Conversion UX

Every element on a screen asks the user a question. The question determines whether they act or hesitate. A screen that "looks fine" usually fails because it asks a hard question (Is this worth $19/month? What will this cost in total? What am I locked into?) where an easy one was available, or because it leaves a real objection unanswered and lets doubt do the deciding.

Remove friction before you reach for persuasion

Work in this order. It is deliberately the reverse of how conversion advice is usually written, and the ordering is the point.

1. Name the decision. Write one sentence: what is the user deciding on this screen, and what would make them say no? Everything below serves that sentence.
2. Clear the documented blockers. The largest measured losses are friction, not weak persuasion — undisclosed costs (40% of cart abandonment), slow or unstated delivery (20%), distrust of the payment form (19%), forced account creation (18%), an over-long form (17%). Fix these first; they have real numbers behind them and the persuasion levers mostly do not.
3. Answer the objections on the screen. Return policy, total price, delivery date, cancellation terms, what happens after the trial. An answered objection beats a persuasive headline.
4. Then choose behavioral levers from references/principles.md, each with its boundary condition.
5. Check the bright lines in references/bright-lines.md. Several standard "conversion tricks" are unlawful in the EU, UK, or specific US states. This check is not optional and not a matter of taste.
6. State what you could not know. Name every choice that is a hypothesis rather than a rule, and say what would settle it. See references/evidence.md for when a test is even possible.

Non-negotiables

These hold regardless of what the brief asks for.