How a video trailer on a facebook ad affects click rates
Focus on the Family
Focus on the Family is a global Christian ministry dedicated to helping families thrive. We provide help and resources for couples to build healthy marriages that reflect God's design, and for parents to raise their children according to morals and values grounded in biblical principles.
Experiment Summary
Timeframe: 04/09/2021 - 06/10/2021
Focus on the Family created some trailers for their video series. This offered the opportunity for a great test to see whether a more engaging form of creative on a Facebook ad will increase engagement. We tested the trailers against their typical screenshot of the video series against the trailer as the medium for the ads. This was a marketplace test, not a strict A/B test on Facebook. We ran both creatives in the Facebook campaign to which one the market prefers and engages with more readily.
Research Question
A video trailer will increase click rates and engagements.
Design
Results
| Treatment Name | Click Rate | Relative Difference | Confidence | |
|---|---|---|---|---|
| C: | Control | 1.1% | ||
| T1: | Treatment #1 | 2.4% | 124.1% | 100.0% |
This experiment has a required sample size of 977 in order to be valid. Since the experiment had a total sample size of 1,237,115, 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:
124.1% increase in traffic
× 0% increase in conversion rate
× 0% increase in average gift
Key Learnings
The video treatment resulted in much higher engagement overall. It increased the click through rate by 124% and the comments by 100%. Facebook used this creative much more often because of the higher results. The video has a 34% lower cost per click, taking the cost from .67 to .44. There was an increase in donation as well because of the increase in traffic, but the proportions of the samples was so vast that the test was invalid.