Everything Counts in Large Amounts: Measurement & Attribution for Bus Media

November 9, 2023

Because out of home is an old-school medium, sometimes people mistake it for an old-fashioned one, especially in terms of measurement. But with its recent technological advances, OOH now stands toe-to-toe with top digital formats in terms of delivering the data and insights needed to determine success. And that's true not just for billboards but transit media too.

That’s why when Tinder brought the “It Starts With a Swipe” campaign to over 750 Los Angeles buses, we partnered with StreetMetrics, an independent third-party measurement and attribution provider, to determine how impactful the campaign was in driving downloads for the dating app. We know that dollar for dollar, out of home drives 4.8x more app downloads than any other medium (SOURCE: Comscore, 2022). But traditionally, there’s been a challenge in measurement and attribution for bus media: it moves. But now we can determine exactly the makeup of the audience our bus advertising reaches, and the difference exposure to it makes.

 
That means that the kind of duly-diligent, research-driven, comparison-shopping approach that someone might use, say, to buy a car – isn't what's happening here.

We sat down with Suzette Jacobowitz, OUTFRONT Director, Marketing Research and Insights, to learn a little more about how bus media attribution works.

“StreetMetrics essentially created a moving view shed,” she said, “so as buses move around the city, they capture [unique IDs of] devices that are within viewing distance of the bus.”

Using anonymized, opt-in geolocation data from mobile devices to measure the audience exposed to a campaign is the first step in executing a control vs. exposed methodology. Control vs. exposed works by measuring two populations with nearly identical characteristics save for one difference: one group has been exposed to the ads and the other, the control group, has not. So how does StreetMetrics identify the control group?

 
“They’ve created what they call a ghost bus,” explained Jacobowitz. “It follows along just a few minutes behind the actual bus. So, it’s still capturing the same type of audience you would find in a specific area. The only difference is, these are people who didn’t have the opportunity to see the bus, because it already passed by.”

The final step in the control vs. exposed process is counting the number of app downloads for each group, using those unique IDs to link devices to one or the other.

We asked Suzette how StreetMetrics protects people’s privacy in this process. “We don’t collect any personal information about anyone,” she answered. “Literally just counting devices along the way. Even the measurement partner doesn’t see anything personal.” In fact, the only data collected is the list of device IDs.

 
Tinder’s campaign ran for six weeks on buses in the San Fernando Valley and the Westside of Los Angeles, with 637 Tails on bus rears, 110 Kings on their sides, and five full wraps. How did the ads perform? Overall, those exposed to the bus media were 79% more likely to download Tinder than the control group, downloading the app 1,045 times.

“The Tinder campaign was our first app download study and it was the perfect one. It was a really big campaign, they used three different media formats, and we were able to analyze results and break it down to the different formats.”

 
“People always question if something costs more, is it really worth it? This campaign used five full wraps, and full wraps are obviously more expensive. But we were able to show on a per download basis, the full wraps actually delivered more downloads than the tails or kings.”

Jacobowitz shared with us some best practices to optimize a campaign for measurement. “I always recommend that if you're looking to measure lower-funnel metrics, such as app downloads (or footfall or website visitation) you want a campaign that has higher-frequency campaign over a longer duration as the repetitive exposure helps to impact a change in behavior. And reach is important, especially for measuring, not because every campaign needs to be large in scale, but all our measurement partners need to be able to see a certain number of devices in order to have a statistically significant sample.”

 
She also stressed the importance of a clear call to action. “Tinder, obviously the goal was download the app. But if it’s Coke and they want to do footfall studies to a supermarket, if there’s nothing in the ad that says, ‘go to the supermarket’ it doesn’t really work.”

While StreetMetrics can measure footfall and app downloads, that’s just the beginning of the measurement and attribution capabilities that OUTFRONT can offer to our advertisers. Tune-in, sales lift, and brand lift are just three of the other ways that brands can measure the impact of out of home.

And we’re ready to make that impact for you. So, let's chat!

Author: Jay Fenster, Marketing Manager @ OUTFRONT

Links to third-party content are not endorsed by OUTFRONT Media.