In many companies, your copy will be put to the test. But wait—don’t panic. What I mean, is that your copy will be put to the test against more of your copy. It’s a pretty common (and very important) process that you need to know about. Read on…
Today’s question is from Ansel R. who asks, “My client wants to do an A/B test with subject lines. I smiled and nodded…What does that mean?”
I know, you thought you left tests behind in school. And you did—these are a different kind of test. And, believe it or not, they’re great.
As you and I both know, writing copy is not an exact science. When you put your copy together, you make sure that you make the benefit to consumer clear, you have a strong CTA, you write in the brand voice and all of the myriad of other techniques that you’ve learned and practice.
But even with all of those boxes to check, it’s still possible—likely, even—that you can come up with a couple of equally viable sets of copy. And the problem is that the only people who can actually know which set of copy will be more effective with the readers…are the readers themselves.
So what do you do? You test them against each other!
Here’s an example of how that goes down. Say you’re putting together an email and you have two different headlines. Well, you (or, more likely, your client or boss) send out one set of that email with headline A to half of the subscriber list and another set of that exact same email with headline B to the other half of the list. Then, after it’s sent, you check the stats to see which version had the better click-through rate. Since the only thing that was different in this email was the headline, you know that the headline with the higher click-through rate was the winner! Isn’t that cool?
In this scenario too, if the client/company is technologically savvy enough and has a big enough subscription list, they could send out the email tests to only 10% of the list. When they find out the winner from that test, they could send the winner out to the other 90%! (Seriously, I love testing.)
A test like this, in which only one variable is changed, is called an A/B test (or a single variant test). If, in this email, you had changed the headline of the emails, but then also a picture in the emails and maybe even the subject line, you’ve changed multiple elements in the email and testing these two versions against each other is called a multivariate test. In multivariate tests, it can be hard to determine exactly what change made the difference (or what combination of changes).
I know that some copywriters worry that testing diminishes their own purpose or authority. But it’s not like your boss or client will be testing your copy against copy generated by a computer or something; they’re testing your copy against your own copy.
And sometimes you can guess the results of a test, but sometimes they come completely out of left field and there’s no way you could have predicted them. That testing (and, sometimes, re-testing and re-testing) just helps you learn more about your audience and how you can best communicate to them to make them take action. And that’s a good thing, isn’t it?
It’s a great thing. Testing isn’t just a tool for the marketing or product teams, it’s very much a tool for you, too. Checking out the results and learning from them can only serve to help you write more effective copy for your client/company and make you a better copywriter overall.
Your turn! How could you apply testing to some of your projects? Tell us in the comments below!