split test marketing

Split Testing: Which TLA (Three Letter Abbreviation) Are You?

For many small businesses, just executing a marketing program is an accomplishment in and of itself. Finding the time, allocating the budget, and understanding the results are all things that are not quite top priorities when there’s so much else to do. And still, if you want successful, cost-effective marketing, testing is the only way short of luck to make it happen.

You may have heard of split testing, A/B testing, or multi-variate testing. These are all basically the same thing – namely, you guessed it, testing. And before we move forward here, I want to define what testing is in the context of marketing your business.

Thinking specifically about digital marketing platforms, testing is often quite straightforward. Google Analytics (which, if you’re not already using, please do — it’s free!) has a website testing module built-in. It lets you automatically push a defined percentage of your web traffic to an alternate version of a web page so you can compare how they influence behavior and convert to leads/sales. That’s just one example.

Other examples:

  • Comparing different ads in your pay-per-click campaigns and how often they induce clicks
  • Sending a marketing email with different subject lines to two random groups and comparing how they yield opens and click-throughs
  • Changing the text in a “button” on your site (for example, using “Get Started” instead of “Pay Now”) and observing if it affects transaction rate.

What these tests have in common is that they compare one option to another so you can identify which is more conducive to business. With that simple definition, it’s easy to apply the idea of testing to offline strategies as well. Distribute two versions of a coupon to see which one is redeemed more often. Print a one-sheet and a brochure version of the same information and see which one is picked up more often. I’ll bet there are things like this that you’ve already tried without actively thinking about them being “tests.”

That said, it’s important to take from these tests whatever information you can. If they’re on a digital platform, there will almost certainly be some built-in analytics to observe to make it easy to determine what has worked best. If it’s offline marketing you’re testing, take some time before you roll out the test to identify what it is you want to measure. It doesn’t have to be complicated. To use the coupon example again, if you just want to measure redemption rates, keep track of how many of each were distributed versus how many were redeemed. You could go a step further by comparing the average total sale associated with the different versions. Maybe one version is somehow encouraging a larger total purchase.

You may know the abbreviation, ABT. It stands for “Always Be Testing.” That sounds a little aggressive, right? Let’s consider how realistic it is, and what your other alternatives are, using other “TLAs” I’m going to make up.

ABT – Always Be Testing

Any expert will tell you that this is the ideal scenario and they’re not wrong. The more you test, the more you learn. The more you learn, the better you are at channeling your marketing dollars toward tactics that work and away from those that don’t. It may seem impossible for a small business owner or a one-person marketing team to adopt a full-time philosophy like this, but it’s really not so difficult. It’s more about a change in approach than an increase in workload. Start by thinking about the marketing tasks you’re already doing and what you want them to accomplish. Then, consider how you can make an adjustment or alternate version. Don’t double your work, just do a small part differently. And finally, remember to observe the results!

SBT – Sometimes Be Testing

Okay, you’re not quite ready to make sure you’re always running some sort of test. You should still do what you can, when you can. The next time you send a marketing email, send half of your list one subject line and send the other half another subject line. Then compare which gets more opens. If that’s all the bandwidth you have in the short term, so be it. Remember that the more often you do this sort of testing, the more you’ll learn and be able to improve your bottom line, but don’t let an inability to fully commit to that approach deter you from doing the things you can make time for.

NBT – Never Be Testing

You might be in this group. Many small business owners and employees are. Do yourself a favor and upgrade to, at least, SBT. Think of it this way: If you’re just doing the same things over and over, all you might know is what your results are — not what they could be and not what they should be. It’s like if you’ve never had a vision test. You don’t know how strong your vision is. You just know how to get by with what you have. But if glasses would make your vision (and your life) immediately better, you’d absolutely make the change. You just need to give it a shot!

Leave a Reply

Your email address will not be published. Required fields are marked *