Why You’re Measuring the Wrong Display Metrics
One of the primary advantages digital advertising has over traditional media is the ability to directly attribute sales to ad impressions. Digital attribution is far from a perfect science, but online ads can be connected directly to revenue. In television or print, that level of attribution is still virtually impossible, despite big steps in the right direction.
Even with the enhanced attribution of digital ads, many advertisers aren’t using meaningful metrics to assess the value driven by their display campaigns. Here are a few do’s and don’ts of display measurement.
TV Metrics Weren’t Built for Digital
Traditional media has its own set of success metricsand they’re far outstripped by anything the web has to offer. One of the biggest mistakes advertisers make online is taking those metrics and attempting to apply them to digital.
Why limit yourself when you have access to so much more?
Even when advertisers do take advantage of the superior performance data digital has to offer, it’s easy to focus on the wrong numbers.
Stop Counting Clicks
When it comes to online advertising, clicks are one of the most common measurements of success. Its tempting to look to clicks as a performance indicator because they represent a very tangible action, but all the reliable evidence suggests that clicks are not an accurate measure of performance for display.
One comScore study, for example, found that clicks had absolutely no correlation with conversions. So even if people click on your display ads, theyre no more likely to convert than those who didnt. Ultimately, advertising is about building a brand or it’s about making sales. A click doesn’t necessarily indicate brand recognition, recall or awareness, and it certainly doesn’t correlate directly to sales.
A significant percentage of Internet users don’t click ads. Ever. Regardless of the demographic makeup of clickers vs. non-clickers (there has been some discussion about whether clickers are or arent valuable targets), there is a significant number of people who just don’t do it, but that doesn’t mean they aren’t influenced by ads. That same comScore study also found that being served an ad does correlate strongly with conversions, indicating that seeing an ad can lead someone to convert who otherwise wouldnt. Failing to recognize that unclicked ads have an impact will understate your campaign performance and lead you to underinvest in channels that are actually driving revenue.
That brings me to the next point: the validity of the view-through conversion.
Don’t Ignore View-Throughs
Often, advertisers only want to attribute sales to a display campaign when a user clicks on an ad and then immediately makes a purchase, fills out a lead form, or completes some other desired action.
The idea that display ads lead to conversions even when users don’t click (referred to as a view-through conversion) is often met with skepticism. But there is strong evidence to suggest that display ads can and do encourage users to return to a website directly or using a search engine.
In addition to the comScore study discussed earlier, Salesforce has reported that running display campaigns led to an 80% increase in branded search, or users searching directly for the brand name, which means that people who saw an ad used a search engine to visit Salesforce.com instead of clicking on an ad.
A more comprehensive comScore study found that retargeting campaigns, on average, led to a 1,046% increase in branded searcha clear indication that retargeted users were using search engines to return to the advertisers site. The comScore study covered 103 campaigns run by 39 different advertisers in 7 industries.
Our own clients who have set up tests to check the validity of view-throughs consistently find that retargeted users convert at higher rate than those who don’t.
Setting up a test to ensure view-throughs are valid can be a good strategy, but the evidence consistently shows that view-throughs are valid.
The success of a given display campaign is dependent upon your goals. A good outcome for one marketer may not be a good outcome for you. Before launching a campaign, consider what you want to accomplish and set your metrics accordingly.