Marketers, here’s how to understand what’s really behind your mobile ad visits metric

Marketers appreciate metrics that give them a clear sense of how their campaigns move the proverbial needle. On its face, the “exposed visits/visitors” metric used ubiquitously throughout the world of mobile ad campaign measurement is as clear as it gets.

Exposed visitors represent the number of devices that were exposed to an ad that subsequently were seen in a campaign location. Essentially, exposed visits help marketers gauge the reach of their campaigns by reflecting the number of users whose devices were spotted at a specific location after they were exposed to an ad. Seems pretty simple, right?

Well, as you may have guessed, there are a lot of details for marketers to consider the next time they see visits data in a campaign measurement report. Why be bogged down by the details? It’s important for smart mobile marketers to understand the nuances behind the numbers they rely on to make important business decisions because the ways in which measurement firms calculate and report visits can vary drastically. Knowing the basics of exposed visits is a great place to start.

The two ways visits are measured

The first thing to understand is that exposed visits can be reported in two distinct ways, and usually it’s up to your vendor to choose how it’s done. Some measurement providers report visits according to raw value, meaning they are measuring the actual number of devices exposed to an ad that have been tracked in campaign locations. Some think of raw value as a gold standard because it is a direct measurement of actual mobile devices that were spotted in a location; however, I’ll explain why it’s not necessarily the most representative measure of what’s actually happening in campaign locations.

Another popular way of reporting visits — through extrapolated exposed visits — is based on a more complex estimating process. Yet, while extrapolation arguably gives us a better representation of campaign results and true store foot traffic, methods can vary wildly from vendor to vendor.

The raw story

The definition of raw value visits is straightforward: It is the actual count of devices exposed to an ad that have been tracked subsequently in campaign locations. But let’s think about how mobile devices can be tracked.

In general, in order for a measurement provider to track a mobile device, it must have a mobile app installed that is sending off location information either through a location software development kit (SDK) or through its ad server, and that app must have been given permission to track GPS location data. Even though most apps require location tracking at the download stage, getting an accurate measure of the number of exposed devices that showed up in a particular location is limited by these technological constraints.

Ever see a campaign report showing a relatively small number of exposed visits? It was probably reporting visits based on raw value. The raw visit value can seem like an insignificant number because it typically represents only those devices using an app featuring a location SDK or serving ads, with GPS enabled; and thus, it under-represents the true location foot traffic. Just think: In real life, in most locations, the number of people on their phones with GPS enabled is a small subset of the actual number of people in that particular location.

So, if someone tells you that raw value-based visits are the most “real” visits number you can get, keep in mind that while this method directly tracks devices, it actually only directly tracks devices a system can see. In truth, even those raw value metrics may vary from vendor to vendor because each company is operating in its own data ecosystem among its own proprietary set of app partners.

Extrapolating the truth

While raw value visits are a direct measurement, extrapolated exposed visits give us an estimate of the physical foot traffic in a location based on the activity of mobile devices exposed to campaign ads. We get there by using a mathematical model, factoring in things like the typical number of people who visit a type of location, the amount of time they tend to spend there and other parameters. Ultimately, extrapolated exposed visits reflect a more representative estimate of the total number of exposed visits seen in a real-world location.

Measurement providers use their own mathematical models to estimate real-world visits using this method. Think of it as their own special sauce, though most are taking the same key criteria into consideration. First, they’re usually looking at the frequency by which mobile devices send out GPS location data through an SDK or ad network. While devices send out location data intermittently, some devices might send out those pings far more frequently than other devices. By adjusting for these variables, measurement providers might try to overcome biases that could arise if devices seen in locations are exceptionally active or inactive.

Then there’s dwell time — the amount of time a device typically is present in a given location. Vendors might consider the average amount of time a device is present in a fast food restaurant (say, around 20 or 30 minutes) as opposed to a car dealership (one to two hours), and factor that into their extrapolated exposed visits equation.

It’s also important to consider the likelihood someone might have their mobile device on or in use while in certain locations. People in a movie theater are less likely to have their phones on or active than while comparing prices at a consumer electronics store. But just because their phones are less likely to be emanating trackable data does not mean that people exposed to an ad for an Oscar-nominated film did not go to see that movie after seeing the ad.

All of these factors and more are modeled to estimate — or extrapolate — how many devices were actually in a location following ad exposure.

Creating transparency

Hopefully, now marketers reading this have a better understanding of a key mobile ad metric — exposed visits — and what’s behind it. As with much of the data marketers rely on every day, there are lots of complexities at work before mobile ad metrics show up in a campaign report.

As I’ve noted here before, my goal is for the mobile location data industry to be as transparent as possible, which means marketers must understand the metrics and hold their measurement providers accountable. Want to learn more? Look for my explanation of control groups and how they affect ad exposure and lift, coming here soon.

Opinions expressed in this article are those of the guest author and not necessarily MarTech Today. Staff authors are listed here.

About The Author

Gladys Kong, CEO of UberMedia, is an expert in mobile technology and data solutions. Gladys is dedicated to innovating and developing new ideas within technology startups. Since joining UberMedia as Chief Technology Officer (CTO) in 2012, Gladys has been responsible for taking UberMedia from social media app development company to a leading mobile advertising technology company and recruiting one of the best data science teams dedicated to consistently producing data solutions that anticipate and respond to today’s diverse marketplace. Gladys’s tenure in technology is extensive: She was CEO and co-founder at GO Interactive, a social gaming firm. Prior to that she was VP of Engineering at, and VP of R&D at Idealab, where she helped create numerous companies, including Evolution Robotics, Picasa, X1Technologies, and many more.

Let’s block ads! (Why?)

Mobile Marketing & Martech – MarTech Today

Leave a Reply

Your email address will not be published. Required fields are marked *