Facebook is giving advertisers more data on how it grades ads - but buyers say it's a step behind Google
- Facebook is giving advertisers more information about how it calculates its relevance score, a measure of how users react to ads.
- Google provides a similar metric called quality score, and some advertisers said Google's metric is more straightforward.
- Facebook is also narrowing how it counts its potential reach, the disputed metric that it's used to compare its audience to US Census Bureau data.
Facebook is giving advertisers more data on how relevant their ads are.
Facebook's 3-year-old relevance score grades ads from one to 10 and tries to predict how an ad will be received by an advertiser's audience based on the feedback users give and the emoji reactions they click on. If an ad has a low relevance score, the marketer may tweak the copy or the page that an ad leads to. The higher the relevance score, the more effective the ad and the lower the ad rates.
Google offers advertisers a similar metric for advertisers called quality score that measures user experience.
Until now, Facebook's relevance score was a single metric. Starting April 30, it will share three stats that make up the metric, representing quality, engagement rate, and conversion rate.
The idea is to give marketers a better understanding of what factors go into calculating relevance score, particularly with performance-based marketers who obsess over data.
"We think this level of granularity will offer reporting that's more actionable for businesses," Facebook wrote in a blog post. "When used together, ad relevance diagnostics will help advertisers understand whether changes to creative assets, audience targeting or the post-click experience might improve performance."
Advertisers often complain that Facebook doesn't offer enough information about how ads perform. Andrew Sandoval, director of biddable media at The Media Kitchen, said the changes to relevancy score are an improvement, but that Facebook still keeps some data in its "black box."
Sandoval also said Facebook's approach isn't as "foolproof" as Google's. Google's score is based on data from search data - including whether someone clicked through and the user experience on the landing page - whereas Facebook's score is based on user reactions.
Facebook is also changing how it calculates potential reach
A couple years ago, Facebook came under fire when analysts pointed out that Facebook was selling an audience that purported to reach more people than the US Census Bureau reports.
Advertisers see the disputed metric, called potential reach, when they set up targeted ad campaigns. Potential reach estimates the number of people that an ad may reach before it runs. Facebook also reports campaign reach, which represents how many people a campaign reached after it ran.
Facebook responded that its audience should not be compared to Census data because people often have more than one account or are of relevance to different marketers when they travel, which justifies counting them more than once.
Until now, Facebook's potential reach was based on all of its monthly active users. Going forward, the company will only consider potential reach to include monthly active users who have viewed an ad in the past month, which in theory will help advertisers better target ads.
The Media Kitchen's Sandoval said that the change could help advertisers fine-tune their campaigns but warned that Facebook's definition of reach should be taken with a grain of salt. Today, he said the agency often adjusts data from both Google and Facebook before clients see it.
"To be fair to the platforms, there's a lot of variables that we can't know."