<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1811696792399086&amp;ev=PageView&amp;noscript=1">

Channels for A/B Testing

Posted by Laura Phelps on 11/3/17 4:30 PM

Understand the best channels for A/B Testing

Learn the small business essentials of digital marketing testing. 

Group 14.png

If you spend time on marketing blogs, you’ll quickly run into strategies around A/B testing. The difficulty of A/B testing runs a large range and it’s possible for you, as a small business owner, to run your own test or for a team to require a large data science team. Testing can get complex, fast. But, the good news is that you likely don’t need to do anything too complex for your own business.


So, what exactly is A/B testing? It’s what it sounds like -- you take one option and stack it against another and see how it does. In fact, you probably do this pretty regularly in your own life (you try both a hamburger and a salad to see which one will make you more full, for example.)

When you apply testing to your small business marketing and product strategy, you’re essentially creating an intended learning plan. Think about your question and then determine the best channel to learn your answer. For example, maybe you want to learn if potential customers are more likely to join your mailing list if they see product A or product B first. In that case, an online ads campaign may be the perfect way to A/B test by running two otherwise identical ads at the same time. Or, maybe you want to test which discount converts more customers? Well, your email list may be the best way test that by sending one discount to an otherwise randomized half of your list and the other discount to the other half.

 

Where do you start?

First, poke around in the tools that you’re already using for your digital marketing efforts. Many email CRMs have build in A/B testing platforms that make it simple -- usually just asking you to drop in the two subject lines or texts that you want to test. Similarly, ad platforms can help walk you through setting up an A/B testing plan specific to that channel. If you want to take your testing plan to the next level, you can consider using tools like Optimizely, VWO or Convert.

 

What are the pitfalls?

When you’re just starting out, keep things simple. When testing something like a subject line or a discount, try to keep everything else in your two messages the same. You don’t want to muddy the data by testing too much at once so it’s unclear what really made the difference.

Find a spreadsheet or Google Doc to keep the results in one place. You should reflect on your testing every month or so and think about what else you want to test and whether you’re truly implementing your learnings. For example, if you learn that most people click on an ad featuring product A instead of product B, then you probably want to feature product A more often in your email newsletter, on your flyers or radio ads and even in your storefront.

Finally, not all A/B tests are conclusive. Don’t try and derive learnings from something that really isn't there. If it comes back and it’s not clear that A is better than B then reevaluate the test. Were the two options different enough? If so, keep running it and adapting it to different channels until you have a conclusion.

Want to dig deeper into A/B testing? Sign up for the three-week, coach-led MainStreet Marketing course by BusinessBlocks. Sign up now to reserve your seat >

 

Topics: Marketing