“Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”
Many attribute this quote to John Wanamaker, the founder of the department store that bore his name. The quote is attributed to others, too, but the source isn’t important. What’s important is that it’s incredibly hard to know if advertising is working or not. Marketers perpetually struggle with this.
In 2009, I saw results of work Google did with Lipton Tea. Google ran ads for a new white tea alongside search results, on YouTube, and on its display network. If these ads had been on TV or radio, it would be difficult, if not impossible, to measure the impact on retail sales. Even with online advertising, it would typically be difficult to make that connection.
But Google and Lipton took their analysis a step further in two ways.
First, They used comScore’s online panel to measure the difference in web engagement with Lipton assets between those exposed to the advertising and those who didn’t see it. They also conducted survey research to understand how the advertising affected attitudes and opinions toward Lipton and this new tea.
Second, and most interesting, they tied online ad viewers (and non-viewers) to Kroger loyalty cards. From there, they measured whether actual purchasing behavior changed. In a real sense, Lipton measured the change in sales attributable to its advertising effort.
Let that sink in for a minute. The researchers could tell if someone had seen an ad. And if a person saw the ad, they could measure whether or not their purchasing behavior changed.
Lipton dollar sales per household increased by 26 percent, which is far from trivial. (See the full results.)
Though this research is now 7 years old, I’m surprised to have not seen it used more broadly in the industry. In the past few weeks, I’ve seen links to not-so-new revelations of Google’s ability to track when users visit specific stores in real life. If you run an ad for a retail location, you can track whether those who saw your ad visited your store or not. See Search Engine Land’s 2015 article on how this works, and another article that mentions how PetSmart and Office Depot use these metrics.
Measuring sales for individual consumers is just getting started and I think there will be much more to come. As ordinary consumers, we’re all going to start receiving much more targeted advertising. (And I think this will make ads more interesting for most people.) And marketers are going to have unprecedented ways to have their messages spread effectively.