Why a trial that requires a card converts differently

Two trials of the same software. One needs an email; the other needs sixteen digits. The card version gets a third of the sign-ups and more paying customers at the end of the month, which reads as a paradox until you notice the two trials are not the same offer.

What everyone sees

Removing the card field is treated as pure gain, since it lowers a barrier and barriers reduce conversion. The barrier is doing two jobs. It selects for intent and it converts the trial from a free sample into a purchase already made, with cancellation as the action requiring effort.

What is actually happening

Shampanier, Mazar and Ariely found that moving a price from a small positive amount to zero produced a disproportionate jump in take-up, with participants treating free goods as a different category rather than a cheaper one — which is why a no-card trial fills with people who would not have paid anything. Gourville and Soman documented the other half through payment depreciation: separating payment from consumption in time weakens the sense of having paid, so a card entered now and charged later sits in an intermediate state that still frames later use as consumption of something bought. The two effects run in opposite directions on volume and the same direction on quality of cohort.

Why it stays hidden

The trade-off hides because the two funnels are usually judged on different metrics by different teams. Growth reports sign-ups, which favour the open trial. Revenue reports paid conversion, which favours the card. Neither number is wrong and neither answers the question of which offer was actually made.