Because saying no costs the business something, and somebody decided friction was worth it. The cancel link buried four screens deep, the pre-ticked box, the decline button in grey while accept glows blue. None of it is an accident. All of it is a design decision that somebody signed off.
These are dark patterns, and they are far more common than most business owners realise, including on sites their own team built. Usually nobody set out to be deceptive. A growth target arrived, an A/B test showed the pushy version won, and the pushy version shipped.
Our position on this is firm and we will defend it. Dark patterns borrow conversions from the future and pay them back with interest in refunds, chargebacks, support load, and lost trust. They are also increasingly illegal. Here is what they are, how common they have become, and what we build instead.
A dark pattern is an interface designed to steer people into choices they would not otherwise make. The Federal Trade Commission describes them as sophisticated design practices that can trick or manipulate consumers into buying products or services or giving up their privacy. The key word is manipulate, not persuade.
The distinction sits in whose interest the design serves. A clear, confident sales page that explains why your product is worth the money is persuasion. A checkout that adds an item you did not choose is manipulation. One respects the decision, the other works around it.
The term now carries legal weight rather than just design criticism. Regulators in the United States and Europe both use it, and both have moved from writing reports to writing rules. That shift is what turns this from an ethics conversation into a business risk conversation.
The FTC's staff report Bringing Dark Patterns to Light, published on 15 September 2022, groups them into four families: misleading consumers and disguising advertisements, making subscriptions difficult to cancel, hiding terms and extra fees, and coercive data collection. Almost every real example fits one of those four.
The disguised advertising family covers ads dressed as editorial content, comparison sites that present themselves as neutral while being paid, and false countdown timers that manufacture urgency for a deal that never expires. The timer one is worth singling out, because it is so common that people have stopped noticing it, and it is straightforwardly a lie about a fact.
The hidden costs family is the one that damages trust fastest. The FTC report describes burying limitations in lengthy terms of service and concealing mandatory charges until late in the transaction. Anyone who has reached the final checkout screen and watched the price jump knows exactly how that feels, and knows that the feeling attaches to the brand, not to the fee.
Much more common than most teams assume. A review by the FTC alongside the International Consumer Protection and Enforcement Network and the Global Privacy Enforcement Network examined 642 websites and mobile apps offering subscription services, across 27 authorities from 26 countries.
The result was that nearly 76% of the sites and apps examined employed at least one possible dark pattern, and nearly 67% used multiple possible dark patterns. That is not a fringe practice by bad actors. That is the majority of subscription businesses, most of which would describe themselves as reputable.
The review identified sneaking and interface interference as the most common techniques, meaning concealing important information and obscuring details or presetting options to guide decisions toward business interests. Presetting options is the one we see most in client work, and it is almost always inherited rather than chosen, sitting in a form nobody has looked at in two years.
In the European Union, several are explicitly prohibited. Article 25 of the Digital Services Act states that providers of online platforms shall not design, organise or operate their online interfaces in a way that deceives or manipulates the recipients of their service, or in a way that otherwise materially distorts or impairs their ability to make free and informed decisions.
The regulation goes further and names practices by name. Article 25 says the Commission may issue guidelines on how the prohibition applies to giving more prominence to certain choices when asking for a decision, to repeatedly requesting a choice that has already been made especially by presenting pop-ups that interfere with the user experience, and to making the procedure for terminating a service more difficult than subscribing to it.
Read that last one again, because it describes something almost every subscription business does. If a customer can sign up in two clicks and must phone you to cancel, that asymmetry is exactly what the text describes. Article 25 also notes the prohibition does not apply to practices already covered by the unfair commercial practices directive or by the General Data Protection Regulation, which means these are overlapping regimes rather than a single loophole.
In the short term, often yes, and that is precisely the problem. A pre-ticked box does increase opt-in rates. A hidden cancel flow does reduce cancellations this month. The test that approved it was measuring the right number over the wrong time period.
What the test does not capture is what happens next. The person who was tricked into a subscription cancels angrily, disputes the charge, leaves a review, and never buys from you again. The signup counted. The refund, the support ticket, the chargeback fee, and the review did not appear in the same dashboard.
Data collected under coercion is worth less too. An email address given by someone who did not realise they were giving it produces poor open rates, high unsubscribes, and spam complaints that damage your sending reputation. You end up with a bigger list that performs worse, which is an expensive way to look successful.
Persuasion gives people accurate information and makes the good choice easy. Manipulation withholds information or makes the alternative artificially hard. The test we use is simple: if you explained the design decision to the user out loud, would they be fine with it?
Run that test on a few common patterns. Putting your best plan first because most customers pick it is fine, and you could say so. Making the cancel button the same colour as the background is not, and you could not say so. The ones that fail the test are the ones you would have to hide.
Urgency is the interesting edge case. Real scarcity is honest and useful, because a genuine deadline helps someone decide. A countdown that resets when you reload is a lie told by a widget. The difference is not the technique, it is whether the underlying fact is true. This is where careful microcopy earns its place, because honest urgency needs precise words rather than pressure.
Try to leave. Go through your own cancellation flow, your own unsubscribe, your own cookie banner, and your own checkout as a customer rather than as an owner. Count the clicks and note anywhere you feel a small resistance. That resistance is the thing.
Cookie banners deserve their own pass because they are where most sites fail without realising it. If accepting is one button and rejecting takes two clicks through a settings panel, that is giving more prominence to one choice, which is exactly the practice Article 25 names. We covered the practical setup in our guide to cookie consent.
Then look at your forms and your pricing. Any box that is ticked by default should be justified out loud. Any fee that appears after the first price is shown should be moved forward. Our piece on pricing page design makes the case that showing the real number early converts better anyway, which is the useful part of this whole argument.
Clarity, symmetry, and reversibility. Say what the thing costs, make leaving as easy as joining, and let people undo a decision without a phone call. Those three rules eliminate almost every dark pattern without anyone needing to memorise a list of them.
Symmetry is the one that gets resisted internally, and it is the most important. If signing up takes two clicks, cancelling should take two clicks. Teams push back because they fear churn, but a customer who can leave easily is a customer who will come back, and one who feels trapped will make sure other people hear about it.
The commercial argument holds up in our own work. Across the projects we have delivered we see an average conversion lift of 3.5 times, and those gains come from clarity rather than pressure. On Ajust we saw sign-ups rise 230%, and on Sport Design Australia enquiry conversions rose 205%. None of that came from hiding a button.
Pick one flow this week and walk it as a customer. Your cancellation flow is the highest value place to start, because it is the one your team never uses and the one regulators are most focused on. Fix what makes you wince, and do it before someone else finds it.
Then set a standard your team can apply without asking permission each time. Ours is that any design decision we would not explain to the user out loud does not ship. It is a low bar and it catches nearly everything, and it removes the awkward conversation about whether a specific tactic is over the line.
If you want a second pair of eyes on a checkout, a signup, or a cancellation flow, we are happy to walk through it with you and tell you honestly what we would change. You can find us at phoenix.studio, and we usually get back to people within 48 hours.
Tell us where you want to go. We'll tell you how we'd get you there.