It’s no secret that app publishers want to know as much about their users as possible. It doesn’t matter whether it is for the sake of serving the most effective ads, or for developing a deeper understanding of user behavior. As a result, many have turned towards the blossoming field of location data.
Individual apps, however, will always have an incomplete understanding of how their users are moving through the world: no app can collect 24/7 data on an individual device because they would burn through a user's battery. And location data, in general, is very messy: just because a phone tells you that it’s at Point X, that does not mean it is actually at point X due to interference, prediction methodologies, and just plain, old error. The best way to effectively use location data is to get your hands on as much as you can, and then sift through the dirt to find the diamonds. To properly understand users in the real world, an app will have to augment their data with a broader ecosystem of location data.
The Targeted Solution
This is where Unacast comes in! Not only do we clean up all the noise so that you can get to insights faster - and cheaper - we can also provide additional insights about your users by attaching it to our entire pool of data. For example, we can tell you what your user is doing when they have your app closed.
Without the data that Unacast provides, you would have to pull in credit card data (an expensive endeavor) or try your hand at online cookie data (cheap, but limited to digital). As a result, most apps tend to approach targeting with a best-guess strategy, slowly attempting to narrow down the most effective ads via A/B testing campaigns.
From Theory to Reality
Say, for example, that you have a “Makeup and Beauty” App. Due to the nature of the app, you might assume that your users purchase more clothing, or maybe even go to the gym more than the average person. These assumptions however, are just guesses-- advertisements for lipsticks or trial memberships for gyms might gain some traction, but there’s nothing exact about the method. Guessing like this probably gets the app 60-70% of the way there, especially if your app has a highly specific or specialized user base, But, would you have guessed that your beauty app users are more likely to go to hotels? What about coffee shops?
As an experiment, we took a Makeup app that falls into the category of Health and Beauty, with approximately 300K users. We then augmented users who had this app with our entire ecosystem of apps to see where they spent their time. After comparing these users with our visit product, we were able to see what categories of stores these users spent their time at. Not only that but we were able to compare that to a larger population, in this experiment ~26M devices over the month of July, to test for significance.
From there it was a simple test of significance to show that the app users were ~20% more likely to go to Shopping Malls and 10% more likely to go to Gyms than the average person. And while we might have guessed that they are somewhere around 22% more likely to go to Beauty Salons, it is surprising to find that they are 15% more likely to go Coffee Shops and 16% more likely to go to Hotels.
What You Don't Know Can Hurt You
This simple and quick experiment shows that Location Data can reveal insights into your users that would otherwise be impossible. Coffee shops, in particular, are something that probably would have escaped even the laser-focused advertiser’s eyes, even if they were a data-guru, because ~34% of coffee purchases are paid for with cash. When a user is outside of their home, not using their credit card, and doesn’t have your app open(which is the case a significant portion of time), there is only one source that can help you reach the most accurate understanding of your users, location data.