A pricing page teardown, in public

We asked three customers to let us rebuild their pricing page in public, with their real conversion data before and after. All three sold subscriptions between $9 and $99 a month.

The most consistent win was reducing tier count. Two of the three ran five plans; cutting to three lifted paid conversion by 18% and 24% respectively. The third already had three and gained nothing from the change, which is a useful control.

Highlighting the middle plan mattered when — and only when — the highlight matched where buyers already clustered. Highlighting a plan nobody wanted moved conversion down by 4%, because it made the recommended option look expensive relative to what people actually chose.

Finally, moving the usage limit out of a comparison table and up beside the price cut support tickets about "which plan do I need" by roughly half. That is not a conversion metric, but it is an hour a week back.