In Mid-2018, Google Ads rolled out Responsive Search Ads (RSA’s) to advertisers. At the time, the general consensus was that RSA’s performed poorly in PPC. Now in 2021, RSA’s are a generally accepted format in rotation with the older Expanded Text Ads (ETA’s).
However, I’m going to challenge the argument that RSA’s aren’t performing well. They are, you’re just looking at them the wrong way.
First, What are RSA’s?
Responsive Search Ads contain the following:
- 15 Headlines: versus the 3 that an Expanded Text Ad can contain. Google dynamically chooses which mix of headlines to show unless you’ve “pinned” certain headlines into certain positions.
- Up to 4 Descriptions: versus the 2 you get with Expanded Text Ads, but these descriptions can also be 90 characters long. Google also dynamically determines which description to show.
RSA’s are intended to give Google’s algorithm more options to choose from in order to help you improve your clickthrough rate, quality score, and ultimately performance.
The Wrong Way of Looking at RSA’s
It seems simple, you should add RSA’s into your ad groups and then compare the performance of the ETA’s vs. the RSA’s right?
Here’s a screenshot of some data, at first glance you’d think the ETA’s outperformed the RSA and we should pause the RSA!
This doesn’t tell the whole story though, now let’s look at the right way.
The Right Way of Looking at RSA’s
In 2019 I got the opportunity to speak with a higher up engineer at Google regarding this very topic. I asked him straight up why the RSA’s are performing so poorly. He got pretty antsy and said “they’re not, it’s such a common misconception!”.
The key sentence was this; the intention of an RSA isn’t to beat the ETA’s, it’s to improve the overall ad group performance.
He then went on to explain that the RSA’s, when introduced to an establish ad group, instead of stealing all the traffic and wrecking performance they’ll take the worst performing traffic and try to improve it.
This means that the right way to measure performance of introducing an RSA to an ad group isn’t to compare it to the ETA’s, but look at the performance of the ad group overall before vs. after.
Let’s go back to that ad group from earlier and look through this lens:
The RSA’s performed worse, but the overall ad group CTR increased including the ETA’s!
So What Do We Do? RSA’s, ETA’s or Both?
We believe in having a couple ETA’s and at least 1 RSA in every ad group. This gives you, the advertiser some control but also ensures Google has some “room to play”.
In 2021, we’re seeing RSA’s even outperform ETA’s more consistently as they’re being introduced on a level playing field vs. being thrown into an established ad group with refined copy. We’re also seeing nice quality score improvements as well.
Not to mention, we can imagine a scenario where RSA’s are eventually the only ad format. Get comfortable with them and learn how to use them properly now while you’ve still got the old ETA’s to use as backup!