Ad Copy A/B Tests You’d Never Think to Try
You hear it all the time in the digital advertising and marketing industry: Always be testing.
It applies to everything from keywords to landing pages to creative assets and more, but for the purpose of this post we’ll focus on ad copy.
Do a quick Google search for ad copy testing and you’ll find information about all sorts of tests you can try. Capitalizing certain words and letters, adding exclamation points, and swapping words for synonyms are just a few ad copy variations that are popular for testing.
But let’s say you’ve already tried everything you can think of. What happens when you run out of changes to test?
…Trick question. You’re never out of changes to test! Ad copy can always be reworked to improve performance, even when you don’t expect to see a change. The only way to keep your campaign on the upswing is to test, test, test.
If you think outside the box—which can be easier said than done—you’ll discover all sorts of ad copy variations worth testing. Here are a few to start you off:
1. Spell out numbers, symbols, and abbreviations (or don’t).
If you’re offering a free month of service, for example, you already have several options to test: Get 1 month free vs. Get one month free, or even Get 1 mo. free. You can also test & vs. and, % vs. percent, # vs. number, etc. (or should we say, et cetera?).
2. Experiment with tense.
Feel stressed? could have a different success rate than Feeling stressed?
3. Test complete sentences against words and phrases.
Do you feel stressed? or Are you feeling stressed? can be tested against Stressed?, Feel stressed? etc.
4. Try active and passive statements.
Grammar nitpickers will tell you never to use the passive voice (i.e., Our product is used by all kinds of people. rather than All kinds of people use our product.), but marketers are allowed to break the rules when certain phrasing resonates better than what’s grammatically correct.
5. Throw something crazy in the mix.
Copy like, Better than mediocre service. or Find out if we suck. is snarky for sure, but it also uses elements of reverse psychology by emphasizing the opposite of the message you’re delivering. If you have a little leeway to be edgy, go for it!
If you’re at the point where you’re testing stuff nobody thinks to test, you probably already know a thing or two about what you’re doing. Or maybe your tests haven’t gone so well in the past and you’re looking for something new to try. Regardless, it’s absolutely crucial to analyze your test results accurately. You’ll need enough data to reach a statistically significant conclusion, eliminate any outside influencers (seasonality, other tests running concurrently, etc.), and avoid anything else that can muddy the results.
Once you determine which ad copy performed better, get rid of the loser and tweak the winning ad copy for another round of testing, followed by another round of testing, followed by another round of testing…