The optimization that raised conversions and cut revenue

The optimization that raised conversions and cut revenue

A store ran a promo banner that lifted conversion rate from 2.8% to 3.6%. Everyone celebrated. Revenue per visitor told a darker story.

Conversion rate counts buyers, not basket size. The clue: the promo pushed a cheap loss-leader, so more people bought, but each bought less.

The change in thinking, not the page: we started judging the test on revenue per visitor instead of conversion rate alone, then re-targeted the banner to a higher-margin bundle.

Under the better lens, the original promo's revenue per visitor had actually fallen from $4.10 to $3.70 despite the higher conversion rate. The re-targeted version brought it to $4.90.

Conversion rate optimizes for yes. Revenue per visitor optimizes for the bank. Pick the wrong North Star and you can win every test while the business quietly loses. Always pair the rate with the value behind it.