Human Coalition

How a radical landing page redesign affects name acquisition rate

Experiment ID: #864

Human Coalition

Experiment Summary

Timeframe: 01/12/2015 - 02/02/2015

Online for Life developed a petition, the Pro-Life Declaration to acquire new email addresses for their house file. Initially, the petition was a flat, content-focused design that looked exactly like…a petition. We wanted to test this against a radical redesign centered around a compelling image combined with the value statement “Life is Precious”. Rather than assume that the site visitor was ready to sign the petition, we assumed that we had to keep selling them once they got on the page. In addition to the new value statement and image, we added a counter at the top to show that by signing the petition, the visitor was joining a much larger group.

Research Question

Does a more evocative landing page with social proof outperform a more straightforward, functional landing page?

Design

C: Standard Landing Page
T1: Radical Redesign

Results

 Treatment NameConv. RateRelative DifferenceConfidence
C: Standard Landing Page 34.3%
T1: Radical Redesign 46.1%34.7% 100.0%

This experiment has a required sample size of 123 in order to be valid. Since the experiment had a total sample size of 71,688, 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
× 34.7% increase in conversion rate
× 0% increase in average gift

Key Learnings

The radical redesign delivered a 34.7% lift in the number of people who signed the Declaration — a huge boost to the campaign. This lift proved that the landing page must appeal to the donor. It also reminded us that visuals help sell the ultimate conversion, and we are not done selling once we have the user on the page.

 


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

Question about experiment #864

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