7 Ways to Fail Better – Your Guide to Jump Starting A/B Testing for PPC Ads

So you’ve started a Google AdWords campaign.  You have a campaign configuration that makes sense. You’ve selected keywords that are relevant to your business, and you’ve narrowed down your geographic targeting and day-parting. Now you want to get the most out of your ads.

A/B Testing is a great way to optimize your click-through-rate, learn what ad creative is most compelling and cater your overall marketing message best to your customers. To get you started we’ve broken down 7 simple ways to get the most out of A/B testing, and improve your overall campaign performance.

1. Only Test 1 Difference

To get the best results from A/B testing, you must isolate only one difference between two ads. For example, only change the headline and keep the ad description lines and destination URL the same. Also, only test 2 ads at a given time.  Testing more than two ads will dilute result and prolong the testing timeline.

Testing more than one variable at a time, such as headline and call to action, is a multi-variant test, and is a different testing method.

2. Run Tests Simultaneously

If you test one version one week and the second the next, you’re doing it wrong. It’s possible that version B was actually worse but you just happened to have more search volume while testing it. Always split traffic between two versions simultaneously.

3. Rotate Ads Evenly

This is a simple setting in most ad interface like Google AdWords, Bing Ads or LinkedIn Ads.  The Rotate Evenly setting gives equal preference to all active ads in your ad group, regardless of how each ad performs.  Other ad rotation settings include optimize for clicks, optimize for conversions, and rotate indefinitely.

4. Collect A Large Enough Sample

There is a concept called “statistical significance” that determines whether your test results are conclusive.  This simple calculation tells you if your test has run for enough time and if you have collected enough data.  There is a simple online calculator available if your current platform doesn’t have one.

5. Trust the Results… Not Your Gut

The winners in A/B tests are often surprising or unintuitive.  Your goal with the test is to improve click-through-rate, so don’t reject the results because of arbitrary judgment.

6. Conclude the Test

The value of A/B testing is lost if you don’t conclude the test. Pause the failed ad, and continue your campaign with the winner of your test.

7. Don’t Stop There

Continue optimizing your PPC ads by testing additional variables.  If you tested the CTA in the ad copy, try testing a different headline or promotion.

These 7 tips are just the tip of the iceberg when it comes to optimizing a PPC campaign.  Learn more about paid media and the advantages of a well-executed campaign here.