Google Ads sunset Expanded Text Ads in favor of Responsive Search Ads in July 2022.

However, there is still a major issue with RSAs, which has persisted since they were first made widely available: a lack of asset-level performance data.

Advertisers can still see the usual metrics (clicks, conversions, etc.) for an ad as a whole. But what if you want to see which headline between “Contact Us Today” and “Speak to an Expert” had a higher clickthrough rate or drove more conversions? When you go to analyze performance, you won’t be able to easily differentiate between them like you could with ETAs.

One solution is to establish a more deliberate approach to RSAs before launching, which will save you time & stress during analysis later on.

 

Establish an RSA Plan 

 

Define Ad Themes 

The first step in establishing an RSA plan is to define your ad themes.

• Review the messaging guidelines, sales sheets, websites, etc. of your organization/client.
  • Note any common phrases or terms that might be usable in your ad copy, and then categorize those phrases based on the overall theme.
  • For example: use cases, important features, and general brand information.
• Choose 2-3 themes you want to test in your ad copy.
• Write copy for at least 2 RSA variations, differentiating them by theme.
  • For example:
    • RSA 1 has headlines and descriptions that speak to common customer pain points, with “Learn More” as the main CTA.
    • RSA 2 has headlines and descriptions that focus on specific product/platform features, with “Try for Free” as the main CTA.
• It’s also important to note that Google Ads has a limit of 3 RSAs per ad group, so prioritize the themes you (or the client) are most interested in.

Take Advantage of Pinning 

Once you’ve written the initial ad copy and defined your ad themes, the next step is to decide on your pinning strategy.

The main benefit of pinning headlines and descriptions is that it keeps your ad focused on a theme. This allows you to test more intentionally, making it easier to understand what types of copy are/aren’t working.

You can develop a pinning strategy any number of ways, including:

• Using multiple asset variations pinned in a specific order. This gives you (and Google) more flexibility to test, while also sticking to a tight theme. For example:
  • Write 15 headlines and 4 descriptions.
  • Pin branded headlines to position 1, CTAs and features to position 2, and general information to position 3.
  • Pin long descriptions to position 1 and short descriptions to position 2.
  • Save and run the RSA.
• Recreating Expanded Text Ads. This could be particularly useful for highly regulated industries, like banking. For example:
  • Write 3 headlines and 2 descriptions.
  • Pin each headline and description to the position you’d like them to show.
  • Save and run the Mock ETA.
  • Mock ETA variations may see the lowest amount of impressions compared to other RSAs, as they have fewer headline and description assets and Google has more flexibility to serve ads without pins. 

Example of Mock ETA Set-Up

The one downside of pinning is that you might generate fewer impressions. However, if you’re comfortable with that, the benefits of control can outweigh the benefits of having high impressions.

• You may also see a lower Ad Strength rating during set-up, but this rating does not actually impact how the ad is served.

Implement Labels 

The last step is to define labels, which is fairly straightforward. Labels help you quickly categorize ads based on their theme, and the “Labels” column can be added to reports you export for analysis.

Create the labels in Google Ads and assign them to the relevant RSAs. Then, whenever you download the Ad Report to analyze performance, make sure to include the Label column. From there you can create Pivot Tables, data visualizations, etc. based on the label to get a sense of what messaging themes are and aren’t working.

Example of Theme Labeling

Conclusion

There are a number of ways to test Responsive Search Ads, but it’s still difficult to work around the lack of headline and description data during an analysis.

One solution is to establish a more consistent, intentional approach to ad copy. Defining pin locations and labels will take a little more work up front, but it makes future analyses much easier.

Lastly, it’s important to note that there is some asset-level data to work with if you click “View Assets Details” under an RSA. It isn’t robust, but you can at least get a little extra insight into which headlines/descriptions might be top performers.

• In this section, Google Ads provides a Performance column. These ratings are fairly vague (Best, Good, or Low), and should really only be used as a general reference.
• There is also an Impressions column. This can serve as an indicator of performance based on the assumption that, if Google chooses to show a certain headline/description more often than others, it is likely a high performer (at least in their eyes).
• Additionally, you can navigate to the Combinations Report to see which combination of headlines and descriptions served together.
  • Ad variations are then listed in order of most impressions to least, so the assumption here is similar to the one above (more impressions = higher performer).