Is your Media Mix correct? Find out how much revenue you're leaving on the table at your current spend with our free audit.
Is your spend allocation correct? Find out how much revenue you’re leaving on the table with our free audit.

4 Best Practices for Responsive Search Ads

April 4, 2019

Google’s responsive search ads (RSAs) have been up and running since mid-2018, and several best practices are emerging that focus on a few key areas:

  • Strategies for getting started
  • Creative variety
  • Pinning
  • Optimization


Here, we detail each of these ways search advertisers can create effective and high-performing RSAs.

1. Just Get Started


The first step is to stop procrastinating—add a few RSAs to existing search campaigns alongside text ads. In the search ads app, there’s a recommendation link on the upper left side of the page that brings up the RSA ads that Google suggests adding to ad groups. If you opt to not use these recommendations, focus on a few high-volume ad groups, with a minimum of 100 weekly impressions.

It’s important to provide Google’s machine learning with enough impressions to test and understand which ads work best with which users. Choose headlines both with and without keywords, either by using dynamic keyword insertion or static headlines that use a variety of terms. When creating headlines and descriptions, think across the funnel—users search differently when they’re top of funnel versus about to make a purchase.

2. Variety Is the Spice of Search


Be sure to max out the assets for RSAs and provide 8-10 headlines and 3-4 descriptions. Include a variety of creative that focuses on products, services, problems solved, benefits, types of users, and calls to action.

Remember that users respond differently to different approaches—one person may be looking for discounts, while another may be focused on luxury brands. The more varied the assets, the more Google’s machine learning can predict which messages will appeal to which users.

3. Pin... sparingly


RSAs are designed to use machine learning to fully customize ads to meet user need based on what Google knows about them. Although pinning may be necessary for legal compliance within the organization, it also constrains the machine-learning algorithms to do what they’re meant to do. When using pins, maintain a variety in each spot. For example, pin two or three variations of a headline or description to each position to let the machine learning work.

4. Optimize for High Performance


RSAs may need fine-tuning or pruning of any low performers. Look for ‘0’ impression assets and see if they’re too similar to other assets (i.e., not enough variety in the headlines and descriptions). Remember to use dynamic and static keyword insertion to leverage great keywords.

And, don’t forget to make sure RSAs are compatible with campaign extensions. RSAs show up just like any other search ad—users don’t see a difference. To measure RSA impact, set campaigns to ‘auto rotate’ for an apples-to-apples comparison with text ads. As data accumulates, tweak and fine-tune RSA assets to continue to improve performance.

RSAs: Part of the Modern Marketer’s Journey


Many machine-learning innovations like RSAs are providing marketers with a valuable advantage: the opportunity to focus on being creative rather than being spreadsheet jockeys. The goal of all marketing is to connect with customers in a meaningful, personalized way. RSAs are one more touchpoint on that journey.

For more information on using RSAs to gain a tactical and competitive advantage, download our guide, Getting Started with Responsive Search Ads.

responsive search ads



________________________________

Feeling overwhelmed by your advertising campaigns? Donate 15 minutes of your time and we’ll show you how MarinOne allows you to easily manage, measure, and optimize all of your ad campaigns from a single location. Request a demo today.

Andrew Todd

Marin Software
By submitting this form, I am agreeing to Marin’s privacy policy.

See why brands have relied on Marin to manage over $48 billion in spend