The Heritage Foundation

How using Facebook’s “Donate” optimization method decreased results

Experiment ID: #11944

The Heritage Foundation

Founded in 1973, The Heritage Foundation is a research and educational institution—a think tank—whose mission is to formulate and promote conservative public policies based on the principles of free enterprise, limited government, individual freedom, traditional American values, and a strong national defense.

Experiment Summary

Timeframe: 05/07/2019 - 05/16/2019

In an attempt to understand whether or not the “Donate” optimization method could produce better results than the “Purchase” optimization method within Facebook, we decided to experiment with these settings to understand how it would impact the results for The Heritage Foundation’s “Religious Freedom eBook” offer.

Research Question

Will the “Donate” optimization setting increase results for acquiring new emails and/or donors?

Design

C: Religious Freedom Booklet - Acquisition | Purchase
T1: Religious Freedom Booklet - Acquisition | Donate

Results

 Treatment NameConv. RateRelative DifferenceConfidence
C: Religious Freedom Booklet - Acquisition | Purchase 0.02%
T1: Religious Freedom Booklet - Acquisition | Donate 0.00%-90.6% 98.0%

This experiment has a required sample size of 19,593 in order to be valid. Since the experiment had a total sample size of 81,105, 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
× 90.6% decrease in conversion rate
× 0% increase in average gift

Key Learnings

With a 97.9% level of confidence, the “Donate” optimization setting decreased donor conversion rate by -90.6%.

Making matters worse, this setting also decreased email registration rates by -65%, as well (with a 100% level of confidence).

Further experimentation may be required with other offers, but when it comes to introducing the “Donate” optimization method against an already running “Purchase” optimization setting, it would appear that results negatively compound to significantly impact performance for the existing campaign.


Experiment Documented by Greg Colunga
Greg Colunga is Executive Vice President at NextAfter.

Question about experiment #11944

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