Conversion window

The window shapes how a business reads its own funnel. Set it too short and genuine conversions that took longer are missed, so a healthy funnel looks weak. Set it too long and a conversion is credited to an action that had little to do with the decision, and old groups keep improving while new groups are compared with a different standard. Either way, comparisons between periods stop meaning anything.

• Look at the actual distribution of time to convert and set the window where most genuine conversions land.
• Keep the window the same in every report. Changing it breaks comparison with earlier periods.
• Match it to the sales cycle: days for a cheap self-serve product, months for a B2B contract.
• Revisit it only when buying behaviour changes, not whenever a number disappoints.
• Read results by cohort, so each group is judged on the same amount of elapsed time.

A conversion window sets how long after a starting action a conversion still counts. In a product funnel, the starting action might be a sign-up and the conversion an upgrade to a paid plan. A thirty-day window counts only upgrades within thirty days of sign-up. On an ad platform, the starting action is an ad click or view, and the window is how long a later purchase or lead is credited to that ad. Many platforms let the window be changed, and click windows often default to around thirty days.

It is a close cousin of the attribution window. The attribution window decides which earlier touch gets credit. The conversion window decides whether a result is counted at all.

• Judging a group before its window has closed, then calling it a failure.
• Using the ad platform's default window without checking it fits the sales cycle.