As marketers, we’re acutely aware that our words carry weight. That they matter. We know this because more often than not, the success of our campaigns ultimately hinges on whether we choose the right words.
Right before clicking ‘post’ ‘send’ or ‘publish,’ it’s always: Will this social post get users to click? Will this subject line entice people to open? Will the text in this ad nudge shoppers to convert?
I’ve often examined how big, dramatic differences in post text can lead to major upticks in performance. This approach is especially important in A/B testing so you can make clear-cut observations based on your results.
I tend to lean toward the dramatic (probably as a result of my long unutilized film degree), but this week, I’m going against my own proclivities to discuss just the opposite: I want to look at the effect small text tweaks have, and how they can drive meaningful improvements in performance. Let’s dive in.
If I’m going to make the case for the value of post text micro-optimizations, it’s important to first provide context. To do that, let’s quickly zoom out.
If you were here last week, you’ll remember that the results of a recent experiment that we ran supported the idea that post text has a significant impact on performance metrics like CPC and CTR. Here’s the TL;DR:
- We tested significantly different post texts and used Facebook’s paid channels to distribute the ads.
- We found that on average, different post texts resulted in a 15% difference in CPC and a 17% difference in CTR.
- To oversimplify: the results showed that differences in post texts definitely make a difference — they’re undeniably a factor in performance.
There are different ways to build off these findings. One way — the way that I’m talking about today — is to see whether slight variations in post text can improve results, and if they do, by how much?
3 Examples of Boosting CTR by 3%
To better understand the degree to which slight variations in post text can impact performance, we dug through our huge database to find some recent wins. Below are three examples where testing out slight variations led to optimized post texts that gave a boost in click-through-rates by about 3%.
The variation with the 3% higher CTR is in bold below.
- Post #1:
- Variation 1: “Try these healthy alternatives the next time a craving hits.”
- Variation 2: “Try one of these healthy alternatives the next time a craving hits”
- Post #1:
- Post #2:
- Variation 1: “Unique baby names have been in high demand!”
- Variation 2: “Unique baby names are on the rise!”
- Post #2:
- Post #3:
- Variation 1: “Kids bored in the house? These toys are just what they need!”
- Variation 2: “Kids bored in the house? These toys will do the trick!”
Striving for Scale
I know what you’re thinking. 3% difference? Why bother? I’ll tell you why – scale. 3% may not seem like a lot. But when you’re working with campaigns that reach 10s or 100s of thousands of people, it can have a big impact. In the examples above, the average difference between CTRs was upwards of 17%, which is a huge margin that could be the difference between hitting your budget and CPA goals or missing them completely.
Let’s use an oversimplified example to illustrate:
- If a post text micro-optimization increased the CTR from 4% to 7%
- For every 1,000 impressions, an increase in CTR by 3% would mean:
- 30 more people visiting your online store
- 30 more potential paying subscribers
- 30 more people in the top of your marketing funnel
When I write any sort of promotional text, I can often agonize over little differences for longer than it takes me to write the original. If you’re anything like me on that front, you may be beginning to dread the expected campaign and traffic explosion just because of the amount of time you’ll be agonizing over text. Can you humor me for a second for a shameless plug?
I’ve been working with Anyword’s AI Language Platform a lot lately, and I’ve been having quite a bit of fun using it as a tool to wordsmith. It’s like it does all of the agonizing for me, and all I have to do is make a choice. To demonstrate this, I put the content of this email into the platform and asked it to generate post text for me. Then I picked one of them to wordsmith:
The top text is what I gave the system to wordsmith, and the score our AI gave it in terms of performance potential. Immediately below it, you’ll see the tweaks the platform generated, and the score for those tweaks. Agony begone! Thank you for humoring me.
Whether you use an AI copywriting platform or a whiteboard and a marker, there’s a clear takeaway here: Wordsmithing matters. Just do it.