A/B Test: Why You Need to Include Photos in Your Tweets

A/B Test: Why You Need to Include Photos in Your Tweets Jade Furubayashi Blogger Extraordinaire Simply Measured

As community managers, we are constantly on the hunt for small tweaks that go a long way. But on Twitter, when the goal is driving engagement, clicks, etc., only having 140-characters to work with can be intimidating.

How can you craft copy that’s eye-catching and creative enough to inspire your audience to click, favorite, or respond? This is certainly possible with clever copy, but nothing helps more than a killer photo.

Over the past few months, just like you, I’ve been running into headlines like this:

Screen Shot 2014-05-07 at 2.23.25 PM

And just like you, I’d say, “Yes, yes we know…But the quality of the content factors into this as well, right?”

I started paying closer attention. I’d include an image with a link to our latest blog post and the RTs/Favorites would start pouring in. So I figured I’d do what we do here at Simply Measured: Test and report my findings.

Tweet Without a Photo:

Tweet With a Photo:

My Tweets would have the same copy, sent at the same time on the same day of the week, but with a link instead of an image. And, you guessed it, the image Tweets would outperform the text-only Tweets every single time.

Here’s the process I followed for this A/B test. I strongly encourage any community manager to run these simple tests on the different components of your content. To be honest, we have a great design team that not every social media team has access to. You need to understand your own unique best practices for driving engagement, and that’s a good thing: You have anability as a marketer to measure every single thing you do, and improve based on the results.

Methodology

To get the best snapshot of how photos were performing in contrast to links, I had to control for a few factors:

  • Tweet copy: Tweaks in copy can make a huge difference, especially if you change your CTAs.
  • Tweet content: Obviously, the content you’re Tweeting has a major impact on your audience’s interest level.
  • Day of the week: Certain days are more engaging than others.
  • Time of Day: Firing off a Tweet at 7:30pm might not be as effective as Tweeting at 9:30am.

I picked a selection of 12 link-only Tweets from Friday, May 2nd, and used them on Friday, May 9th. However, on May 9th, these 12 Tweets included a link AND a photo.

Findings

After diving into my Twitter Account Reports for the respective days, I eliminated any Tweets that weren’t directly part of the experiment. As you can see below, I posted the same Tweets at the same time, on the same day of the week, but with different media types.

Screen Shot 2014-05-10 at 1.47.40 PM

The difference in engagement was significant.

In our A/B test, photo Tweets with the same copy, same link, at the same time of day, on the same day of the week lifted engagement by 33%.

Granted, this is only one A/B test, and I always try to exercise caution around making these assumptions. Let’s take a look at how photos affect engagement on a bigger scale:

Overall, Photos Average 13.2 Engagements Per Tweet vs. 5.9 for Links

At the end of the month, I always analyze our Twitter performance. I ask myself questions like:

  • “What was our top Tweet?”
  • “What time of day was the best for engagement?”
  • “What drove the peaks and dips?”
  • “What was our most engaging content type?”

In April, photos outperformed links and videos with ease.
Screen Shot 2014-05-07 at 2.57.19 PM
Although I Tweeted twice as many links as I did photos, photos drove nearly same total engagement. Engagement per Tweet was much higher for photos: 13.2 vs. 5.9… a 110% increase.

Twitter is becoming an increasingly more visual platform. The redesign, which everyone billed as a Facebook replica, is clearly setting the stage for a more visual experience. Right now, many brands are lagging behind. Photos show up every now and then, but the majority of Twitter is still text based. This is a great time to score a ton of engagements, before more brands catch on.

Looking to analyze your own Twitter account? Simply Measured can help. Click the link below to start a FREE trial of our reporting software, covering every major social network, and helping tie your social media engagement to direct web traffic.

Request a trial

 

Jade Furubayashi

My name is Jade and I'm the Social Media Manager for Simply Measured. We can find common ground in Beyoncé and Chipotle burrito bowls.