Tips on testing responsive search ads

Cara deBeer of Catalyst discussed ad testing at SMX West last week.

Have you been testing Google’s responsive search ads, but are still not sure what to make of them? You’re not alone.

Cara deBeer, paid search director at Catalyst, and Brad Geddes, co-founder of Adalysis, discussed what they’ve seen so far from testing responsive search ads (RSAs) against enhanced text ads (ETAs) in a session at SMX West last week.

In her presentation, deBeer discussed some of the limitations of RSAs (namely detailed reporting) and ways to evaluate how your RSAs are really doing. Looking at case studies from client accounts, deBeer showed test results at the aggregated account level that seemed to indicate mixed results — lower click-through rates (CTR), higher conversion rates (CVR) — for RSAs compared to ETAs. Looking at various segments, however, the data looked a bit different.

Label and segment for deeper insights

“We need to dig into various segments to see areas for opportunity, even if it looks like RSAs aren’t doing well,” said deBeer.

She showed results segmented by device, match type, brand and non-brand, at the product level by geographic market to identify pockets of insights.

Her parting advice was, “Everyone should test RSAs, but not everyone should test them all the time.”

If you’re in a regulated area like pharma or need to tightly control your ad copy, RSAs are not for you. But, deBeer said, if you want to take advantage of the additional real estate of RSAs, are able to label the different ad formats for analysis and have a handful of top, high volume ad groups that will allow you to use existing reporting and be able to segment your results, then she recommends testing RSAs.

Geddes presented data looking across thousands of accounts. This aggregated data suggested using RSAs “if you want high CTRs on broad matched keywords and aren’t conversion focused” and to use ETAs “if you are conversion focused.” On the whole, Geddes says they see RSAs driving higher CTR and lower CVR than enhanced text ads. In other words, RSA can function well for driving upper-funnel traffic.

Back to basics

Both deBeer and Geddes stressed that results will vary and everyone should run their own tests.

To truly have success with RSAs — or any text ad format — Geddes said, advertisers need to put thought into their ads and get back to basics of messaging benefits versus features. When ETAs first came on the scene, for example, Geddes said people complained they didn’t work, but often were just tacking on their first part of a description line into the second headline rather than revamping their creative approach to the new format.

That includes the understanding that your third headline may not show and that ads should engage people.

“Most web CTAs are terrible,” he noted, to an audience of nodding heads. Instead of “Subscribe to our newsletter,” Geddes suggested something more useful such as, “Subscribe to powerful marketing tips.”