. Delay clauses and force majeure clauses often exist separately. Force majeure typically extends the timeline and excuses liquidated damages, but does not entitle the contractor to additional compensation for idle resources.

    Force Majeure Red Flags

    When reviewing any contract, flag these issues. For a broader framework on spotting contract problems, see our red flags checklist.

    No force majeure clause at all. This leaves both parties relying on common law doctrines (impossibility, impracticability, frustration of purpose) that are narrower and less predictable than a well-drafted contractual provision. Our free AI contract review tool flags missing force majeure provisions as part of its standard analysis.

    Pre-2020 boilerplate. If the clause lists “acts of God, war, terrorism, earthquake, flood, fire” and nothing else, it was written before anyone thought about pandemics, cyberattacks, or government-ordered shutdowns. It needs an update.

    One-sided force majeure. Only one party can invoke the clause. Unless there is a specific commercial reason for this asymmetry (rare), force majeure should be mutual.

    No mitigation requirement. Without a mitigation obligation, the invoking party can stop performing entirely and wait indefinitely. This creates moral hazard.

    No duration limit or termination trigger. Zombie contract risk. If force majeure can suspend performance forever, neither party can plan, and the non-affected party is trapped.

    “Including but not limited to” without specific events. Sounds broad but may be too vague to enforce. Courts want specificity. A list of specific events followed by a catch-all is more reliable than a catch-all alone.

    Force majeure that excuses payment obligations. Unusual and risky for the non-affected party. In most commercial contexts, force majeure should excuse delivery or service obligations, not payment for work already performed.

    Sample Modern Force Majeure Clause

    Here is a balanced, post-2026 force majeure clause with annotations explaining each provision:

    Force Majeure. Neither party shall be liable for any failure or delay in performing its obligations under this Agreement (other than payment obligations for services already rendered) to the extent that such failure or delay is caused by a Force Majeure Event, provided that the affected party: (a) provides written notice to the other party within seven (7) business days of becoming aware of the Force Majeure Event; (b) uses commercially reasonable efforts to mitigate the impact of the Force Majeure Event; and (c) resumes performance promptly upon cessation of the Force Majeure Event.

    “Force Majeure Event” means any event beyond the reasonable control of the affected party, including but not limited to: pandemic, epidemic, or public health emergency; government order, regulation, sanction, or embargo; war, armed conflict, terrorism, or civil unrest; natural disaster; cyberattack or critical infrastructure failure; supply chain disruption; labor shortage, strike, or lockout; or utility failure — provided that such event was not reasonably foreseeable at the time this Agreement was executed and is not caused by the affected party’s acts or omissions.

    If a Force Majeure Event continues for more than sixty (60) consecutive days or ninety (90) cumulative days in any twelve-month period, either party may terminate this Agreement upon thirty (30) days’ written notice, without liability for such termination.

    This clause covers the essential elements: specific triggering events with a catch-all, a foreseeability qualifier, a payment carve-out, notice requirements, mitigation obligations, resumption obligation, and a termination trigger. Adapt it to the specific deal dynamics, contract type, and risk allocation appropriate for your transaction. Need to check whether an existing contract’s force majeure clause meets this standard? Upload it to Clause Labs for a free AI analysis — the tool compares your clause against each of the 10 elements above in under 60 seconds.

    How AI Handles Force Majeure Review

    AI contract review tools are particularly effective at force majeure analysis because the task is largely structural: is the clause present, what events does it cover, does it include required sub-provisions?

    Clause Labs specifically:
    – Detects presence or absence of a force majeure clause
    – Flags outdated (pre-2020) language that lacks modern triggers
    – Identifies missing elements (no notice requirement, no mitigation obligation, no termination trigger)
    – Checks for one-sided provisions
    – Compares the clause against current market standards

    Upload your contract free to check whether your force majeure clause meets current standards. The analysis takes under 60 seconds — 3 free reviews per month, no credit card required.

    Frequently Asked Questions

    Does every contract need a force majeure clause?

    Not necessarily, but most commercial contracts benefit from one. Without a force majeure clause, parties must rely on common law doctrines — impossibility, impracticability, or frustration of purpose — which have higher thresholds and less predictable outcomes. If the contract involves ongoing performance obligations (service delivery, supply, construction), a force majeure clause is strongly recommended. For simple, one-time transactions, the risk may not justify the drafting investment.

    Can I invoke force majeure for economic hardship?

    Almost certainly not. Courts consistently hold that increased costs, financial difficulties, or market downturns do not constitute force majeure events. Norton Rose Fulbright’s analysis of COVID-related force majeure claims confirms that economic hardship, standing alone, is insufficient. A contract becoming unprofitable is not the same as performance becoming impossible.

    What if my contract was signed before COVID — is the force majeure clause still valid?

    Yes, the clause is valid as written. The question is whether it covers the event you want to invoke. A pre-COVID clause that lists “pandemic” is fine. A pre-COVID clause that lists only “acts of God, war, earthquake” may not cover a pandemic — depending on the jurisdiction and whether the court applies ejusdem generis to the catch-all language. The clause does not need to be re-drafted to be valid, but it may need to be re-drafted to be useful.

    Does force majeure apply to payment obligations?

    In most contracts, no. The standard position is that force majeure excuses performance obligations (delivery, service, construction) but not payment for work already performed. A force majeure clause that excuses payment obligations is unusual and should be a red flag in any contract review. Some contracts split the difference: payment obligations may be deferred (not forgiven) during the force majeure period.

    Can both parties invoke force majeure simultaneously?

    Yes, and it happens more often than you might expect. A supply chain disruption can simultaneously prevent the manufacturer from delivering and the buyer from receiving goods. When both parties invoke force majeure, the clause should specify how mutual invocation is handled — typically, both parties are relieved of their respective obligations for the duration of the event, with either party able to terminate if the event exceeds the duration threshold.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

  • Limitation of Liability Clauses: How to Spot, Negotiate, and Draft Them Right

    Limitation of Liability Clauses: How to Spot, Negotiate, and Draft Them Right

    Limitation of Liability Clauses: How to Spot, Negotiate, and Draft Them Right

    A single missing carve-out in a limitation of liability clause cost one SaaS vendor’s customer $4.2 million in unrecoverable losses after a data breach. The vendor’s liability was capped at 12 months of fees — $36,000 total. The customer had no recourse for the remaining $4.16 million because their lawyer never negotiated a data breach carve-out.

    Limitation of liability (LOL) clauses determine the maximum financial exposure under any contract. According to World Commerce & Contracting, poor contracting practices erode an average of 9% of annual revenue, with losses exceeding 15% in complex industries. LOL clauses sit at the center of that loss — they define what you can and cannot recover when things go wrong.

    This guide breaks down every component of limitation of liability clauses, covers negotiation strategy by contract type, and gives you sample language you can use in your next deal. If you need an AI second opinion on the LOL clause in a contract sitting on your desk right now, try Clause Labs free — it flags cap mismatches, missing carve-outs, and one-sided exclusions in under 60 seconds.

    What Is a Limitation of Liability Clause?

    A limitation of liability clause is a contractual provision that caps how much one party can recover from the other for breach. Without one, liability exposure is theoretically unlimited — governed only by whatever a court might award.

    Every LOL clause has two distinct components:

    The liability cap sets a maximum dollar amount on direct damages. Common formulations include a fixed dollar figure (“liability shall not exceed $500,000”), a formula tied to fees (“liability shall not exceed fees paid in the prior 12 months”), or a per-incident or aggregate structure.

    The consequential damages exclusion disclaims liability for indirect losses — lost profits, lost revenue, lost data, business interruption. This component often matters more than the cap itself because consequential damages frequently dwarf direct damages in commercial disputes.

    If you review only one clause in any contract, make it this one.

    The Two Types of Liability Limitations

    Liability Caps on Direct Damages

    Liability caps come in several structures, each with different risk profiles:

    Cap Structure Example Language Risk Level for Claimant
    Fixed dollar amount “shall not exceed $500,000” Moderate — predictable but may be too low
    Formula-based (12-month fees) “shall not exceed fees paid in prior 12 months” Variable — depends on deal size
    Formula-based (total fees) “shall not exceed total fees paid under this agreement” Better — scales with relationship
    Per-incident “shall not exceed $100,000 per claim” High — limits recovery on each event separately
    Aggregate “shall not exceed $500,000 in the aggregate” Highest — total pool depletes across all claims
    Annual reset “shall not exceed $200,000 per contract year” Moderate — replenishes annually

    What’s market-standard depends entirely on contract type and deal size. For SaaS agreements, 12 months of fees paid is the most common cap structure. For professional services, the cap typically ranges from 1x to 3x fees paid under the applicable statement of work.

    The critical question for your client: does the cap reflect the actual exposure if the other party breaches? A $36,000 cap on a contract governing data for 100,000 customers is a red flag no matter how “market-standard” the formula looks.

    Consequential Damages Exclusions

    Consequential damages — also called indirect, special, or incidental damages — include lost profits, lost revenue, lost data, lost business opportunities, and business interruption. The distinction between direct and consequential damages is notoriously unclear, and courts across jurisdictions define the boundary differently.

    Most commercial contracts exclude consequential damages mutually, meaning neither party can recover indirect losses from the other. This creates a predictable risk allocation — but it also means that if a vendor’s software failure destroys your client’s revenue for a quarter, your client may only recover the subscription fees, not the lost revenue.

    When to accept a mutual consequential damages exclusion:
    – Both parties face roughly equal risk of indirect losses
    – The direct damages cap is adequate to cover realistic exposure
    – Specific high-risk scenarios (data breach, IP infringement) are carved out

    When to push back:
    – The exclusion is one-sided (only protects the other party)
    – There are no carve-outs for high-impact scenarios
    – Your client’s primary risk is exactly the type of loss being excluded

    The 8 LOL Negotiation Points That Actually Matter

    1. Cap Amount

    The most obvious negotiation point, but lawyers often accept “12 months of fees” without analyzing whether it’s adequate.

    What to evaluate: Compare the cap to realistic damages scenarios. If your client’s potential loss from a breach is $2 million and the cap is $50,000, the clause is functionally an exculpation — the breaching party faces no meaningful financial consequence.

    What to push for: Higher multiples for higher-risk contracts. For a SaaS agreement governing critical business data, push for 24 months of fees or a fixed dollar minimum (e.g., “the greater of $500,000 or 12 months of fees”).

    2. Cap Structure — Per-Incident vs. Aggregate

    An aggregate cap depletes over time. If your client suffers three separate breaches and the aggregate cap is $500,000, the third claim may find the cap already exhausted.

    What to push for: Per-incident caps for ongoing relationships, or aggregate caps that reset annually.

    3. Carve-Outs and Super-Caps

    Carve-outs exclude certain obligations from the general liability cap. Standard carve-outs include:

    • IP indemnification — almost always carved out
    • Confidentiality breach — increasingly carved out, especially post-GDPR/CCPA
    • Data breach — the most heavily negotiated carve-out in 2026
    • Willful misconduct and fraud — typically carved out by law regardless
    • Indemnification obligations — often subject to a separate, higher “super-cap”

    A super-cap sets a higher ceiling for carved-out obligations. For example: general liability capped at 12 months of fees, but IP indemnification and data breach obligations capped at 24 months of fees.

    4. Mutual vs. One-Sided

    A one-sided LOL clause only protects one party. If you’re reviewing a vendor contract where the vendor’s liability is capped but your client’s isn’t, that’s a fundamental imbalance.

    What to push for: Mutuality. If the vendor insists on capping their liability at 12 months of fees, your client’s liability should be capped at the same amount. The rare exception: if one party’s risk profile is genuinely asymmetric (e.g., a data processor handling millions of records for a small fee).

    5. Consequential Damages Scope

    The words matter here. “Indirect, special, incidental, and consequential damages” is broader than just “consequential damages.” Some clauses add “lost profits” to the exclusion explicitly — which may otherwise be classified as direct damages in certain jurisdictions.

    What to watch for: Broad exclusions that use “including but not limited to” followed by a list that captures both indirect AND potentially direct damages.

    6. Gross Negligence and Willful Misconduct

    Should the cap apply when a party’s breach results from gross negligence or willful misconduct? In most jurisdictions, courts will refuse to enforce caps that protect a party from its own intentional wrongdoing. But contractual language matters.

    What to push for: Explicit carve-out: “The limitations in this Section shall not apply to damages arising from a party’s gross negligence, willful misconduct, or fraud.”

    7. Data Breach Liability

    Data breach carve-outs are the single most negotiated LOL term in 2026. With all 50 states now requiring breach notification and notification timelines ranging from 30 to 60 days, the costs of a data breach extend far beyond the contract value.

    What to push for: At minimum, a super-cap for data breach obligations (2x-3x the general cap). Ideally, full carve-out from the cap for breaches involving personal data.

    8. Insurance Alignment

    The LOL cap should align with the insurance requirements in the contract. If you require the other party to carry $5 million in professional liability insurance, but their liability is capped at $50,000, the insurance requirement is meaningless — they’ll never face a claim that exceeds the cap.

    What to push for: LOL cap at or near the required insurance limits, or at minimum, ensure insurance covers the carved-out obligations.

    Limitation of Liability Red Flags

    These provisions should trigger immediate review and pushback:

    • No LOL clause at all — unlimited exposure for both parties
    • Trivially small cap relative to potential damages (e.g., $10,000 cap on a contract governing $5 million in services)
    • One-sided cap that only protects the vendor
    • Broad consequential damages exclusion with no carve-outs — especially for data breach, IP infringement, and confidentiality
    • Cap includes indemnification — this may effectively nullify the indemnification clause entirely
    • Aggregate cap that doesn’t reset in multi-year contracts — the cap depletes over time
    • Exclusion of “all indirect damages” — broader than just consequential
    • No carve-out for willful misconduct or fraud — may be unenforceable anyway, but the ambiguity creates litigation risk

    For a structured approach to catching these red flags across entire contracts, see our contract red flags checklist.

    LOL by Contract Type: What’s Market-Standard

    SaaS and Software Agreements

    • Standard cap: 12 months of fees paid
    • Standard carve-outs: IP indemnification, confidentiality breach, data breach
    • Consequential damages: Typically excluded mutually
    • Negotiation range: 12-24 months of fees; push for super-cap on data breach at 2x-3x
    • Key issue: Ensure data breach liability isn’t swallowed by a general cap that’s pegged to a relatively small subscription fee

    For a deeper analysis of SaaS-specific risks, see our guide on how to review SaaS agreements.

    Professional Services and MSAs

    • Standard cap: Total fees paid under the SOW, or 12 months of fees
    • Standard carve-outs: Gross negligence, willful misconduct, IP infringement
    • Negotiation range: 1x-3x fees for the general cap
    • Key issue: Order of precedence — does the MSA cap apply to individual SOWs, or is there one aggregate cap across all SOWs?

    Employment Agreements

    • LOL is uncommon in employment agreements
    • When it appears, it’s typically in arbitration provisions limiting remedies
    • Key issue: Statutory rights (FLSA, Title VII, state wage laws) cannot be contractually limited. Courts will void LOL provisions that attempt to cap statutory damages.
    • Reference: ABA Model Rules require attorneys to ensure clients understand what rights they’re waiving

    Vendor and Supplier Agreements

    • Standard cap: Purchase price or 12 months of purchases
    • Product liability: Usually carved out (and often non-waivable under UCC Section 2-719 for personal injury in consumer goods)
    • Warranty claims: May be subject to a separate cap
    • Negotiation range: 1x-2x annual purchase volume

    Commercial Leases

    • Landlord liability limitations are common and often aggressively one-sided
    • Tenant liability limitations are rare — landlords resist them
    • Property damage and personal injury: Usually carved out
    • Key issue: “Exculpatory clauses” in leases face heightened scrutiny in residential contexts and may be void by statute in some jurisdictions

    How LOL Interacts With Other Clauses

    Limitation of liability doesn’t exist in isolation. It interacts with — and sometimes contradicts — other risk allocation provisions.

    LOL + Indemnification: The most contentious interaction. If indemnification obligations fall within the general liability cap, a party’s entire indemnification protection may be worth less than the legal fees to enforce it. Push for indemnification to sit outside or above the general cap. For a complete analysis, read our indemnification clause guide.

    LOL + Insurance: The cap should align with insurance requirements. If the contract requires $2 million in errors and omissions coverage but caps liability at $50,000, there’s a fundamental mismatch.

    LOL + Warranties: If warranty breach counts against the general cap, a significant warranty claim could exhaust the cap and leave nothing for other claims.

    LOL + Data breach provisions: Is data breach inside or outside the cap? Given that 29% of most AI-related breaches stem from third-party SaaS platforms, this question has real financial consequences.

    Best practice: Review the LOL clause, indemnification, insurance requirements, and warranties together as a system. A change to one affects all the others.

    Sample LOL Clause Language

    Standard Mutual Limitation (SaaS)

    LIMITATION OF LIABILITY. EXCEPT FOR (I) OBLIGATIONS UNDER SECTION [INDEMNIFICATION],
(II) BREACH OF SECTION [CONFIDENTIALITY], (III) LIABILITY ARISING FROM A PARTY'S
GROSS NEGLIGENCE OR WILLFUL MISCONDUCT, OR (IV) A PARTY'S DATA BREACH OBLIGATIONS
UNDER SECTION [DATA PROTECTION]:

(A) IN NO EVENT SHALL EITHER PARTY BE LIABLE TO THE OTHER FOR ANY INDIRECT, SPECIAL,
INCIDENTAL, OR CONSEQUENTIAL DAMAGES, REGARDLESS OF THE FORM OF ACTION; AND

(B) EACH PARTY'S TOTAL AGGREGATE LIABILITY UNDER THIS AGREEMENT SHALL NOT EXCEED
THE AMOUNTS PAID OR PAYABLE BY CUSTOMER IN THE TWELVE (12) MONTHS IMMEDIATELY
PRECEDING THE EVENT GIVING RISE TO THE CLAIM.

FOR CLAIMS ARISING UNDER SECTIONS [INDEMNIFICATION], [CONFIDENTIALITY], OR
[DATA PROTECTION], EACH PARTY'S LIABILITY SHALL NOT EXCEED TWO TIMES (2X) THE
AMOUNTS PAID OR PAYABLE IN THE TWELVE (12) MONTHS PRECEDING THE CLAIM.

    This structure provides a general cap with a super-cap for high-risk obligations. Note that it’s mutual, includes standard carve-outs, and creates a two-tier system.

    Aggressive Vendor-Favorable Limitation (For Comparison)

    IN NO EVENT SHALL VENDOR BE LIABLE FOR ANY INDIRECT, SPECIAL, INCIDENTAL,
CONSEQUENTIAL, PUNITIVE, OR EXEMPLARY DAMAGES, INCLUDING LOST PROFITS, LOST DATA,
OR BUSINESS INTERRUPTION. VENDOR'S TOTAL LIABILITY SHALL NOT EXCEED THE FEES PAID
BY CUSTOMER IN THE THREE (3) MONTHS PRECEDING THE CLAIM.

    Red flags: one-sided (only limits vendor liability), extremely low cap (3 months vs. 12), no carve-outs, includes “lost data” in the exclusion (dangerous for data-dependent services).

    When LOL Clauses Are Unenforceable

    Courts may refuse to enforce LOL clauses in several circumstances:

    • Unconscionability — gross disparity in bargaining power combined with unreasonably harsh terms. Under UCC Section 2-719(2), limitation of consequential damages for personal injury from consumer goods is prima facie unconscionable.
    • Personal injury or death — most jurisdictions prohibit contractual limitations on bodily injury liability
    • Fraud or intentional misconduct — a party generally cannot limit liability for its own fraud
    • Violation of statutory rights — employment discrimination damages, consumer protection claims, and other statutory remedies typically cannot be capped by contract
    • Failure of essential purpose — when a limited remedy “fails of its essential purpose” under UCC 2-719(2), the broader limitation may fall with it. As the Masuda Funai analysis explains, courts split on whether a failed exclusive remedy also voids the consequential damages exclusion.

    Jurisdiction note: Enforceability standards vary significantly by state. Negotiate as if the clause will be enforced — but understand that a court in your jurisdiction might reach a different conclusion. As the Lexology analysis of U.S. contractual liability limitations notes, commercial contracts between sophisticated parties face a lower unconscionability hurdle than consumer agreements.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    How AI Analyzes Limitation of Liability

    AI contract review tools can evaluate LOL clauses against market standards in seconds. When you upload a contract to Clause Labs, the AI identifies the cap amount, cap structure, carve-outs, consequential damages exclusions, and checks for interactions with indemnification and insurance requirements. It flags missing LOL clauses, one-sided limitations, caps that appear too low for the deal size, and missing carve-outs for data breach and IP obligations.

    The free tier includes 3 contract reviews per month — enough to test the analysis on a real contract before committing to the Solo plan at $49/month for 25 reviews.

    Frequently Asked Questions

    What’s a reasonable limitation of liability cap?

    It depends on the contract type and deal size. For SaaS agreements, 12 months of fees paid is standard, with 24 months becoming more common for enterprise deals. For professional services, 1x-3x total fees is typical. The key question: does the cap bear a reasonable relationship to the potential damages if the other party breaches?

    Should limitation of liability be mutual?

    Generally, yes. If one party’s liability is capped, the other party’s should be too. A one-sided cap suggests the drafting party is trying to shift disproportionate risk. The exception: genuinely asymmetric risk profiles, such as when a data processor handles massive volumes of personal data for a small processing fee.

    What’s the difference between limitation of liability and indemnification?

    Limitation of liability caps total exposure. Indemnification creates an obligation to compensate for specific losses (typically third-party claims). They interact in critical ways — read our indemnification clause guide for the full analysis. The most important question is whether indemnification obligations fall inside or outside the liability cap.

    Should data breach be carved out of the liability cap?

    In 2026, the answer is almost always yes. Data breach costs regularly exceed contract values by orders of magnitude. At minimum, negotiate a super-cap (2x-3x the general cap) for data breach obligations. For contracts involving significant personal data, push for a full carve-out.

    Can you have no limitation of liability?

    Yes — if neither party includes an LOL clause, liability is unlimited (subject to whatever a court would award). This is more common in simple agreements, but it’s risky for both parties. Even if you’re the party with more bargaining power, unlimited liability creates unpredictable exposure.

    What happens if the cap is reached — can I still make claims?

    Once the cap is exhausted, the capped party has no further financial exposure for claims subject to the cap. However, obligations carved out from the cap (IP indemnification, data breach, willful misconduct) remain enforceable up to their respective limits. This is why carve-outs and annual cap resets matter.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

  • Indemnification Clauses Explained: What Every Lawyer Should Negotiate

    Indemnification Clauses Explained: What Every Lawyer Should Negotiate

    Indemnification Clauses Explained: What Every Lawyer Should Negotiate

    Indemnification disputes are among the most litigated contract provisions in commercial practice. According to World Commerce & Contracting, poor contract management costs organizations an average of 9.2% of revenue, and indemnification is consistently one of the top three clauses driving post-signature disputes. When a Jones Day analysis of indemnity disputes notes that “even with detailed indemnification clauses, disputes can still get messy because of fights about whether a claim is even the subject of indemnification,” the message is clear: getting this clause right during negotiation is worth orders of magnitude more than litigating it later.

    This guide breaks down indemnification clauses into their component parts, identifies the seven negotiation points that actually matter, flags the red flags that should trigger immediate pushback, and provides sample language you can adapt. Whether you are reviewing an NDA, an MSA, or a SaaS agreement, the frameworks here apply.

    What Is an Indemnification Clause?

    In plain terms, an indemnification clause is a contractual promise by one party to compensate the other for losses arising from specified events. It is the primary mechanism for allocating risk between contracting parties.

    Example in simple terms: “If you get sued because of something I did under this contract, I’ll cover your legal costs and any damages.”

    Indemnification clauses appear in virtually every commercial agreement: MSAs, SaaS subscriptions, employment agreements, vendor contracts, commercial leases, and M&A purchase agreements. They are the most negotiated clause type in transactional practice, and for good reason. A poorly drafted indemnification clause can expose your client to uncapped liability for events they cannot control.

    For a broader overview of contract risk analysis, see our contract red flags checklist.

    The Anatomy of an Indemnification Clause

    Every indemnification clause has four components. Weakness in any one of them creates risk.

    Who Indemnifies Whom

    • One-way indemnification: Party A indemnifies Party B, but not vice versa. Common in vendor agreements where the vendor bears more risk.
    • Mutual indemnification: Both parties indemnify each other for their respective obligations. Generally considered more balanced.
    • The trigger question: What event activates the indemnification obligation? A breach of representations? Third-party claims? Negligence? Willful misconduct?

    What Is Covered (Scope)

    The scope defines which losses trigger the indemnification:

    • Third-party claims (most common): Party A covers Party B when a third party sues Party B because of something Party A did
    • First-party losses (less common, more contentious): Direct damages between the parties themselves
    • Specific triggers: IP infringement, breach of representations, data breach, negligence, willful misconduct, regulatory violations

    Each trigger should be evaluated independently. IP indemnification, for instance, is standard in technology agreements because the vendor is in the best position to know whether its software infringes. Data breach indemnification has become essential post-GDPR and CCPA.

    Indemnification Procedure

    The procedural mechanics determine whether the indemnification clause actually works when you need it:

    • Notice requirements: How quickly must the indemnitee notify the indemnitor? In what format? What are the consequences of late notice?
    • Control of defense: Who selects the attorneys? Who makes litigation strategy decisions?
    • Settlement approval: Can the indemnitor settle without the indemnitee’s consent? (They should not be able to admit liability on your client’s behalf.)
    • Cooperation obligations: What must the indemnitee do to support the defense?

    Financial Limitations

    Indemnification does not exist in a vacuum. It interacts with other financial provisions:

    • Liability caps: Does the indemnification obligation sit inside or outside the overall limitation of liability? This is often the single most contentious negotiation point.
    • Carve-outs: Is indemnification carved out from the liability cap entirely? (Common for IP indemnification and data breach.)
    • Insurance backing: Do the insurance requirements in the contract align with the indemnification exposure?

    The 7 Indemnification Negotiation Points That Matter

    These are the provisions worth fighting over. Everything else is detail.

    1. Scope of Covered Claims: Narrow vs. Broad Triggers

    The issue: The breadth of the trigger language determines how much risk the indemnifying party assumes.

    Narrow (indemnitor-friendly): “Indemnify for third-party claims to the extent directly resulting from Indemnitor’s material breach of this Agreement.”

    Broad (indemnitee-friendly): “Indemnify for any losses arising out of or in connection with the services provided under this Agreement.”

    The phrase “arising out of or in connection with” is the broadest possible trigger. It captures claims with even a tangential relationship to the contract. Push for “directly resulting from” or “to the extent caused by” instead.

    2. Defend vs. Indemnify vs. Hold Harmless

    These three terms are not synonymous, despite being used interchangeably in many contracts. According to a Morgan Lewis analysis, the legal distinctions are significant:

    • Defend: Obligation to hire attorneys and manage litigation from the moment a claim is filed. This is the most immediate and expensive obligation.
    • Indemnify: Obligation to pay damages or losses after a judgment or settlement. Does not arise until the end of a case.
    • Hold harmless: The promise to absorb the consequences of a covered claim. Most courts treat this as duplicative of “indemnify,” but California courts have interpreted “hold harmless” as a distinct defensive right.

    Practical implication: If your client is the indemnitee, you want all three. If your client is the indemnitor, “indemnify” alone is less costly than “defend, indemnify, and hold harmless.”

    The ABA Litigation Section’s analysis of indemnity obligations emphasizes that drafters should be explicit about whether the duty to defend is included, rather than relying on courts to interpret ambiguous language.

    3. Knowledge Qualifiers

    The subtlety that changes everything:

    • “to the extent arising from” = broadest
    • “to the extent resulting from” = moderate
    • “to the extent directly caused by” = narrowest

    Also watch for:
    – “to [Party]’s knowledge” qualifiers (limits scope to known issues)
    – “material breach” vs. “any breach” triggers
    – “sole negligence” vs. “negligence” vs. “any act or omission”

    4. Notice Requirements

    Notice provisions seem procedural until someone misses a deadline. Key elements:

    • Timing: “Promptly” is vague and litigable. “Within 15 business days of becoming aware” is specific and enforceable.
    • Format: Written notice to a specific address or email? Or any reasonable communication?
    • Consequence of failure: Does late notice eliminate the indemnification obligation entirely, or just reduce it by any prejudice caused by the delay?

    Best practice: Negotiate for notice that reduces the obligation only to the extent the indemnitor was actually prejudiced by the delay. An absolute forfeiture for late notice is a trap.

    5. Control of Defense

    When your client is being indemnified:
    – Who picks the lawyers? The indemnitor usually controls the defense, but the indemnitee should have approval rights over counsel selection.
    – Who approves settlements? The indemnitee should never be forced to accept a settlement that admits liability on their behalf or includes non-monetary terms (like injunctions) they have not approved.
    – What if interests conflict? If the indemnitor and indemnitee have conflicting interests in the litigation, the indemnitee should have the right to retain separate counsel at the indemnitor’s expense.

    6. Liability Cap Interaction

    This is where indemnification negotiations get genuinely contentious, and where the most money is at stake.

    Inside the cap: The indemnification obligation counts against the overall limitation of liability. If the cap is $1M and an indemnification claim costs $800K, only $200K of cap remains for other claims.

    Outside the cap (carved out): The indemnification obligation is uncapped or subject to a separate, higher cap. This is standard for IP infringement, data breach, and confidentiality breach indemnification.

    The negotiation: Indemnitors want everything inside the cap. Indemnitees want carve-outs for the highest-risk items. The market compromise varies by contract type (see Section 5 below).

    For a deep dive on how liability caps work and interact with indemnification, see our limitation of liability guide.

    7. Survival Period

    How long does the indemnification obligation last after the contract expires or terminates?

    • Indefinite survival: Maximum protection for the indemnitee, maximum exposure for the indemnitor
    • Fixed period (common: 12-24 months post-termination): Balanced approach for general indemnification
    • Statute of limitations: Survival tied to the applicable statute of limitations for the underlying claims
    • No survival clause: Dangerous. Arguably, the indemnification dies with the contract.

    Best practice: Negotiate survival periods that match the realistic timeline for claims to surface. IP infringement and data breach claims can emerge years after contract termination and should survive longer than general breach claims.

    Indemnification Red Flags

    These provisions should trigger immediate pushback in any contract review. If you are using Clause Labs’s AI contract review, these are the types of issues the risk analysis will flag automatically.

    Red Flag Why It Is Dangerous
    Unlimited indemnification with no cap Theoretically infinite financial exposure
    One-sided indemnification for mutual risks Unfair risk allocation that may not survive judicial scrutiny
    “Arising out of or in connection with” trigger Broadest possible scope; captures tangential claims
    No notice requirements Allows ambush claims months or years later
    Indemnitor has no control of defense but must pay All the cost, none of the ability to manage it
    Indemnification survives indefinitely Open-ended exposure with no endpoint
    No duty to mitigate damages Indemnitee has no incentive to minimize losses
    Consequential damages included in indemnification scope Can dwarf direct damages by orders of magnitude

    Indemnification by Contract Type

    What constitutes “standard” indemnification depends entirely on the contract type and the relative bargaining positions of the parties.

    SaaS and Software Agreements

    • Standard: Vendor indemnifies customer for IP infringement claims. Customer indemnifies vendor for misuse of the platform.
    • Market position on IP indemnification: Universal. If a SaaS vendor will not provide IP indemnification, that is a major red flag.
    • Data breach: Increasingly carved out from liability caps. The data processor (vendor) typically indemnifies the data controller (customer) for breaches caused by vendor’s failure to maintain required security.
    • Typical cap interaction: IP and data breach indemnification carved out from the general liability cap, often subject to a separate “super cap” of 2-3x annual fees.

    For more on SaaS-specific risks, see our SaaS agreement review guide.

    Professional Services and MSAs

    • Standard: Mutual indemnification for third-party claims arising from each party’s breach, negligence, or willful misconduct.
    • Key issue: The service provider’s indemnification for professional errors (errors & omissions) and whether it overlaps with or replaces standard warranty remedies.
    • Insurance alignment: The indemnification scope should mirror the service provider’s E&O insurance coverage.
    • Subcontractor indemnification: The primary contractor should indemnify the client for subcontractor actions, and the subcontractor agreement should contain a back-to-back indemnification.

    Employment Agreements

    • Standard: Employer indemnifies employee for claims arising from authorized conduct in the scope of employment (similar to D&O indemnification for executives).
    • Employee to employer: Indemnification for breach of restrictive covenants, breach of confidentiality, or IP assignment violations.
    • Key issue: State law often limits the enforceability of employee-to-employer indemnification, particularly where the employee has limited bargaining power.

    Commercial Leases

    • Standard: Tenant indemnifies landlord for personal injury and property damage occurring in the leased premises. Landlord indemnifies tenant for common area issues.
    • Anti-indemnity statutes: Multiple states (including Texas, New York, and California) restrict or void certain indemnification provisions in construction and lease contexts. Always check your jurisdiction.
    • Insurance coordination: Indemnification obligations should align with the commercial general liability policies required under the lease.

    For a broader framework on how to systematically review contracts for these and other issues, see our guide on how to review a contract in 10 minutes.

    M&A Purchase Agreements

    • Standard: Seller indemnifies buyer for breach of representations and warranties. Buyer indemnifies seller for post-closing liabilities.
    • Key features: Baskets (deductible before indemnification kicks in), caps (often 10-20% of purchase price for general reps, 100% for fundamental reps), escrow accounts, and rep & warranty insurance.
    • This is the most complex indemnification context. M&A indemnification alone could fill an article. The principles here provide a foundation, but M&A-specific counsel is essential.

    Indemnification vs. Other Risk Allocation Mechanisms

    Indemnification does not work in isolation. Review it as part of the full risk allocation framework:

    Indemnification + Limitation of Liability: Often in tension. Ensure the contract explicitly addresses whether indemnification obligations count against the liability cap. Ambiguity here is the single most common source of indemnification disputes.

    Indemnification + Insurance: The indemnification scope should not exceed the indemnitor’s ability to pay. Insurance requirements should back up indemnification obligations. If a party indemnifies for data breach but carries no cyber insurance, the indemnification may be worthless.

    Indemnification + Warranty Remedies: Clarify whether breach of warranty claims flow through the indemnification or through a separate warranty remedy. Having both without coordination creates overlap and confusion.

    Indemnification + Liquidated Damages: They can coexist, but clarify the relationship. Liquidated damages typically address agreed-upon amounts for specific breaches (e.g., SLA failures). Indemnification addresses third-party claims and uncapped losses.

    Sample Indemnification Clause Language

    Balanced Mutual Indemnification

    Each party ("Indemnifying Party") shall defend, indemnify, and hold harmless the
other party and its officers, directors, employees, and agents ("Indemnified Party")
from and against any third-party claims, damages, losses, and expenses (including
reasonable attorneys' fees) to the extent arising from:
(a) the Indemnifying Party's material breach of this Agreement;
(b) the Indemnifying Party's negligence or willful misconduct; or
(c) the Indemnifying Party's violation of applicable law.

The Indemnified Party shall provide written notice of any claim within fifteen (15)
business days of becoming aware of such claim. Failure to provide timely notice
shall not relieve the Indemnifying Party of its obligations except to the extent
actually prejudiced by such failure.

    Annotations: This is a balanced starting point. The “to the extent arising from” trigger is moderate. The notice provision is specific but not punitive. The duty to defend is included explicitly.

    IP-Specific Indemnification (Vendor to Customer)

    Vendor shall defend, indemnify, and hold harmless Customer from any third-party
claim that the Services, as provided by Vendor and used by Customer in accordance
with this Agreement, infringe any United States patent, copyright, or trade secret.

Vendor's obligations under this section shall not apply to claims arising from:
(i) Customer's modification of the Services;
(ii) Customer's use of the Services in combination with materials not provided
    or approved by Vendor; or
(iii) Customer's use of the Services after Vendor has provided a non-infringing
     alternative.

    Annotations: The exceptions are standard and reasonable. They protect the vendor from liability for the customer’s modifications while covering the vendor’s core IP warranty.

    How AI Identifies Indemnification Issues

    Modern contract review AI evaluates indemnification clauses across multiple dimensions simultaneously:

    • Scope analysis: Is the indemnification one-sided for risks that should be mutual?
    • Missing provisions: Are notice requirements, defense obligations, or settlement procedures absent?
    • Cap interaction: Does the contract address whether indemnification counts against the liability cap?
    • Market comparison: How does this indemnification compare to standard provisions for this contract type?

    When we uploaded a mutual NDA with a one-sided indemnification clause to Clause Labs, it flagged the imbalance in 28 seconds, identified the missing notice procedure, and noted the absence of a survival period. That kind of pattern recognition across multiple interacting provisions is where AI contract review adds the most value.

    For a comparison of AI tools that perform this type of analysis, see our AI contract review tools guide.

    Want to see how AI handles the indemnification clause in your next contract? Upload it to Clause Labs for free — 3 reviews per month, no credit card required. Solo plan starts at $49/month for 25 reviews when you need more.

    Frequently Asked Questions

    What is the difference between indemnify, defend, and hold harmless?

    Defend requires the indemnifying party to hire attorneys and manage litigation from the moment a claim is filed. Indemnify requires payment of damages after a judgment or settlement. Hold harmless is treated as synonymous with “indemnify” by most courts, though some jurisdictions (notably California) treat “hold harmless” as a distinct defensive right. In practice, include all three terms when drafting for the indemnitee’s protection.

    Is indemnification the same as insurance?

    No. Insurance is a product purchased from a third-party carrier. Indemnification is a contractual obligation between the parties. They should work together: indemnification obligations should be backed by adequate insurance coverage, and insurance requirements should align with the scope of indemnification. An indemnification clause without insurance backing is only as good as the indemnitor’s balance sheet.

    Can you cap indemnification obligations?

    Yes, and many contracts do. Common approaches include setting a separate “super cap” for indemnification (e.g., 2-3x annual contract fees), tying the cap to available insurance coverage, or placing indemnification inside the general limitation of liability. The appropriate cap depends on the contract type, deal size, and relative bargaining power.

    What happens if the indemnifying party cannot pay?

    The indemnification obligation exists on paper, but collecting is a separate matter. This is why insurance requirements matter: even if the indemnifying party becomes insolvent, their insurance carrier may still cover the claim. In M&A contexts, escrow accounts and rep & warranty insurance address this risk directly.

    Should indemnification survive contract termination?

    Almost always yes. Claims covered by indemnification (IP infringement, data breach, negligence) frequently surface after the contract has expired. A minimum survival period of 12-24 months is standard for general indemnification. IP and data breach indemnification should survive for the applicable statute of limitations period.

    Is mutual indemnification always fair?

    Not necessarily. “Mutual” indemnification is fair only when both parties have roughly equal exposure. In a SaaS agreement, the vendor has IP risk and data handling risk. The customer has misuse risk. The scope of each party’s indemnification should reflect their actual risk profile, not a false equivalence.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

  • How to Set Up AI Contract Review at Your Firm in Under 30 Minutes

    How to Set Up AI Contract Review at Your Firm in Under 30 Minutes

    How to Set Up AI Contract Review at Your Firm in Under 30 Minutes

    The average solo lawyer spends 91 minutes on administrative overhead before even opening a contract. Setting up an AI contract review tool should not add to that number. It does not. With most modern platforms, you can go from zero to a completed AI-assisted contract analysis in less time than it takes to draft a standard engagement letter.

    According to the ABA’s 2024 Legal Technology Survey, AI adoption among solo attorneys increased 55.5% year-over-year, yet 13% of solos still say they don’t know enough about AI to form an opinion. This guide closes that gap. Here is how to set up AI contract review at your firm in 30 minutes, with no IT department, no data migration, and no learning curve worth mentioning.

    Try Clause Labs free — no credit card required

    The 30-Minute Setup: Timed Steps

    Minutes 1-5: Create Your Account and Get Oriented

    Start at your chosen platform’s signup page. For this walkthrough, we will use Clause Labs as the example.

    What you are doing:
    – Enter your email and create a password
    – No credit card required for the free tier (3 reviews per month, permanently free)
    – Verify your email
    – Set up your profile: firm name, primary practice areas, default jurisdiction

    Why this matters: Your jurisdiction setting affects how the AI scores state-specific risks. A non-compete flagged as “high risk” in California (where they are largely unenforceable under Bus. & Prof. Code Section 16600) would score differently in Florida (where they are enforceable under specific conditions per Fla. Stat. Section 542.335).

    The whole process takes less time than setting up a new Westlaw login.

    Minutes 5-10: Upload Your First Contract

    Choose a contract you have already reviewed manually. This is important — you want to compare the AI’s analysis against your own work, not fly blind.

    Step by step:
    1. Drag and drop a PDF or DOCX into the upload area (or paste the text directly)
    2. Select the contract type (NDA, MSA, employment, SaaS agreement) or let the AI auto-detect
    3. Click “Analyze”
    4. Wait 30-60 seconds while the AI processes

    What is happening behind the scenes: The AI parses the document, identifies individual clauses, classifies each clause by type, assesses risk on each clause against market standards, identifies missing provisions, and generates a structured risk report. For scanned PDFs, OCR processing adds an extra 30-60 seconds.

    Pro tip: Upload a contract where you remember the issues you flagged. You are going to compare your manual findings against what the AI catches.

    Minutes 10-15: Review Your First AI Analysis

    This is the moment most lawyers describe as the “aha moment.” Here is what you will see:

    Risk Score (X/10): An overall risk assessment for the contract. A 3/10 is relatively clean. An 8/10 means significant issues. The number itself matters less than the clause-by-clause breakdown beneath it.

    Clause-by-Clause Analysis: Each clause the AI identified, with:
    – A risk rating (Critical, High, Medium, Low, or Informational)
    – A confidence score showing how certain the AI is about its assessment
    – A plain-English explanation of why this clause is risky
    – Suggested alternative language where applicable

    Missing Clause Detection: Provisions the AI expected to find based on the contract type but did not. For NDAs, this might be a missing carve-out for independently developed information. For employment agreements, it might be a missing garden leave provision.

    Your comparison exercise: Check the AI’s findings against what you caught in your manual review. Most lawyers find the AI caught everything they did, plus 2-3 issues they missed — typically in definitions sections, buried cross-references, or missing provisions.

    Minutes 15-20: Customize Your Settings

    Now that you have seen what the AI can do, configure it to match your practice:

    Jurisdiction preferences: Set your primary state so risk scoring reflects local law. If you practice in multiple states, you can adjust per review.

    Contract type priorities: Tell the system which types you review most frequently (NDAs, MSAs, employment agreements). This helps prioritize your dashboard and refine the AI’s suggestions over time.

    Notification preferences: Choose how you want to receive alerts about completed reviews, especially useful if you run analyses while doing other work.

    Team setup (if applicable): On the Professional tier ($149/month, up to 3 users) or Team tier ($299/month, up to 10 users), you can invite associates or paralegals now.

    Minutes 20-25: Run a Second Contract

    Upload a different contract type than your first one. If you started with an NDA, try an employment agreement or a SaaS agreement.

    Why a second analysis matters:
    – AI performance varies by contract type — some types have deeper training data
    – You are building confidence with a second successful analysis
    – You can compare how risk scoring and missing clause detection differ across contract types
    – Two data points are always better than one for evaluating a tool

    At this point, you have invested 25 minutes and have two complete analyses to evaluate.

    Minutes 25-30: Design Your Ongoing Workflow

    This is where the setup becomes permanent. Decide where AI review fits into your existing process:

    Option A: AI First, Then Human Review (Recommended for Most Lawyers)
    Upload the contract immediately upon receipt. While the AI processes (30-60 seconds), grab a coffee. Review the AI’s output, then do a focused human review concentrating on the flagged issues and deal-specific context. This is the approach most lawyers settle into by end of week one.

    Option B: Human Scan First, Then AI for Deep Analysis
    Do a quick 5-minute read-through to understand the deal structure. Then upload to AI for the detailed clause-by-clause analysis. Compare your initial impressions against the AI’s findings.

    Option C: Parallel Review
    Run both simultaneously and compare. Useful during your first week when you are calibrating trust in the AI’s output. Less efficient long-term, but good for building confidence.

    Your 30-minute setup is complete. Bookmark the tool, or better yet, add it to your browser toolbar.

    What to Expect in Your First Week

    Based on data from Clio’s Legal Trends Report, the average contract review takes 2-3 hours when done manually. Here is how that changes across your first week of AI-assisted review:

    Day 1: Everything feels new. You will double-check everything the AI says against the source document. Each review takes about the same time as manual review because you are verifying. This is normal and appropriate.

    Day 2-3: You start trusting the clause identification. The AI consistently labels clauses correctly, and you spend less time verifying classifications. You still scrutinize every risk rating. Reviews drop to 45-60 minutes.

    Day 4-5: You develop a rhythm. AI does the scanning, you do the thinking. You spend your time on what actually requires a law degree: evaluating risks in context, advising the client on strategy, and deciding what to negotiate. Reviews drop to 20-30 minutes for standard contracts.

    By end of week: You wonder how you reviewed contracts without it. The shift from “reading every word looking for problems” to “reviewing AI-flagged issues with context” is significant.

    Telling Your Clients About AI

    You may not be required to disclose AI use in most jurisdictions for transactional work — but it is worth considering. For a detailed breakdown by state, see our guide to AI disclosure requirements for lawyers.

    If you choose to disclose, here is sample language for your engagement letter:

    “Our firm uses AI-assisted review tools as part of our quality assurance process. These tools assist with clause identification and risk analysis. All AI-generated analysis is reviewed, verified, and supplemented by attorney judgment before inclusion in any client deliverable.”

    This mirrors the approach recommended in ABA Formal Opinion 512, which addresses lawyers’ ethical obligations when using generative AI tools, including the duties of competence under Model Rule 1.1 and communication under Model Rule 1.4.

    If clients ask directly, frame it as a quality measure: “We use AI-assisted review tools to ensure no clause is overlooked — the same way we use legal research databases to ensure no case is missed.”

    Common First-Week Questions (and Honest Answers)

    “The AI flagged something I disagree with.”
    Good. That means you are supervising properly. AI tools err on the side of caution — they would rather over-flag than miss something. Your professional judgment is the final authority. For a structured approach to reviewing AI output, see our framework for supervising AI outputs.

    “The AI missed something I would have caught.”
    This happens. No AI tool catches everything, just as no human reviewer catches everything. The combination of AI-plus-human catches more than either alone. According to Gartner’s research on legal technology, AI-assisted review reduces missed issues by 40-50% compared to manual-only review, but that remaining 50-60% is why human review remains essential.

    “My contract type isn’t perfectly supported.”
    AI contract review tools perform best on common contract types (NDAs, MSAs, employment, SaaS). For unusual agreements — merger documents, complex licensing, multi-party joint ventures — the AI still provides useful clause identification and risk flagging, but your expertise becomes proportionally more important. The system learns from your feedback over time.

    “My associate or paralegal should be doing this, not me.”
    Absolutely — have them use it. AI review tools make junior staff faster and more consistent. Under ABA Model Rule 5.3, you remain responsible for supervising their work, but the AI adds a second safety net for quality control.

    When to Upgrade from Free

    The free tier (3 reviews per month) is enough to evaluate the tool. Here is when upgrading makes financial sense:

    Scenario Recommended Tier Cost Break-Even
    Reviewing 5-25 contracts/month as a solo Solo ($49/mo) $49/month Less than 12 minutes of billable time at $250/hour
    Small firm with 2-3 reviewers, up to 100 contracts/month Professional ($149/mo) $149/month Less than 36 minutes of billable time at $250/hour
    Firm needing batch review, obligation tracking, Clio integration Team ($299/mo) $299/month One contract review that previously took 3 hours

    At $49/month, the Solo tier costs less than a single hour of most lawyers’ billable time. If AI saves you even one hour per month — and it will save far more than that — the tool pays for itself on day one.

    See pricing and start your free account

    Security: What Happens to Your Client’s Data

    Before uploading any client document, you need to know how the tool handles data. This is not just good practice — it is your obligation under Model Rule 1.6 (confidentiality of information).

    Questions to ask about any AI contract review tool:
    – Does it retain my documents after analysis? (Clause Labs: no long-term data retention)
    – Does it train its AI on my uploads? (Clause Labs: never)
    – Is data encrypted in transit and at rest? (Clause Labs: yes, both)
    – Is there a SOC 2 compliance pathway? (Clause Labs: on our roadmap)

    For a deep dive on data security across AI legal tools, see our guide on how to use AI ethically for contract review.

    Frequently Asked Questions

    Do I need any technical skills to set up AI contract review?
    No. If you can attach a file to an email and read a report, you can use AI contract review. There is no code, no configuration files, no IT setup. The most technical step is dragging a file into a browser window.

    Does it work on Mac, Windows, and mobile?
    Yes. Modern AI contract review tools are browser-based, so they work on any device with a web browser. Full analysis is available on desktop (Mac and Windows). Mobile viewing works for reviewing results, though uploading and in-depth review are better on a larger screen.

    What if I don’t like it?
    The free tier requires no credit card and no commitment. Try it on three contracts. If it doesn’t add value, you’ve lost 30 minutes of setup time and nothing else. If you upgrade and change your mind, cancel anytime.

    Is my data safe?
    Look for tools that offer encryption in transit and at rest, zero training on user data, no long-term data retention, and a clear privacy policy. These are the baseline requirements recommended by the ABA and state bars for cloud-based legal tools.

    How does AI contract review affect my malpractice insurance?
    Most malpractice carriers have not issued specific exclusions for AI-assisted contract review. The key is documentation: show that you used AI as a supplement to (not replacement for) your professional judgment, and that you supervised the output. According to the ABA’s guidance, the risk profile is similar to using any other legal research or document review technology.

    This article is for informational purposes only and does not constitute legal advice. Consult a qualified attorney for advice specific to your situation.

    Start your 30-minute setup — sign up free at Clause Labs