MOUNTAIN VIEW, CA: Simple A/B testing of digital search ads should be retired in favour of a multi-level option that is now enabled by ad rotation, according to a leading Google executive.
Writing in Search Engine Land, Matt Lawson, the Director of Performance Ads Marketing for Google, explained that optimized rotation of ads allows marketers to have more ads present in an ad group and increase their chances of finding the right match across all of the variables that Google AdWords considers.
“Going from one ad to a minimum of three ads in an ad group can give that ad group up to a 5% to 15% lift on average in both impressions and clicks (with incremental uplift for each additional creative),” he stated.
In terms of testing, he suggested that users could test one ad (A) against an experiment ad group with four ads (A, B, C and D) with rotation set to optimized.
That way “you don’t need to test ads A and B against each other anymore. Those two ads now work together instead”, he said.
So instead of finding a “winner” and scrapping the “loser”, users “should plan on having options A, B, C and D all active at any given time” with an ad only being deleted when it generates minimal clicks at which point it can be replaced with a new one.
Lawson also cautioned against focusing on CTR to assess an ad’s effectiveness. “CTR alone can be misleading because ads show on all sorts of queries in all sorts of contexts,” he pointed out.
“There’s no way to control for all of the different devices, locations, situations and everything else that goes into one auction.”
Accordingly, he recommended using ad group-level metrics, including impressions, clicks, conversions and CTR.
Writing in Search Engine Land, Matt Lawson, the Director of Performance Ads Marketing for Google, explained that optimized rotation of ads allows marketers to have more ads present in an ad group and increase their chances of finding the right match across all of the variables that Google AdWords considers.
“Going from one ad to a minimum of three ads in an ad group can give that ad group up to a 5% to 15% lift on average in both impressions and clicks (with incremental uplift for each additional creative),” he stated.
In terms of testing, he suggested that users could test one ad (A) against an experiment ad group with four ads (A, B, C and D) with rotation set to optimized.
That way “you don’t need to test ads A and B against each other anymore. Those two ads now work together instead”, he said.
So instead of finding a “winner” and scrapping the “loser”, users “should plan on having options A, B, C and D all active at any given time” with an ad only being deleted when it generates minimal clicks at which point it can be replaced with a new one.
Lawson also cautioned against focusing on CTR to assess an ad’s effectiveness. “CTR alone can be misleading because ads show on all sorts of queries in all sorts of contexts,” he pointed out.
“There’s no way to control for all of the different devices, locations, situations and everything else that goes into one auction.”
Accordingly, he recommended using ad group-level metrics, including impressions, clicks, conversions and CTR.
Sourced from Search Engine Land; additional content by WARC staff
