3 Tips to Protect Your Mobile Campaign From Inaccurate Location Data - Thinknear by Telenav | Location-based Mobile Advertising Solutions

3 Tips to Protect Your Mobile Campaign From Inaccurate Location Data

header-new-york-fraud-data

Mobile ad spending is growing, expected to hit $42 billion in 2018 in the U.S. alone. With that kind of growth, there are going to be publishers who outmaneuver the system—striving for financial gain by passing dubious data through ad exchanges, either knowingly or unknowingly. Since location-based mobile ads are especially valuable, location data can be a particularly hot commodity to target. In fact, in our most recent Location Score Index report, we found that 63% of ad requests did not contain location data accurate enough to effectively target a mobile user. Not only are we huge fans of making the industry better, but we do all we can in our own technology to make sure we are only using the most accurate location data.

So, how do we uncover publishers who are trying their hardest but are still coming up short? Or outsmart those just looking to make a quick buck? Unfortunately, it’s not quite as easy as outsmarting the so-called Rolex salesperson on the street; but it is just as possible. Simply knowing what inaccurate location data looks like can arm you with the knowledge to ask the right questions and ensure you’re truly getting what you paid for. Nothing could be worse than thinking you’re getting location data from people on the streets of New York City when, in actuality, your mobile ads are only being shown to people on the outskirts of New Jersey. In case you’re interested in all the things you should be asking of your mobile advertising partner, download our guide with answers to the 12 questions IAB recommends you ask your partner.

For now, here are three tips for identifying the defective data that could be lurking in your location-based mobile ads if you’re not asking the right questions.

Tip #1: Keep an eye out for IP-based data

App publishers passing inaccurate location data are not usually doing so maliciously. Sometimes the publisher simply doesn’t know the correct protocol. Whatever the reason, it still happens. IP-based data is some of the easiest to pass, because it requires no user interaction whatsoever. IP data is simply assigned by either the internet provider (for WiFi) or the cellular network provider (think AT&T or Verizon), and it can only be mapped to with a high degree of accuracy at a state-level, at best.

Luckily, this type of data is fairly easy to spot. IP-based data is often converted to a “centroid”, making it quickly apparent that it is inaccurate; but more on that later.

Tip #2: Look out for user registration information

Publishers will sometimes convert a user’s zip code—gathered at the time of registration—to location data. To do this, publishers will often convert the zip code location to a “centroid,” finding the center of the zip code and passing along those latitude and longitude coordinates as a user’s location.

For instance, if a user entered their Culver City ZIP code, 90232, the app would convert it to {34.019191, -118.39255} (the center of 90232) before sending it to the ad exchange. These latitude-longitude coordinates look to be very precise, reliable data. However, if the user is actually at Thinknear’s office in the same zip code, that latitude and longitude would be inaccurate by more than a mile, no longer counting as hyperlocal data. Since registration data is gathered only one time, it is possible for the data to be off by hundreds (or thousands) of miles if the user has since moved, entered the incorrect zip code, or simply is not currently in their zip code region. Additionally, registration data tells us nothing about the places a mobile user visits in their day-to-day life—data that is key for behavioral targeting.

img-map-fraud-blogpost-2

Tip #3: Be wary of cached locations

We consider using stale, cached location data to be bad. Yes, way worse than eating that rock-hard Starburst you found under your couch cushion last night—at least the wrapper was still intact, right? Wrapper or no wrapper, cached data is old. The duration between when the data was “pulled” and then sent can range from a couple hours to days, weeks, or even months.

Ad targeting should be based on only the most recent of data—freshly plucked from the mobile device that very second. Unfortunately, the current structure of location data does not account for freshness, which is part of the reason why we created the Location Score, so we can verify the quality of the data packets.

Over the past year, we’ve found that the quality of location data from app publishers has remained, largely, unchanged. Be sure to download our most recent Location Score Index report to discover why location data hasn’t improved and what location data in the ad exchange ecosystem currently looks like.

While data quality begins to improve over the next few years, you need to rely on a trusted data partner to ensure the quality of the location data your campaign is using. Inaccurate data can take any of these forms, and is the reason our Location Score platform sorts hundreds of billions of data sets every month to identify only the most accurate data. If the data isn't accurate, we throw it out...and so should you!

Contact us. Let's create magic together.

Our Newsletter is good. Sign up so we can deliver the goods. (Not bad, huh?)

Request a call