Here’s What We Learned From Our Own (Unofficial) A/B Test

 In Data & Analytics, Design, Digital Marketing

We recently posted a blog post about colors and conversions essentially, how to use color to appeal to your audience and get results. During our research for this blog post, we found that the thing that matters most in terms of color and marketing is how much your color choices reflect your brand’s identity. So when an opportunity presented itself, we wanted to see if that was really true for us, and so we tried an (unofficial) A/B test to find out.

The plan was simple: create two blog graphics for our “5 Reasons Ads Just Aren’t Enough Anymore” blog post, identical except for the colors, and promote both to multiple audiences. We created one image that strictly corresponds to our branding, and one that’s a little more colorful. You may remember seeing these around on social media:


We tried these on four ad sets: two on Facebook, and two corresponding sets on Instagram. We wanted to know which would be more effective: the brighter, more colorful graphic, or the image that more closely follows our brand identity, but is a bit more subdued. Everything else – post text, image, audiences, time promoted remained the same.

Of course, this wasn’t an official, completely fair test. Facebook’s algorithms interfere almost immediately with any ad or promoted post, so it’s possible that early response to the ads potentially dictated how Facebook distributed the rest of them.

That said, our results were still very interesting. When our ads were done running, here’s what we found:

Based on these numbers, it would appear that the more colorful images work best. Facebook Ads served more “color” ads to three out of four of our audiences, which means a lot more of those were seen than our “on-brand” ads.

Post Engagement which refers to likes, comments, shares, etc. on our ads matches impressions for the most part. Interestingly enough, it’s clear here that Facebook users interacted more with the more colorful ads, while Instagram audiences were more engaged with the on-brand graphics. However, when comparing with the number of impressions each audience was served, these results aren’t surprising.

When we looked at how many users actually clicked through to our blog post, however, the numbers got a lot more interesting:

So what does all this mean?

Well, it’s difficult to say for sure. As we acknowledged before, this isn’t a perfect test, and our results are certainly fallible. However, it would seem that though Facebook deemed our colorful graphics more engaging, viewers were much more likely to actually click (or tap) through on the image that more closely reflects our branding. In fact, 72% of our clicks came from the “on-brand” ads. And while impressions and post engagements are worth tracking, click-throughs — measuring actual visits to our website — matter much more.

It seems the research we found was correct — while Facebook may have judged the more colorful ads as more engaging, the users who did click through to our blog post preferred the one that looked most like us.


 Invite Change Media Group to your inbox! You’ll get our start-up series of emails, then one email a month of must-know info. No spam ever, unsubscribe any time.




As CMG’s Crafter of Strategic Communication, Amy Libka combines strategy and creativity to help clients communicate as effectively as possible. She is passionate about clear, effective communication, and enjoys crafting messages that help individuals and groups achieve their goals. She specializes in creating content and managing digital advertising campaigns for a diverse group of clients.

Recommended Posts
Contact Us

Want to get in touch? Shoot us a message and we'll get back to you, asap!

Not readable? Change text. captcha txt