Google Ads: 3 Responsive Search Ads Best Practices

Are you applying best practices in your responsive search ads?

Not sure?

Well, it’s your lucky day.

I went down the RSA rabbit hole and figured I’d share. After all, they are where the rubber meets the road regarding kick-ass search campaigns.

Here are some takeaways and resources on properly creating and testing responsive search ads.

Starting with what might be a surprise to some PPCers…

Account Structure Matters

The first thing to know is, that it’s not just about the ads themselves.

Before you even create the ads, you need to take impressions into account.

According to Google, it takes 2,000 impressions to get a “Low,” “Good,” or “Best” rating.

If possible, you need to structure your campaigns and ad groups so that they are getting enough traffic.

I say “if possible” because you may be unable to hit that volume with hyper-local or small niche campaigns.

Here are the ad assets/RSA guidelines to keep in mind when planning your campaign.

Determine the amount of RSA assets to use to acquire sufficient data:

• 4 headlines — 15k impressions
• 6 headlines — 30k impressions
• 10 headlines — 35k impressions gets you to 80% sufficient data
• 15 headlines — rarely get above 30% sufficiency

Then, when you do have enough data:

• Replace low-performing assets.
• Remove low impression assets.

It’s not the A/B pick-a-winner test from the glory days of AdWords, but it’ll have to do.

And if I really, REALLY want to test the low-impression assets, I’ll add them to something called…

Fake ETAs

This old PPC dog learned a new term recently: fake ETAs.

If you haven’t heard, these are “pinned” RSAs dressed up like expanded text ads:

• 2–3 headlines
• 1–2 descriptions

By pinning specific headlines into specific positions, you can effectively force the ad to show a controlled combination — giving you cleaner test data on individual assets.

Measuring Responsive Search Ads Performance

Ed Leake has an excellent technique for measuring ad performance in his God Tier Ads course.

He uses custom metrics called CPI (for lead gen) & RPI (for ecom) to measure results.

• CPI = Conversion Per Impression
• RPI = Revenue Per Impression

Per Ed, “Blends two metrics you should care about, CTV & CTR.”

The higher number is better. This gives you a single number to compare ads across ad groups and campaigns — much cleaner than juggling CTR and CVR separately.

Additional Resources

Brad Geddes

Google Ads OG Brad Geddes has a post on RSA testing strategies you’ll want to read and bookmark. It’s a few years old, but the information is still very relevant. He outlines how he uses different ad themes, pinned vs unpinned assets, and more.

Read How to Test Responsive Search Ads

Search Engine Land

Be sure to check out Search Engine Land for excellent graphics to help you visualize more of the above outline. Bookmark it because the content updates regularly.

Barely Scratched The Surface

Again, these were just a few takeaways. RSAs are a HUGE topic, so I’m sure I missed something.

If you have any questions, you can contact me here. I’ve been running Google Ads campaigns for about 17 years (ten of them full-time). If I don’t have an answer, it’s likely I know someone who does.

If you want to master Google Ads independently, I can recommend a few courses, or we can set up some training for you.

Book a Free Strategy Session

Or grab my free Small Budget Manufacturer’s Google Ads Blueprint.

If you are looking for strategies you can use to optimize your Google Ads account but aren’t ready for a professional deep dive, you can grab my Small Budget Blueprint. It’s the same framework I use across all small budget campaigns.

Download the Google Ads Small Budget Blueprint

Until next time…

Have a good one,

Corey