How a new serial ask decreased donor conversion rate
Heritage Action for America
Experiment Summary
Ended On: 03/26/2019
In reviewing the Serial Ask results from a recent experiment where the social proof elements increased gift matching and revenue from serial ask donors from The Heritage Foundation, the team at Heritage Action wanted to test bringing the popup language which had previously increased donor conversion rates into the primary page so that visitors not on a desktop device could also see it.
Research Question
Will bringing the popup language used to previously increase desktop serial ask donor conversion rates work in the primary messaging for visitors on all device types?
Design
Results
| Treatment Name | Conv. Rate | Relative Difference | Confidence | |
|---|---|---|---|---|
| C: | Control | 1.7% | ||
| T1: | Treatment #1 | 0.87% | -50.3% | 99.4% |
This experiment has a required sample size of 1,302 in order to be valid. Since the experiment had a total sample size of 5,008, and the level of confidence is above 95% the 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
× 50.3% decrease in conversion rate
× 0% increase in average gift
Key Learnings
With a 99.4 level of confidence, we saw a decrease -50.3% in donor conversion rate.
This treatment also decreased revenue/visitor by -76%.
Our hypothesis is that the serial ask is successful when emotional decisions are made. The treatment language leans heavily upon social proof, which is more of a logical argument. The additional clarity to position the difference between the two organizations may have also played a role in the donor conversion rate and revenue decreases, as well.