Easy Ways to Start A/B Testing

February 6, 2018 Sanaa Hajoui

A/B testing allows you to learn important insights about your customers and what they respond to online. It means that in all of your digital marketing materials, from website to search and display ads and social media, you’re putting your best foot forward and using creative and deployment strategies that have been proven to work. What makes them proven? Why, you’ve tested them!

If you’ve never done A/B testing before here are some easy ways to get started (and if you’re not clear on what A/B testing is, or how it can benefit your business, find out what you need to know here).

1. Determine what to test

From website copy to e-newsletters and display ads, it’s possible to scientifically test nearly every aspect of your digital marketing strategy. The question, really, is where to start. To get the best bang for your buck, make a list of all of your digital marketing initiatives and think about how they contribute – or could potentially contribute – to your bottom line.

For example, while there are some neat hacks for A/B testing your Facebook posts, if you use the platform primarily for the purpose of creating engagement with your brand, but not to actually drive sales (though Yellow Pages can help you with that!), this may not be where you want to spend your time. Elements such as the placement and wording of website calls to action, newsletter subject lines or the timing of your deployment, and search and display ads, can really move the needle for your business.

2. Form your hypothesis

Have a hunch about how to change an element of your campaign to impact your leads or sales? That’s your hypothesis, and the item you’ll want to test. Let’s say you have a barbeque repair company and the image on your display ads is of a grill – but you think a photo of a tasty steak might do better. That’s your hypothesis. If you’re testing your newsletter, your hypothesis may have to do with the types of subject lines you use, the sender of the email (for example, Steve from Grill Company or just Grill Company), or the timing of your newsletters (say, Saturday morning vs. Monday morning). Your test will compare the status quo with your new hypothesis.

3. Pick your test

Once you’ve identified which marketing initiative and element you want to test, you’ll need to figure out your methodology. If you use Google Analytics you can use Content Experiments to test up to 10 variations of elements such as images, headlines, page layouts and calls to action for free. Email deployment services such as MailChimp allow you to test newsletter elements including subject line, sender name, content, layout and send time, while Yellow Pages can help you use A/B testing to optimize your search and display ads for maximum performance. You’ll also need to determine how long you want the test to last and whether you want to target the test to specific audiences.

4. Deploy and measure

Once you’ve identified what you want to test, why you want to test it, and how you’re going to do so, it’s time to get testing! Deploy your A/B test and wait for the results to come in. Yellow Pages Analytics tracks interactions and contacts, making it even easier to compare performance. Once you have the data, you’ll be able to crunch the numbers to determine which version produced the best results. You can then continue testing other elements of your marketing initiatives or start to make changes based on the learnings of your first A/B test.

Testing delivers better results

No matter what element of your digital marketing campaign you are looking to test, it’s easy to get started on A/B testing if you break it into four easy steps (and even easier if you let Yellow Pages experts design and implement the testing for you).

Previous Article
Infographic: 7 Tips for Creating Strong Calls-To-Action
Infographic: 7 Tips for Creating Strong Calls-To-Action

What's a call to action? A CTA is a digital button or link on your website encouraging users to take a spec...

Next Article
New Website Budgeting Template
New Website Budgeting Template

With the cost of a new website ranging from free (if you build it yourself using a basic template on a plat...