Skip to content
Expert Guide Series

What Happens if Users Don't Read My Terms of Service

Almost nobody reads your app terms of service. You already know that. The scroll bar moves in one fast drag, the button gets tapped, and the person is through to the thing they opened the app for in the first place. Product teams tend to treat this as a legal worry, and it's a reasonable one, so let's deal with it straight away. A user who agreed without reading is still bound by what they agreed to, as long as you put the agreement in front of them properly.

Nobody reads the terms, and everybody reads the screen that holds them.

That settles the legal half of the question and leaves the half that actually affects your numbers. Your terms screen sits at the exact moment a person decides whether to carry on or back out. They're not reading the clauses, but they are reading the screen, the tone, the length, the ask, and the sense of whether this company is being straight with them.

We spend most of our time on the behavioural side of that moment, working out where people hesitate and why. So this piece covers both. What makes an agreement hold, what breaks it, why copying somebody else's document is a bad idea, and what your users are responding to instead of the words.

The legal position: yes, they are still bound by terms they did not read

Courts in the UK, the US and across the EU have landed in roughly the same place. Someone who taps "I agree" has agreed, whether or not they read a line of it. The logic is practical. If genuine reading were the test, almost no online contract would survive, because almost nobody reads.

What courts do care about is whether the user had a fair chance to read. Was the agreement put in front of them at the point of signing up? Was the link obvious and did it work? Did they have to take a deliberate action to accept? Those questions decide far more cases than the wording of clause 14.

Where the answer gets less comfortable

Consumer protection law sits over the top of all of this, and it doesn't care that someone clicked. A term that's unfair to a consumer can be struck out even when the acceptance was perfect. Hiding something significant in dense text makes it more likely to be treated as unfair, not less.

We're a design consultancy rather than a law firm, so the document itself wants a solicitor who knows your jurisdiction and your business model. What we can tell you about is the screen it lives on, and that's the part most teams hand to nobody.

What makes app terms of service legally binding (and what breaks it)

The strength of an agreement comes down to how clearly you presented it and how deliberately the user accepted it. Those two things, more than anything in the document, decide whether it holds.

How app terms of service are usually presented

MethodWhat the user doesHow well it holds
ClickwrapTicks a box or taps "I agree" next to a working linkStrongest
Scroll-wrapScrolls through the text before the accept button activatesStrong, and slower
Sign-in wrapReads a small line under the button saying that continuing means agreeingDepends entirely on how visible that line is
BrowsewrapNothing, the link sits in a footerWeakest, often fails

Four things break an otherwise sound agreement. A link that's grey on grey, or small, or below the fold. A pre-ticked box, which removes the deliberate act that makes acceptance meaningful. Changing the terms later without telling anyone and without asking for fresh agreement. And a mismatch between what the document says and what your product does, because a judge reading both will notice.

Keep a record of who accepted which version and when. If you ever need to rely on the agreement, you'll need to show that specific person saw that specific wording.

UX/UI design built around real psychology

We design app interfaces around how people actually think and behave. User research, psychology-driven UX/UI design and technical specs delivered as one complete package.

See how we work Get started

No commitment

Can you copy another app's terms and conditions?

No, and the reasons stack up quickly. A terms document is a written work with an owner, so copying one is copyright infringement. Law firms draft these and bill for them. Taking that work and dropping your name on it is the same as lifting their marketing copy.

The commercial reason is worse than the legal one. Their terms describe their company, their data practices, their refund policy, their subscription mechanics and their jurisdiction. A skincare subscription and a hotel booking platform have almost nothing in common at the clause level. Copy one into the other and you've made promises you can't keep, and granted yourself permissions you never built for.

Copied terms describe somebody else's product, which means they describe promises you can't keep.

There's a trust cost too. Plenty of users never read a word, but the ones who do tend to be the ones who care most, and they're often the ones who'd pay you the most. Finding another company's name in clause nine tells them exactly how much thought went into the rest of the product.

Use other apps the way you'd use a competitor review. Look at structure, at what they've chosen to make prominent, at how they explain data use in plain language. Then get your own drafted around how your product actually works.

Why almost nobody reads terms of service, and why that will not change

Reading the full document takes real effort, and the user gets nothing back for it. They can't negotiate a clause. They can't accept most of it and decline the rest. The only outcome of reading carefully is either agreeing anyway or not using the app, and they already decided they want the app. Skipping is the rational move.

Survey work backs up what the behaviour suggests. A Brookings Institution survey found that nearly three-quarters of online users say they rarely read business terms of service, scrolling to the bottom and clicking "I Agree". That's self-reported, so the real number is probably less flattering.

So the honest design position is to stop treating the document as a communication channel. It's a legal record. Anything a user genuinely needs to know before they commit has to appear somewhere they'll actually look, in words they'll actually process, at the moment it matters to them.

Write a short plain-language summary of the three things your terms ask of the user, and put it on the screen itself. Keep the full document one tap away for anyone who wants it.

That summary isn't a replacement for the terms and conditions for apps you still need drafted. It's the version that does the communicating while the legal version does the recording.

What users read instead of your document

People make a judgement about whether you're trustworthy long before the agreement screen, and they make it from much smaller signals. How much you ask for at sign-up. Whether the price on screen matches the price at checkout. Whether a permission prompt explains itself. The tone of your microcopy. Whether anything feels deliberately vague.

Language level does more here than teams expect. On a genetics product we worked on, we found the usable range was narrower than anyone assumed. Make the language too simple and people distrust it, because they start to suspect there's no real science underneath. Go too technical and too medical and they disengage and stop part way through.

What worked was making the science credible but optional. Plain language carried people through the flow, and the medical depth sat there to be explored by anyone who wanted it. The rigour was visible enough to lend credibility without becoming a condition of using the product.

Your terms sit in the same band. Nobody wants to read 4,000 words of defined terms, and a two-line summary that says "we respect your privacy" reads as though you've got something to hide. A short, specific explanation of what you collect, what you do with it and what you never do with it will be read far more often than the document it summarises.

Asking for too much, too early: the location-sharing drop-off

We worked on a map-based fitness social network that connected people for runs and cycle rides. Users were dropping off at one clear point, the moment the product asked them to share their precise location with a potential match. The terms screen wasn't the problem. The sequence was.

We were asking for that location before anyone had the chance to have a conversation or learn anything about the other person. Precise location is a high-trust disclosure, and it was being requested from a stranger on the strength of a map pin. People did the sensible thing and stopped.

What we changed was the order. Instead of divulging an exact location, the product showed an approximate proximity, something like within 200 to 500 metres, so you knew there was somebody nearby you could run with. From there you could look at their profile and talk to them through the built-in messaging. Once you felt confident enough, you shared your precise location and arranged to meet.

Conversion improved because the flow finally matched how human trust actually builds. Nothing was removed from the product and no clause was reworded. The high-stakes ask simply moved to a point where people were ready for it.

Map every point where your product asks something of the user, then rank those asks by how much trust each one requires. Anywhere a high-trust ask sits early in the flow, look at your drop-off there first.

Fee ambiguity: why showing more built more trust on a travel booking product

On a travel booking product, we decided early on to wrap the platform's Stripe booking fee into the total price. The reasoning seemed sound. Travellers want one clean number, so give them one clean number and keep the screen quiet.

What we found was the opposite of what we expected. Users expected to see a platform fee as a line item, because that matches their mental model of how apps and platforms work. By not showing it, even though it was already covered inside the cost, we created a worry that an extra fee would appear later in the process. The silence read as something being withheld.

So we broke everything out and showed exactly what each fee was. People had far more confidence despite having more to read.

Showing every fee gave people more to read and left them with more confidence.

We saw the same pattern on a marketplace checkout project. Confusion about whether a platform fee was added on top of a price or already included in it produced measurable hesitation and drop-off, and the fee amounts were small. The size of the money wasn't driving the behaviour. The ambiguity was. Even a minor cost that people can't quite account for is enough to make them stop at the point of paying.

What the terms screen itself signals in three seconds

Look at how Headspace opens. There's no sign-up gate and no diagnostic questionnaire, and there's no "how are you feeling today?" prompt, which would presume an intimacy the user hasn't worked up to yet. It removes the two things that most reliably break trust on a health app by asking for nothing at all up front.

Your agreement screen is read the same way, in about the same amount of time. Before anyone processes a single clause they've registered the length of the scroll, how many things they're being asked to accept at once, whether the accept button is obviously bigger and brighter than everything else, and whether the wording sounds like a person or a legal department protecting itself.

Three patterns do the most damage. Bundling marketing consent into the same tick box as the terms, which reads as a trick. Hiding the privacy policy link while the accept button glows. And presenting an enormous wall of text at the very first screen, before the user has any reason to care about your company.

A screen that asks clearly, separates the optional from the required and looks like it expects to be read will carry more people through than one engineered to be skipped.

Reading hesitation at the point of agreement

Most teams can't see hesitation because their analytics aren't granular enough to show it. Funnel completion tells you how many people got through. It says nothing about what the ones who got through went through to do it.

What's worth capturing is time on screen, people entering a screen then leaving it and coming back, and scrolling behaviour, particularly repeated up-and-down scrolling through terms or fee detail. That pattern signals either a lack of comprehension or a lack of trust, and both are fixable once you can see them.

Hesitation only means something where you're asking for something. Someone pausing on a content feed is just reading. Someone pausing while being asked for their photo library, their contacts, their card details or their agreement is doing a calculation, and the higher the stakes of the ask, the more attention that pause deserves.

Before you redesign anything, work out whether the hesitation is a trust problem, a comprehension problem or a communication problem. The fix is different in each case, and the wrong fix makes a comprehension gap look like a trust gap forever.

Testing won't surface this on its own. In a usability session people assess a payment or data disclosure rationally, because no money is leaving their account. Live analytics tell a different story. People are feeling okay with stuff until they're not, and the emotional state at a real moment of commitment is where something that looked fine erodes trust just enough for them to bail out.

A practical checklist for terms and conditions for apps

Here's the order we'd work through it, with the legal and behavioural parts handled side by side rather than one after the other.

  1. Get your own document drafted by a solicitor who knows your jurisdiction, your business model and your data practices.
  2. Present it with clickwrap, a visible working link and an empty tick box the user has to act on.
  3. Separate marketing consent from acceptance of the terms, always, in two distinct controls.
  4. Write a short plain-language summary of what you collect, what you do with it and what you never do with it.
  5. Log the version accepted, the date and the user, and ask again when you make a material change.
  6. Move every high-trust request to the latest sensible point in the flow.

Then instrument the screen. Time on it, re-entries, scroll behaviour and drop-off at the accept button are the four numbers that tell you whether the moment is working.

Answering one more question is worth the effort too. Does anything in the document contradict what the product actually does? Mismatches between a terms page written eighteen months ago and a product that's moved on since are common, and they weaken the agreement at exactly the moment you'd need it to hold.

Conclusion

The legal answer is the easy part. Your users are bound by terms they didn't read, provided you presented them fairly and they accepted deliberately, and copying somebody else's document puts both of those things at risk along with your copyright position. Get it drafted properly, present it honestly, keep the records.

The part that changes your numbers sits elsewhere. On the fitness social network it was the order of the asks. On the travel booking product it was a fee we'd helpfully hidden inside a total. On the marketplace checkout it was ambiguity over a few pounds. None of those were clause problems, and no amount of redrafting would have found them.

So treat the document as the legal record it is, and treat the moment of agreement as a design problem with real behaviour attached. Ask for less, earlier. Explain more, plainly. Watch where people stop and work out why before you change anything.

If you want help finding where trust is breaking in your flows, and what to do about it, let's talk about your sign-up and agreement flow.

Frequently Asked Questions

Are my terms of service still binding if users never read them?

Yes. Courts in the UK, the US and the EU have broadly agreed that tapping "I agree" counts as agreement, whether or not a single line was read. What matters is that the user was given a fair chance to read the terms before accepting.

What does a court actually look at when deciding whether an agreement holds?

It looks at how clearly you presented the terms and how deliberately the user accepted them. That means asking whether the agreement appeared at the point of sign-up, whether the link was obvious and working, and whether the user had to take a positive action to accept.

Which method of presenting terms is the strongest?

Clickwrap is the strongest, where the user ticks a box or taps "I agree" next to a working link. Scroll-wrap also holds well but slows people down, sign-in wrap depends entirely on how visible the notice line is, and browsewrap, with just a footer link, is the weakest and often fails.

What are the most common mistakes that break an otherwise sound agreement?

Four things tend to do the damage: a link that is hard to see because it is small, low contrast or below the fold, a pre-ticked box that removes the deliberate act of accepting, changing the terms later without telling users or asking for fresh agreement, and a mismatch between what the document says and what the product actually does.

If a user clicked "I agree", does that protect every clause in my terms?

No. Consumer protection law sits above the acceptance mechanism, so a term that is unfair to a consumer can be struck out even when the acceptance was flawless. Burying something significant in dense text makes it more likely to be treated as unfair, not less.

Can I just copy another company's terms of service?

It is a bad idea. Borrowed terms rarely match your actual product, your data practices or your jurisdiction, and that mismatch is one of the things that weakens an agreement when it is tested.

Do I need a lawyer, or can a design team handle this?

The document itself needs a solicitor who understands your jurisdiction and your business model. A design consultancy can help with the screen the terms live on, which is the part most teams end up assigning to nobody.

Why does the terms screen matter commercially if nobody reads it?

It sits at the exact moment someone decides whether to carry on or back out. People are not reading clause 14, but they are reading the tone, the length and the size of the ask, and forming a view about whether your company is being straight with them.