Skip to main content
In this article
  1. The screen is the record
  2. Four patterns and how they fare
  3. Why arbitration clauses live or die on this question
  4. Limits on what the terms can achieve
  5. Proving what the user saw
  6. Common questions
  7. Reviewing a signup flow, in order
Technology, Privacy & Media

Terms of Service and Clickwrap: When an Online Agreement Binds

Courts deciding whether online terms bind rarely argue about the clauses. They argue about the screen: what it showed, where the link sat, and what the user had to click.

A signup screen on a laptop showing a checkbox beside a link to terms and conditions
Original illustration by Beacon Legal Newsroom.

Key points

  • Clickwrap, which requires an affirmative click next to the terms, is routinely enforced, while browserwrap with no assent step frequently fails.
  • Sign-in wrap outcomes turn on screen design: link prominence, colour contrast, proximity to the button, and whether notice appears where the user acts.
  • Arbitration clauses and class waivers depend entirely on formation, which is a question of state contract law even under the federal arbitration statute.
  • Some terms are void by statute regardless of assent, and businesses cannot contract around statutory duties such as advertising disclosure rules.

When a court decides whether a company's terms of service bind a user, it spends very little time on the terms themselves. It spends its time on the screen. What did the page look like when the user signed up? Was there a link to the terms, and was it visible? Did the user have to do something — tick a box, click a button beneath a notice — or could they have finished the transaction without ever meeting the agreement? Formation needs reasonable notice and manifest assent, and online that means interface design decides the case.

The screen is the record

Traditional contract doctrine asks whether the parties agreed. Online, that collapses into a design question. A user who scrolls past a grey nine-point footer link has not had meaningful notice. A user who clicks a button labelled "I agree to the Terms of Service", with the link immediately above it, plainly has.

Everything in between is argued on specifics: the size and colour of the link text, how close the notice sits to the action button, whether the user had to scroll to see it, whether competing elements pulled attention away, and whether the notice appeared on the screen where the user acted.

This is why disputes here are unusually evidentiary. The company has to reconstruct an interface as it existed at a past moment, for a particular user, on a particular device. Courts have refused to enforce agreements where the company could not show what the user saw — not because the terms were unfair, but because nobody could prove the screen.

Four patterns and how they fare

The vocabulary is informal and courts use it loosely, but the categories are useful shorthand for how much assent an interface actually collects.

Common assent patterns and their typical treatment
PatternWhat the user doesTypical outcome and why
ClickwrapTicks a box or clicks a dedicated "I agree" control next to or above the termsRoutinely enforced. A discrete act of assent tied to a visible link
Sign-in wrapClicks "Create account" under a line saying that doing so accepts the termsGenuinely mixed. Turns on prominence, contrast, proximity and clutter
BrowserwrapNothing. Terms are linked somewhere on the site, often the footerFrequently fails. No assent step and often no reasonable notice
Terms sent after the transactionReceives terms by email or in the box after payingWeakest position. Assent to terms first seen after the deal is hard to show

The middle row is where most litigation lives. Sign-in wrap is the default for consumer apps because it reduces friction, and its enforceability is a spectrum, not a rule. Two such screens can produce opposite results: one with the notice in dark text above a lone button, another with the same sentence in light grey below three login options and a banner.

Practical step: If enforceability matters — and it does if you rely on an arbitration clause, a liability limit or a license grant — the cheapest fix is moving from sign-in wrap to a real checkbox. It costs a little conversion and removes the most contested issue.

Why arbitration clauses live or die on this question

Arbitration clauses and class-action waivers are why online formation gets litigated so heavily. They are usually the most valuable clauses in the document and depend entirely on formation: if the agreement was never formed, neither was the arbitration clause, and there is nothing to compel.

The layering is worth stating carefully. The Federal Arbitration Act makes written arbitration agreements enforceable and limits the grounds for refusing to honour them. But whether an agreement exists at all is a question of ordinary state contract law. The federal statute governs how such an agreement is treated; state law governs whether one was made. That is why the same clause can be compelled in one state and refused in another on identical screens.

The practical consequence is that a motion to compel arbitration becomes a mini-trial about the signup flow, with records of the interface, the account creation and any later changes to the terms all central. Where the parties negotiate instead, the process looks like our overview of mediation and court-ordered settlement conferences.

Limits on what the terms can achieve

Even a perfectly formed agreement does not give a business unlimited drafting power.

Clauses that let one side rewrite the deal

Nearly every set of online terms reserves the right to change them at any time. Courts scrutinise those clauses, and some have refused to enforce changes imposed without notice and a genuine opportunity to reject. The reasoning is straightforward: a promise one side can redefine at will can look like no promise at all. A company relying on a modified term does better if it can show conspicuous notice and that continued use afterwards was a meaningful choice.

Unconscionability

Unconscionability is a state-law doctrine with two halves. The procedural half looks at the circumstances of formation — surprise, hidden terms, absence of real choice. The substantive half asks whether the terms are unreasonably one-sided. Most states require some of both, but the weighting differs, and a clause struck down in one state may survive next door.

Terms that are void regardless of assent

Some clauses fail however carefully the user assented, because a statute says so. The federal Consumer Review Fairness Act voids clauses in form contracts that bar or penalise consumers for writing honest reviews. That is a rule about the term itself, not about notice, and clicking "I agree" does not rescue it — as discussed in our piece on online reviews, opinion and platform issues.

The broader principle is that businesses cannot contract around statutory duties. Terms do not displace advertising disclosure obligations, and a clause purporting to relieve a company of them accomplishes nothing — see the FTC's advertising and marketing guidance and business guidance hub. Billing works the same way: burying a recurring charge in the terms runs into the rules in our explainer on negative-option billing and canceling recurring charges.

State privacy statutes add another overlay. Where terms purport to authorize data collection or sale, state law may require notice and opt-out mechanisms independent of the contract; the California Attorney General's privacy pages show one regulator's approach. As of mid-2026 the number of states with comprehensive privacy statutes keeps growing, so this overlay follows the user's residence, not the contract.

Proving what the user saw

Because the fight is about notice, the record a company keeps is often worth more than the wording it drafts.

  • A dated, versioned archive of the terms, showing which version was live on a given day.
  • Captures of the signup and checkout screens as they appeared in each release, desktop and mobile.
  • Release records tying interface changes to dates, so a screen matches a specific signup.
  • Account-level logs of the acceptance event: timestamp, version presented, and device or client used.
  • Records of any later notice of modified terms, how it was delivered, and any acknowledgement.
  • Evidence of continued use after that notice, where the argument rests on acceptance by conduct.

Terms do work outside consumer disputes too. They are the main tool used against automated collection of site data, where contract, access and copyright arguments overlap — covered in our analysis of web scraping and automated data collection. In business software deals the same discipline extends to protections such as source code escrow in vendor agreements. Consumers approaching from the other side can start with the FTC's consumer information site.

Common questions

I never read the terms. Am I still bound by them?

Usually yes. Contract law does not require that you read an agreement, only that you had reasonable notice of it and did something manifesting assent. Ticking a box beside a visible link is generally enough. Not reading matters only indirectly, as part of an argument that the notice was inadequate or that a term was so surprising and one-sided that unconscionability reaches it.

Where does the link to the terms have to sit on the page?

There is no fixed rule, but proximity and prominence carry the analysis. Notice on the same screen as the action, close to the button, in text that contrasts with the background and does not compete with promotional elements, is the pattern that survives challenge. A footer link on an unrelated page fails. Mobile layouts are judged on what the phone displayed.

Can a company change its terms after I sign up?

Many terms reserve that right, and courts scrutinise how it is exercised. Changes imposed silently, with no notice and no chance to decline, are the most vulnerable. Companies in a stronger position give conspicuous notice, explain what changed, and leave a real option to stop using the service. Whether continued use amounts to acceptance is a state-law question with different answers in different courts.

The company says I agreed but cannot show the old signup screen. What happens then?

That gap is often decisive. The party asserting the agreement carries the burden of establishing formation, and reconstructing a past interface takes versioned archives, release records and account-level acceptance logs. Where a company can produce only today's screen, or a generic template, courts have declined to find that a specific user had reasonable notice. Missing records defeat enforceable terms more often than bad drafting.

Reviewing a signup flow, in order

  1. Open your own signup on a phone. Look at what shows above the button, at real size, on a real device.
  2. Find the assent act. Name the precise moment a user manifests agreement. If you cannot, you probably have browserwrap.
  3. Check contrast and proximity. The notice and link should be legible and adjacent to the control pressed.
  4. Start versioning the terms. Archive each version with the dates it was live, outside the system serving the current page.
  5. Log the acceptance event. Record which version was presented, when, and on what client.
  6. Audit the clauses against statute. Remove anything void by law regardless of assent, and stop using terms to displace statutory duties.
  7. Write the modification process down. Decide how users learn of changes and what a real opportunity to decline looks like.

Sources

  1. FTC — business guidance hub
  2. FTC — consumer information site
  3. California Attorney General — state privacy authority, CCPA pages
  4. FTC — advertising and marketing guidance for businesses

This is general information, not legal advice. Beacon Legal News is a publication, not a law firm, and reading it creates no attorney–client relationship. Law differs by state and changes; check the linked primary sources or speak with a licensed attorney in your jurisdiction before acting.

Beacon

Beacon Legal Newsroom

Beacon is an independent legal-information publication. Articles are researched against primary sources and revised when the law moves. How we source · Corrections