How the position of an acquisition offer in an article page impacts conversion
Care Net
Experiment Summary
Timeframe: 05/11/2018 - 05/25/2018
Care Net learned through previous testing the kind of content that works best for an acquisition offer in the article pages of their blog. They learned that they can turn their blog traffic into names and potentially donors for their organization by offering people the opportunity to sign the Pro-Life 101 Pledge and start their online course. After learning this, they hypothesized that the placement of the offer in the article might impact conversion. To test this, they developed a treatment that placed the acquisition offer in the middle of the article instead of at the end of it.
Research Question
How would the placement of the offer in the article impact conversion?
Design
Results
| Treatment Name | Conv. Rate | Relative Difference | Confidence | |
|---|---|---|---|---|
| C: | Control | 17.4% | ||
| T1: | Mid-Article | 24.2% | 39.3% | 100.0% |
This experiment has a required sample size of 276 in order to be valid. Since the experiment had a total sample size of 2,727, 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
× 39.3% increase in conversion rate
× 0% increase in average gift
Key Learnings
By placing the offer in the middle of the articles instead of at the end of them, they increased name conversion by 39%! This was a bit surprising as we had seen with tests similar to this the opposite result. What we can take away from this experiment are a couple of things:
1. The placement of the offer mid-article was disruptive enough to the reader that it caught their attention and appealed to their beliefs.
2. The pledge offer has a low cost to the reader and doesn’t require a big commitment from them.
3. The pledge at the end of the article on the control may have gotten too “lost” in the busyness of the other items in the footer of the articles.
4. By offering it higher on the page we were possibly able to catch a higher volume of people who were reading the article at that point than the number of people who actually finished the article.
5. This leads us to examine how much of the articles people are reading and propose a new round of experiments that will help keep people engaged and on the page for a longer period of time.