Did you know that traditional A/B testing doesn't work with Responsive Search Ads (RSA)? In this article, we're going to show you how to A/B test RSAs according to the latest best practices so that you can achieve the best results.

What Is A/B Testing When It Comes to Ads?

Even inexperienced PPC advertisers who have no specific ad testing goals have most likely already done a significant amount of A/B testing by simply adding several ads to an ad group. This way Google can show the ads that perform better more often thanks to the optimized default ad rotation settings.

Advanced advertisers typically select the option to rotate advertising indefinitely and use their own selection method to choose the best and the worst ads.  Most of the time, advertisers are more concerned with conversion rate than with CTR. However, concentrating exclusively on conversion rate misses the chance that an ad with a high conversion rate could receive relatively fewer clicks because of a poor CTR, resulting in fewer total conversions for the advertiser. Advertisers gain a better measure of which ad is best at accomplishing one of their goals by combining CTR and conversion rate into a metric called conversions per impression (CTR * conversion rate = conv/imp) (which is to maximize conversions).

Unfortunately, traditional A/B tests don't work for RSAs because they don't usually include impressions as a metric. So, without further ado, let's jump to Responsive Search Ads A/B testing best practices.

1. Include Impressions as a Metric

To be able to measure meaningfully, any A/B ad test, including RSAs, should include impression data, especially impressions volume. Even if two RSAs run on the same days and have been active for the same amount of time, impressions may differ because one is stronger than the other at qualifying itself for more auctions. In all of your evaluations, keep impression volume in mind. In this manner, you'll be able to measure the outcomes more accurately.

2. Use Ad Strength Indicators

As you create new RSAs, Google will offer instant feedback via the Ad Strength indicator. Ad strength is a best practice score that determines the relevancy, amount, and uniqueness of your responsive search ad content before your ads run. According to Google presentations, every improvement in ad strength translates to a 3% increase in clicks. Going from 'poor' to 'average' should result in roughly 3% more clicks, and rising from 'average' to 'good' should result in another 3% more clicks.

3. Use Headline Variations

The Ad Strength indicator considers how many headline and description versions you supply. Because coming up with 15 headlines can be difficult, we considered whether it was worthwhile to include more than a handful of headlines for each RSA.

4. Experiment with Different Ideas

Many ads will not get enough top impressions in a 30-day period to provide performance statistics, but there are still techniques to improve if this is the case for you. Remember that Google permits up to 3 RSAs per ad group, and that specific text can be pinned to different spots in the ad. You can use this to build some useful experiments.

If you only remember only one point from this article, it should be that a basic assumption regarding ad testing must be altered. Impressions are now heavily influenced by the ad itself, rather than just the keywords in an ad group. It's crucial to factor in for this potential variation in impressions when A/B testing RSA advertising and strive to discover the ad that delivers the most incrementality.