Many advertisers require a user’s latitude and longitude coordinates (lat/long) be sent with each ad request to them so they can target users in specific geographic locations or who are near certain stores or landmarks. The importance of this location data has allowed advertisers to target at a very granular level, but how accurate is the data being sent to the advertisers and is it reliable?
Publishers know that advertisers want location data, so they do what they can to provide it, even when it is not the most accurate. This can lead to advertising campaigns that are not as successful as they could be because advertisers are not targeting users where they think they are located. That begs the obvious question of what is considered accurate and reliable location data?
The most reliable location data is GPS lat/long sent directly from the mobile device. This is the most accurate and is what advertisers want as often as possible. Other forms of location data are either derived or provided from other means which are not as accurate. Below are some of the main sources and causes of inaccurate location data.
When GPS lat/long is not available, the most common method to provide it is to derive it from the IP address. This method can be accurate but there are a lot of potential pitfalls with it that can lessen it’s accuracy. Where is the IP coming from that is being used to derive lat/long from? Is it a proxy URL? If that is the case the user could be in another country or on the other side of the world from where the IP is defined. The IP could be one for a cell tower in which case it could be within the same city but still not very accurate when an advertiser is trying to target a user located outside a specific store. All of these scenarios can lead to varying degrees of inaccuracy.
Another common cause of inaccurate location data is caching. How often is the location data updated? Sometimes an app will pull the location data when the app is first started and not update it, but just continue to send the same location data over and over with each request. There is a good chance that the user has moved locations from when they first turned on the app, but since the cached location data is always sent the advertiser never knows and the publisher could miss out on potential ad opportunities.
The last common cause of inaccurate location data is a lazy approach used by publishers. Sometimes they will send the location data that a user provides when they sign up for an app or website. Some apps or sites require users to register and this process may require them to enter a zip code. The publisher then derives the lat/long from the zip code. As you can guess, this can become extremely inaccurate over time, particularly as the user moves about with their mobile device. Thankfully, this approach is not used very often.
In order to combat the inaccuracy of location data, some advertisers now require publishers to provide information on the source of the data and where it came from, such as: GPS, derived from IP, user supplied, etc. A publisher may think inaccurate location data is not impacting them if they have a good fill rate, but if advertisers are targeting users in a specific location they are expecting those users to convert at a certain rate, and if the user is not really near the location that their location data says they are then they will most likely have a low conversion rate. Advertisers will notice this trend over time and they will stop filling as many ads for that particular publisher and eventually their fill rate will suffer.
Location data is very important to advertisers and it only benefits a publisher to provide as accurate information as possible. The more accurate location data a publisher can provide the greater the amount of ad opportunities they may be eligible to display and thus increase their revenue.