As a consultant specializing in digital advertising, I have one particular conversation over and over again: clients want to know exactly which people in our custom audiences we’re reaching and where we’re reaching them.
My answer, that we can’t legally know exactly who in an audience has seen our ads, is never sufficient. So it becomes an endless quest if our ads are performing as we work hard to make sure they do, how do we make that clear to our clients?
Our solution is two-fold. On the one hand, we have access to specific metrics that help us articulate the impact our ads are having. On the other hand, there are more and more tools available that allow us to tell a story, using our ad data.
When it comes to metrics, we look for a few key markers. In Facebook ads, we pay attention to estimated ad lift recall rate: the rate at which audiences are likely to remember our ads two days after seeing them. It’s a rough estimate and god only knows how Facebook calculates it, but in our experience, it’s a good place to start.
If we’re running video ads, across any platform, we pay close attention to video completion rates. If more than half of the impressions we deliver end with people watching the video all the way through, then we know the content is engaging, timely, and impactful. Tracking these rates over time also helps us understand when our video content stops having the impact we need.
Next, we consider the overall goal of the campaign. For example, if we’re running a list acquisition campaign, we’re paying attention to link clicks and the link click-through rate (CTR), as well as reach. If the reach of these ads is low and stays low, then I’m probably sending ads to the same people over and over again. If link clicks are low, then no one’s engaged enough with our content to click through and sign up.
In a way, we use metrics to articulate how we aren’t failing and to change tack when we see our metrics begin to slip. Taking a step beyond our specific ad goals, we take the client’s goals into account.
In 2018, we worked with the ACLU to turnout low-propensity voters (LPVs) in three of California’s congressional districts. For this project, we spoke to voters regularly for months leading up to the election. Knowing that the LPVs would need to hear from us more times and more often than likely voters would, we focused on delivering impressions, layered data from different sources to reach as many people in an audience as possible, and ignored frequency.
As we aggregate these specific types of data, we have to keep in mind that how we deliver data to our clients is almost as important as what kind of data we deliver. When we send reports, we compare our data to results of previous ad campaigns we’ve done for the client, to industry standards, and to the averages we see on other, similar campaigns we run to offer background and context.
The tools at our disposal are largely dictated by our client’s budgets. With many of them, we have to get creative.
Tracking data over time and across platforms is key to articulating our impact. The cheapest way to do that is to build a spreadsheet and manual enter target data every week. With a bit more money, you can build or commission someone to build a custom spreadsheet that automatically loads and organizes your data. It’s rudimentary but gets the job done. Then there’s DoubleClick Campaign Manager, the Rolls-Royce of tracking ad data. DCM aggregates data from all ads across all platforms into a central location. It allows you to track, run, and report all from one place, but it’ll cost you a pretty penny.
About a year ago, Facebook rolled out “Brand Lift studies,” which splits your Facebook and Instagram audiences into two groups, runs your ads to one group, then polls both groups to see a side-by-side comparison that helps analyze your brand lift and ad recall. Of course, this only helps with Facebook and Instagram ads and I, for one, am skeptical about Facebook’s own tools that usually come with a few faulty features and no transparency whatsoever.
Some consultants and pollsters are starting to build and run their own polls to understand the impact of digital ads. This involves very specific skill sets, but probably has fewer glitches than a tool designed by Facebook.
When it comes to articulating data and our overall impact to our clients, we focus first and foremost on offering context. We use a combination of the above tips, tools, and ideas to do more than regurgitate numbers back to our clients.
While we might not be able to answer the question of which specific, individual people we’ve reached with our ads, we can still demonstrate our impact by articulating the data to our clients so they see it within a framework, rather than seeing numbers floating in the ether.
Emily Gittleman is the digital director at 50+1 Strategies, a political advocacy/campaign consulting firm based in Oakland, Calif.