Responsive Search Ads: Text Ads Enhanced by AI

Responsive Search Network Ads, commonly known in English as “Responsive Search Ads”” and often abbreviated as “RSA” , are Google's latest innovation for search ads.

Responsive Search Ads are an ad type that is optimized by Google through automatic optimization involving Artificial Intelligence (AI) to deliver custom-tailored and relevant results to users. These new ads aim to sustainably enhance both the advertiser's performance and the user experience.

In May 2018, Google Inc. introduced its newest form of text ad. Currently, they are still in the BETA version on the Google Ads platform, but RSAs are expected to be available to all Google Ads users by the end of the year at the latest. For now, this new ad format is only accessible to selected advertisers.

Additionally, Responsive Search Network Ads are available in the latest version of the Google Ads Editor.

What are Responsive Search Ads?

The differences between Responsive Search Ads and traditional Expanded Text Ads (ETA) can already be seen in their structure. Instead of fixed text ads, RSAs are made up of individual text components. These are then algorithmically assembled by Google during successful delivery to create a customized search ad.

Fig. 1: Example of a classic Expanded Text Ad.

Fig. 2: Example of the structure of a Responsive Search Ad

As before, Responsive Search Ads consist of headlines and descriptions. The difference, however, lies in their number and length. Instead of two, there are now up to 15 headlines each with 30 characters. There are up to four descriptions (previously only two!), which can even be up to 90 characters instead of 80. This innovation gives advertisers more freedom in designing their ads.

Another innovation is the free combination of individual text components. In RSAs, Google Ads recombines the headlines each time and delivers them with a new version of bundled descriptions.

In Expanded Text Ads, these are always fixed and delivered as a static, pre-crafted ad by the advertiser. This new concept entails a restructuring of existing ad texts and should definitely be considered by the creator.

Why the integration of Artificial Intelligence in ad formats is so important

The enormous potential that Artificial Intelligence brings not only to marketing has become indispensable for Google Ads. Through machine learning, ads are not only more personalized and relevant for the user, but also convince with higher quality and an optimal user experience.

Machine learning helps to optimally tailor ads to users after a short learning phase, offering ads that suit their interests individually. This allows better targeting of different audiences and optimal, individual targeting.

With the help of Artificial Intelligence, creativity can be combined with efficiency to provide a better advertising experience for the user. This effect can generate up to 15% more clicks, ultimately directly benefiting the advertiser.

Artificial Intelligence thus offers advantages not only in terms of individualization to users. Advertisers benefit in the form of financial advantages (e.g., better budget planning, lower CPCs), with a clear focus on improving customer retention.

How do Responsive Search Ads work?

In Responsive Search Ads, the previously created headlines and descriptions are automatically combined by Google. This way, user-tailored variants can be offered and the best modification delivered. Therefore, the process is considered very user-friendly.

During creation, up to three of the 15 pre-made headlines are selected, and afterward, two of the four descriptions are displayed in sequence. This process increases the number of over 43,000 possible combinations significantly, increasing the delivery probability.

Fig. 3: Division of headlines and descriptions in the new Responsive Search Ads format

Good to know: What to consider when using Responsive Search Ads

Unfortunately, RSAs are still not usable by all advertisers and are still in the BETA phase. However, it can be assumed that these will be made accessible to all advertisers by the end of the year. It also means that BETA version users should always thoroughly check performance, as data for performance improvement still needs to be collected.

The basic idea of RSA is to offer users AI-adapted ads based on their previous user behavior to sustainably enhance their user experience. On the other hand, not the entire ad scope is delivered, especially for better visibility on mobile devices or frequently searched results pages (SERP).

Fig.4 : Example of an RSA in mobile preview

Even when not all headlines and descriptions are delivered under all circumstances, essential parts of an ad, like a call-to-action, should be present in all delivery options to ensure high ad quality.

To achieve this, it is possible to pin headlines and descriptions. This can be done both on the Google Ads interface and in the Google Ads Editor to set the order of headlines and descriptions. These are then always played at this position, allowing a small form of control by the advertiser.

Fig. 5: View of creating Responsive Search Ads in the Google Ads Editor with the option to pin

However, pinning also prevents gaining new important delivery insights. Through Google's ad delivery, for example, popular and interesting text components can be filtered out, particularly appealing to the user, which can later also be included in ETAs.

Pinning one headline can also lead to a reduction in testing possibilities by 75%, while pinning two titles can result in a 99% reduction, significantly limiting the machine learning capabilities on which RSAs are ultimately based.

With too many pins, data losses that could contribute to performance improvement must be expected.

In the end, Google's recommendation should be followed to run Responsive Search Ads alongside Expanded Text Ads (ETAs) and see them more as an addition. It is best to use one RSA and at least one, preferably three ETAs per ad group.

Checking RSAs: The Google “Ad Strength” Indicator

Since early September, Google Ads has gradually introduced the “Ad Strength” indicator to evaluate Responsive Search Network Ads and better assess expected results.

This is done by an evaluation through factors like relevance, delivery quantity, and text component variety. Unlike the quality factor, which is based on a numerical rating of 1 to 10, a scale from “poor” to “excellent” is used for evaluation.

Additionally, RSA previews will soon be possible on the Google Ads interface, to better understand the user's perspective and make optimizations.

Fig. 6: The Google Ad-Strength Indicator can help improve the quality of RSA ads

What are our tips? Discover best practices here

To make the best use of Responsive Search Ads, we’ve compiled a list of tips and best practices for you.

• We recommend setting up at least five instead of the required three headlines. They should differ in their quality through diversity. Additionally, a keyword should be included in two of these titles.

• Try out different lengths of headlines! Don’t always use the full length, as this might overwhelm the reader.

• Pay attention to different content in descriptions, as there is a chance that when combining descriptions, duplicate information may appear, diminishing the user experience. Better ensure different statements in descriptions to guarantee delivery compatible with other description alternatives. Redundant information in ads leads to system restrictions in Google and, in the worst case, will not be played anymore. Focus on various USPs to persuade customers with diverse advantages.

• Thanks to their length, Responsive Search Ads have a large visual scope upon delivery. As a result, more of the user's attention is drawn to the ad. Use this attention to your advantage!

• Be careful when dealing with call-to-actions in descriptions. Too many special characters like exclamation marks can endanger delivery. Also, too many prompts in the ad can intimidate the user as a buyer or someone interested in your ad.

• For more guidelines and instructions on creating Responsive Search Network Ads, visit the  Google Ads Support Page.

And what does it look like in practice?

In practice, it quickly becomes apparent that RSAs need time to play in and deliver suitable data to users. Once there is a learning effect through AI in delivery, important KPIs develop significantly in a positive direction, and costs per individual conversion can sustainably decrease.

Furthermore, it can be observed that the bounce rate decreases with RSAs. A greater number of user-relevant information displayed in RSAs helps the user decide on a follow-up click on the ad. As a result, a longer session duration can be expected, as the user is more conscious when choosing the ad.

As mentioned previously, RSAs should definitely be used in parallel with ETAs to recognize and evaluate developments. Google also plans to soon introduce a format for Responsive Display Ads to apply the concept to the Google Display Network as well.

Currently, RSAs are only available to selected customers and in the BETA version. We have also successfully tested RSAs with our clients and noted differences from ETAs. For example, despite the short use of RSAs, we can also record an increase in CTR by 5-15% caused by RSAs.

Conclusion: Ad optimizations through Artificial Intelligence: The Google trend is now more evident than ever in Responsive Search Ads. The extensive potential of these ads combined with Expanded Text Ads helps advertisers better reach their audience and enhance their online experience.

What can we do for you?

Looking for support with ad optimization? We are happy to advise you on the topic of text ad delivery and your options.

We look forward to your inquiry.