FamilyLife

How length of ad copy in a Facebook ad affects conversion rate

Experiment ID: #2691

FamilyLife

FamilyLife® has been committed to helping individuals find biblical help for their marriage and family relationships. Through the Weekend to Remember® marriage getaways, FamilyLife Today® radio broadcasts, The Art of Marriage® video event, and the many other resources and content, God has used FamilyLife to restore hope for millions of couples and transform their lives.

Experiment Summary

Timeframe: 08/11/2017 - 08/18/2017

First, having a “read more” link doesn’t guarantee more clicks or conversions.

Second, in creating a connection with the reader through describing the problem (i.e. the reason for the offer), we have to discover the minimum effective dose. Anything more than that is unnecessary in the mind of the reader. Unfortunately, there are few methods outside of testing that teach us what the minimum effective dose of copy is needed to build the problem.

Research Question

Does shorter ad copy increase conversion rate?

Design

C: Long Copy
T1: Shorter Copy

Results

 Treatment NameConv. RateRelative DifferenceConfidence
C: Long Copy 0.42%
T1: Shorter Copy 0.54%29.8% 100.0%

This experiment has a required sample size of 24,658 in order to be valid. Since the experiment had a total sample size of 176,123, 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
× 29.8% increase in conversion rate
× 0% increase in average gift

Key Learnings

FamilyLife was promoting a name acquisition offer surrounding a free ebook on Facebook. They wanted to see if longer, value proposition-centric body copy or a shorter, to-the-point ad copy would increase conversion rate. All other elements of the ad remained the same.


Experiment Documented by Allison Autrey

Question about experiment #2691

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