Jerusalem Prayer Team

How increasing the lowest gift amount affects donation revenue

Experiment ID: #19241

Jerusalem Prayer Team

Experiment Summary

Timeframe: 06/07/2019 - 06/22/2019

Jerusalem Prayer Team had run a microdonation campaign to raise money to feed Holocaust survivors in Israel. A radical redesign had generated a 79% lift in revenue by, among other things, adding a $45 button in the gift array between $3 and $90. They were curious how revenue might be impacted if they increased the lowest gift amount to bring it more in line with the other two amounts. They launched a version of the page where the only difference was a $9 ask on the gift array and launched a test to determine the impact.

Research Question

How will an increase to the lowest gift amount in an array affect donation revenue?


C: Radical Redesign
T1: Radical Redesign ($9)


  Treatment Name Revenue per Visitor Relative Difference Confidence Average Gift
C: Radical Redesign $4.14 $22.66
T1: Radical Redesign ($9) $6.64 60.6% 96.0% $30.10

This experiment was validated using 3rd party testing tools. Based upon those calculations, a significant level of confidence was met so these experiment results are valid.

Flux Metrics Affected

The Flux Metrics analyze the three primary metrics that affect revenue (traffic, conversion rate, and average gift). This experiment produced the following results:

    0% increase in traffic
× 20.9% increase in conversion rate
× 32.8% increase in average gift
60.6% increase in revenue

Key Learnings

Not only did the treatment (with the $9 array) produce a slight, non-valid lift in donations, it also increased revenue per visitor by 60.6% due to a compounding increase in average gift. It’s an important reminder that when conversion AND average gift increases, the increase to revenue is not additive, it’s compound. This small test produced a big lift in revenue, but there’s much more to be tested on this page—including the value proposition.

Experiment Documented by Jeff Giddens
Jeff Giddens is President of NextAfter.

Question about experiment #19241

If you have any questions about this experiment or would like additional details not discussed above, please feel free to contact them directly.