Stay marketing and tech-savvy. Get the latest in martech - subscribe to MarTech Today.
How location data accuracy leads to stronger personalization
Inaccurate location data can be a headache for marketers. Contributor Amy King explains how you can ensure precise data, resulting in more personalized location data-based marketing campaigns.
Location intelligence reveals a great deal about who we are. By observing the stores consumers visit, their areas of residence, where they work and how far they commute, marketers gather valuable insights into consumer preferences and interests.
Ultimately, this observed, real-world location data is the bridge that links a person’s online activities with their online behavior.
Location data comes from a variety of sources, including IP (Internet Protocol) addresses, GPS (Global Positioning System) coordinates, WiFi signals, apps and beacon data. With US consumers using their mobile devices for over five hours each day on average and touching their cell phones an average of 2,617 times per day, according to data from dscout, our phones have become the most telling source of how we spend our time.
However, each of the above sources faces distinct challenges that influence the accuracy of the signals they provide (from physical barriers to inherent technology limitations), further demonstrated with Forrester reporting in Verve-funded research (PDF) that 34 percent of marketers cited inaccurate location data as a key marketing challenge.
With this, below are various data sources — and their related challenges — when it comes to defining location and ways in which marketers can ensure precise data and better-performing marketing initiatives.
1. IP & WiFi
An IP address is a unique string of numbers that’s linked to a person’s online activity and reveals the location of the device that is accessing the internet. Just as someone needs your mailing address to send you a letter, computers need an IP address to communicate with one another.
The challenge lies in the fact that IP addresses rotate, sometimes as often as every 30 days, which makes it difficult to match a device to a location over an extended period. For example, if a marketer is targeting a device with a specific IP address for a two-month campaign, a portion of the audience may disappear when the IP address changes.
Ultimately, this means that for the advertiser, using IP address data to determine location can result in their ad being shown to an audience in the wrong place at the wrong time, which wastes impressions and decreases their return on ad spend (ROAS).
2. GPS data
Mobile devices constantly send GPS coordinates via cell towers and WiFi hubs, meaning marketers should have constant insight into their audiences’ coordinates, right? Not so fast.
These devices need a clear line of sight to GPS satellites to send and receive information accurately. The signal reflects off physical structures, such as buildings or bridges, which may inaccurately place a device at point A when it’s actually at point B (often called the “Urban Canyon” effect).
Sending and receiving signals indoors is also a challenge, as the transmission degrades considerably as it passes through different barriers such as walls and windows.
For advertisers, relying on GPS signals in urban areas could mean a diminished ability to reach consumers due to physical barriers interrupting the path between satellite and device.
User-determined device settings and activities influence when signals are available for detection. These primarily include location services, background app refresh and the length of time an app has been open.
Ultimately, this makes it difficult for advertisers to capture their audience due to varying user settings:
- When it comes to location services, new mobile operating systems have removed the “always” or “never” options for tracking services. Now, users can change their device settings and choose “only while using” for each app, meaning location information is available only when they are physically using the app.
- If an app doesn’t have access to quickly refreshing location information, the data available will be stale and not a true representation of current movement. Background refresh is important for apps by enabling them to keep content current. However, background app refresh is a power-hungry feature that many users are opting out of.
So, what do all of these challenges mean? Marketers must find ways to work with providers to overcome the inherent limitations of each source, get access to a large volume of data across many sources and filter out “dirty data” to provide meaningful intelligence.
For marketers seeking more precise location data, I recommend seeking vendors that utilize the following:
A. Data volume
It comes down to the simple idea that more is better — in this instance, the “more” is data. Some companies utilize data provided by a panel for their analyses. This method requires a group of users to opt into an app or service, allowing marketers to observe their behavior and extrapolate insights to inform their decisions. This leads to insights that are directional and limited in scale.
Instead, data providers should be offering aggregated non-personally-identifiable data from the entire device population — over 1 billion unique devices per month. Instead of monitoring a small number of opted-in devices, this allows marketers to see every device, making for observations and analyses that are truly representative of how consumers move throughout their days.
B. Diversified sources
Diversified sources — such as apps that don’t offer advertising (i.e., navigation and subscription apps) or network beacons that provide signals when a device is in a retail location — provide validation and more complete data sets. To confirm location data points’ validity, multiple sources should be cross-checked. If more than two sources report a location data signal at the same coordinate, you will know with certainty that the point is accurate.
Additionally, more data sources deliver a larger inclusive sample and a comprehensive view of location behavior. Each source sees how a segment of devices appears to move over time; combining these views across sources deepens our knowledge of consumer behavior, including where consumers shop, how often and where else they spend their time.
C. Quality filters
The advertising ecosystem is saturated with fraudulent data — in fact, ad fraud costs the industry $8.2 billion per year, according to the Interactive Advertising Bureau (IAB). A thorough filter is critical toward removing bad information and protecting advertiser investment.
Sophisticated location providers have access to the same data sources and receive similar signals from device operating systems, bid exchanges and other partnerships. What separates great partners from good ones is their ability to synthesize large amounts of data from diversified sources, filter out fraudulent signals and interpret the remaining information in the context of a map.
When selecting a partner, it’s important to ask how they determine a device’s location, what methodologies they employ to remove bad data, if they have the scale to execute your program, and how they tie location signals to the real world.
With this, you’ll be well on your way to stronger, more personalized location data-based marketing campaigns.
Opinions expressed in this article are those of the guest author and not necessarily MarTech Today. Staff authors are listed here.