Freelance contract red flags rarely announce themselves. Most bad client contracts look like professional documents with reasonable-sounding language, and that’s exactly why they’re dangerous. The clauses that carry the most risk are often buried in boilerplate that nobody reads, using phrases that sound standard until you understand what they actually transfer. Reading a client-sent contract is a skill, and it’s faster to learn than most freelancers realize. You’re looking for specific language patterns, not legal expertise.
Scope Language That Functions as a Blank Check
Vague scope language is the most common contract problem and often the most benign in intent, clients copy it from templates without thinking through the implications. But the implications are real.
Watch for phrases like “and any related work,” “as Client may reasonably request,” “at Client’s direction,” and “including supporting materials.” These phrases extend the scope of your obligations beyond what you discussed and quoted. A contract that says “Contractor shall provide copywriting services for the website, and such other content as Client may reasonably request” has just made your agreed deliverables a starting point, not a ceiling.
What you want instead: scope defined by specific deliverables, specific formats, and specific quantities. “Five website pages: homepage, about, services, team, and contact, each up to 600 words, delivered as Google Docs” is a scope. “Website copywriting as needed” is not.
When you see vague scope language, propose a deliverables list as an attachment to the contract. “I’d like to add a scope exhibit that lists the specific deliverables, this protects both of us from scope disputes.” Most reasonable clients accept this readily. The ones who resist open-ended scope should make you wonder why.
IP Grabs: When They Want to Own Everything
The standard IP grab looks like this: “All work product, including but not limited to drafts, concepts, methods, and tools used in the creation thereof, shall be the sole property of Client.”
Read that again. “Methods and tools used in the creation thereof.” This language doesn’t just claim ownership of the deliverable, it claims ownership of your process. A developer’s reusable code libraries. A designer’s custom brush sets. A writer’s research frameworks. If a court gave this clause full effect, you’d be assigning not just the project output but the instruments of your practice.
The background IP variant is even more aggressive: “including any pre-existing intellectual property incorporated into the deliverables.” This clause reaches back before the project started. If you used any element you created prior to this engagement, a template, a plugin, a design element, the client is claiming ownership of it.
For both variants, the negotiation is the same: limit the assignment to “final deliverables as specified in the scope of work” and add explicit carve-outs for pre-existing materials and tools. A counter-clause: “The Client acknowledges that certain pre-existing materials, tools, and methodologies owned by Contractor may be incorporated into the deliverables. Contractor grants Client a license to use those elements as incorporated into the deliverables, but retains all ownership rights in such pre-existing materials.” For a full breakdown of how IP assignment works and what you’re actually transferring, the assignment versus license distinction is worth understanding before you negotiate.
Unlimited Liability and Indemnification Clauses
This is the highest-stakes clause most freelancers sign without understanding it. The language looks like this: “Contractor shall indemnify and hold harmless Client from any and all claims, damages, losses, and expenses, including attorneys’ fees, arising out of or related to Contractor’s performance under this agreement.”
In plain English: if anything goes wrong, anywhere, for any reason related to your work, you’re responsible for the client’s legal costs, settlements, and damages. Not just for your direct mistakes. Not just up to the project fee. Any and all claims, unlimited.
The exposure is asymmetric in a way that should be alarming. You invoice a client for $5,000. Their marketing campaign uses your copy, performs badly, and their investors sue them for misrepresentation. Under an unlimited indemnification clause, the client can pass their legal costs to you. The project fee and the liability exposure have no relationship.
What you want is a mutual, capped indemnification clause. Mutual means both parties indemnify each other for their own actions. Capped means your liability is limited to the project fee paid, or to a specific amount. A counter-clause: “Each party’s liability under this agreement is limited to the total fees paid by Client to Contractor in the three months preceding the claim.” Clients who use standard legal templates often accept this counter without much resistance, their lawyers know capped mutual indemnification is the market standard.
The clause to refuse outright: unlimited personal liability. As an individual freelancer, unlimited liability exposure is not insurable at a cost that makes sense relative to typical project fees. If a client’s contract cannot be negotiated to include a liability cap, that’s a deal-breaker, not a red flag.
Unilateral Termination Without Compensation
This clause appears in many forms, but the core language is: “Client may terminate this Agreement at any time, for any reason, without notice, and without any further obligation to Contractor.”
What this means: the client can cancel the project the morning after you’ve completed three weeks of work, owe you nothing beyond what was invoiced prior, and face no consequence for the disruption to your pipeline or the income gap the cancellation creates.
Paired with net 30 payment terms and no deposit, this clause means a client can: start a project, receive weeks of work, cancel before any payment is due, and walk away owing nothing for the work completed since the last invoice.
The counter: a kill fee clause and a payment-for-work-completed provision. “In the event of termination by Client, Client shall pay for all work completed to date at the agreed rate, plus a kill fee of [X]% of the remaining uncompleted contract value as compensation for opportunity cost and disruption.” Thirty percent is a reasonable kill fee for most project contexts. The percentage is negotiable, the concept should not be.
Also negotiate for a notice period. “Client may terminate with 14 days’ written notice” gives you time to fill the pipeline gap rather than absorbing the loss immediately.
Payment Terms That Favor the Client
The most dangerous payment clause: “Payment is contingent upon Client’s complete satisfaction with all deliverables.”
In legal terms, this is a condition precedent that may never be satisfied, because subjective satisfaction is impossible to prove or compel. Courts often strike these clauses down, but that requires litigation, and the cost of litigation usually exceeds the invoice value.
Adjacent problem clauses to watch for:
“Payment shall be made upon approval by Client’s management”, introduces a third-party approval chain you have no visibility into or control over.
“Client reserves the right to withhold payment pending resolution of any dispute”, gives the client a unilateral veto over payment whenever they want to use it.
“Client may deduct costs of any errors or revisions from the final invoice”, gives the client the right to make unilateral financial adjustments with no defined limit or process.
Counter with objective approval language: payment triggers tied to delivery of specified files in specified formats, with a deemed-approval clause if no written objection is raised within a defined review period. “Payment is due upon delivery of the deliverables as specified in Exhibit A. If Client does not provide written objection within 10 business days of delivery, deliverables are deemed accepted.” Subjective satisfaction as a payment condition is something to remove entirely, not modify.
Confidentiality Clauses That Go Too Far
Standard confidentiality provisions are reasonable, you agree not to disclose the client’s business information, strategic plans, or internal data. That’s a legitimate and proportionate ask.
What goes too far: “Contractor agrees to maintain the existence of this agreement in strict confidence and shall not disclose the nature, scope, or content of services provided hereunder to any third party.”
This clause means you can’t mention the client exists. You can’t name them in a portfolio, tell a prospective client you’ve worked in their industry, or reference the engagement in any professional context. Combined with a broad IP clause, it means you created work you can’t show and signed a contract you can’t reference.
Non-disparagement clauses buried in confidentiality sections are worth reading carefully. “Contractor agrees not to make any statements, written or oral, that disparage Client’s products, services, or business practices” can prevent you from warning other freelancers about a genuinely problematic client. This is a clause written entirely for the client’s benefit, with no reciprocal protection for you.
For portfolio rights specifically, ensure the confidentiality clause includes an explicit carve-out: “Notwithstanding the foregoing, Contractor retains the right to display the final deliverables in Contractor’s portfolio and for self-promotional purposes.” If you can’t get that carve-out, you’ve identified a clause worth pricing into the engagement. What to do when a client sends a bad freelance contract covers how to negotiate these specific terms without losing the project.
Non-Compete Clauses That Overreach
Some non-compete provisions in client contracts are reasonable: you won’t approach the client’s specific named customers to offer competing services during and for some period after the engagement. That protects a legitimate business interest and is usually proportionate.
The overreaching version: “Contractor agrees not to provide services substantially similar to those provided hereunder to any competitor of Client in the [industry] sector for a period of 24 months following termination.”
If you’re a web developer, this clause might be claiming you can’t work with any company in e-commerce for two years. If you’re a copywriter, any competitor in consumer goods. These clauses are often unenforceable in many jurisdictions precisely because they’re disproportionate, but “unenforceable” doesn’t mean costless. The client can still send a cease-and-desist letter or attempt to enforce it, and you’ll absorb the cost of responding regardless of outcome.
The counter is a narrow, specific non-solicitation clause in place of a broad non-compete: “Contractor agrees not to directly solicit Client’s named customers listed in Exhibit B during the term of this agreement.” That’s a legitimate protection. Restricting your ability to work in your entire industry is not.
Red Flag vs. Deal-Breaker
Not every bad clause is worth walking away from. The distinction is whether the clause is negotiable and whether a reasonable client will accept a counter.
Red flags worth negotiating: vague scope language, broad IP transfers that include pre-existing tools, one-sided termination without a kill fee, payment terms tied to subjective approval, overly broad confidentiality with no portfolio carve-out.
Deal-breakers worth walking from: unlimited personal indemnification that can’t be capped, payment contingent on satisfaction with no objective standard, background IP grabs that claim ownership of pre-project work, and non-competes that restrict your ability to work in your core field.
Raising concerns professionally is standard business practice. “I’ve reviewed the contract and have a few items I’d like to discuss, I’ve marked up the IP and indemnification sections specifically, and I think we can find language that works for both sides” is a reasonable professional communication, not a confrontation. Clients who treat contract negotiation as disqualifying are clients whose contracts contain language worth being concerned about.
The behavioral red flags that precede the contract stage are a companion to the contract red flags covered here, what a client does before signing tells you nearly as much as what they write into the agreement. And once you know what a bad clause looks like, building your own contract around the clauses that actually protect you becomes a different exercise, you know what you’re protecting against because you’ve seen it.