Every dark pattern named, why teams keep shipping them, and a five-minute audit of your own checkout.
CTFM Team
A dark pattern is an interface designed to get you to do something you did not intend to do. Not a confusing design. A deliberate one.
The term was coined by Harry Brignull, a user experience specialist who began cataloguing these techniques and now documents them publicly. The name stuck because it described something everyone had experienced and nobody had labelled.
They work. That is the uncomfortable part. Here is each one, why it converts, and what it costs.
Why a marketing site is writing about this
Because most people using these do not know they are. They inherited a checkout flow, or copied a competitor, or accepted a test result that increased conversion without asking why. Naming them is how you notice them in your own product.
Making the decline option insulting. "No thanks, I prefer paying full price." "I don't care about growing my business."
It works through mild social discomfort, even though nobody is watching you click, and it also produces a very particular kind of irritation that people remember and screenshot.
The honest version: make the decline neutral. "No thanks" is a complete sentence.
Fees that appear only at the final step. Shipping, service charges, booking fees, taxes presented as a surprise.
It lifts completion at the top of the checkout. It raises abandonment at the bottom. And it produces a very particular kind of angry review. In several markets, requirements to show total price up front are now regulated, so check your jurisdiction.
The honest version: show the total early. You lose a few people sooner, and you keep a lot more of the ones who carry on to the end.
A free trial that requires a card and converts silently to a paid subscription with no warning.
The revenue looks great. A good chunk of it is people who forgot. That money comes back as chargebacks, refund requests and one-star reviews with the word "scam" in them.
The honest version: email before charging. Companies that do this report lower revenue and much lower dispute rates, and they keep the customers who actually wanted the product.
Easy to get in, deliberately hard to get out. Cancellation buried several levels deep, or only available by phone during limited hours, or requiring you to speak to someone.
This is an active enforcement area. Regulators in several markets have gone after companies whose cancel flow is far harder than their signup flow.
The rule that is emerging across jurisdictions
The broad principle regulators keep applying is symmetry: cancelling should be about as easy as subscribing. If signup is two clicks online and cancellation is a phone call, you are on the wrong side of that principle. Check the specific rules for the markets you operate in.
The honest version: let people cancel where they signed up. Offer a pause option, which retains more people than an obstacle course does.
A timer that resets on reload, or restarts for each visitor, on an offer with no real end date.
Works once. Then someone reloads the page.
The honest version: real deadlines. A cohort that actually starts, a price that actually rises, stock that actually runs out. These convert better because the urgency survives inspection.
Add-ons, insurance, warranties or newsletter signups already ticked, relying on people not reading.
The honest version: unticked, with one clear line on why someone might want it. You sell fewer add-ons to people who did not want them, which was never revenue you could keep.
Content dressed up as a menu. Download buttons that are really ads. Sponsored posts styled to look exactly like the real articles.
The honest version: label it. Disclosure requirements exist in most markets, and beyond compliance, tricking someone into a click produces a visitor who is already annoyed.
Advertising one thing and delivering another. The advertised price applies to a configuration nobody wants, or the advertised item is permanently out of stock.
The honest version: advertise what you will actually sell.
This is the genuinely interesting part, and it explains why smart teams ship these.
Dark patterns almost always win the test you are running. They lose the test you are not running.
What you measure
What you miss
Checkout completion rate
Refund rate, chargeback rate, dispute costs
Trial to paid conversion
Cancellation on first charge, support volume
Add-on attach rate
Reviews mentioning being tricked
Newsletter signups
Spam complaints, deliverability damage
This quarter's revenue
Repeat purchase rate, referral rate
One change that fixes most of this
Add refund rate, dispute rate and support ticket volume to the same dashboard as your conversion metrics. Not in a separate report owned by a different team. On the same screen. Most dark patterns stop looking clever the moment their costs sit next to their benefits.
