Location Accuracy’s Magic Number: 93 Feet

That’s about the length of seven Volkswagen Beetles lined up by a curb. But for PlaceIQ, the level of geo-data accuracy isn’t the most important thing — it’s how marketers use it.

As the use of geofencing and proximity marketing play a more integral part in omnichannel marketing, the level of accuracy that geo-data can provide has been subject to a great deal more scrutiny.

Seeking to put the state of location accuracy into proper perspective, geo-data specialist PlaceIQ worked with offline search engine Findyr to examine 150 physical locations across five major U.S. cities. On average, the duo found that location data obtained via mobile smartphones is accurate up to 30 meters. (Download the report here; registration required)

That’s roughly 93 feet. To visualize that, PlaceIQ says to imagine seven Volkswagen Beetles parked bumper to bumper. That’s the average range that location data placed a mobile device it attempted to reach in.

Layers of Intelligence

In other words, location data isn’t perfect, but it’s fairly close. However, if a McDonald’s was right next door to a Starbucks, the geo-data alone would not be able to definitively say which one a consumer with a mobile device was standing in front of.

As such, Steve Milton, PlaceIQ’s CTO, suggests that marketers need to view geo-data as offering valuable insights alongside other data sets. Aside from credit card data from stores, location data is the most privacy-sensitive way of knowing whether someone is a near a physical shop or not.

“Location data can paint a vastly different picture depending on the situation,” says Milton. “Think of location as the sum of multiple parts and circumstances, each bringing different layers of clarity.”

The analysis in PlaceIQ’s report illustrates those levels of clarity and insight. Taking into account building height, composition, and density — things that marketers and data gatherers obviously can’t control — all contribute to the ability of mobile devices to emit an accurate signal to generate location, Milton notes.

City By City

Here’s a breakdown of the cities analyzed in PlaceIQ’s report, Location Data Accuracy Revealed, show:

The average finding for location data accuracy varied in each of the five U.S. cities:

  • Boston, MA – 21 meters
  • New York City, NY – 27 meters
  • Austin, TX – 28 meters
  • Washington, D.C. – 29 meters
  • Chicago, IL – 38 meters

Again, it’s worth pointing out that these findings reflect the complicated connection between multiple factors that affect location data accuracy, such as signal source (GPS signals,wifi, cell tower triangulation), environment (area density, skyline view, indoor or outdoor location), and personal use (location data access enabled, type of mobile app used, operating system usage).

“Cities like Chicago and New York are both very dense and offer opportunities for devices to connect to wifi signals, which aids in determining accuracy,” Milton says. “What they also have however, are tall buildings that prevent quick connections to GPS signals, which traditionally provide the most accurate location data.”

A common example of these inconsistencies are ride-hailing apps that determine location less effectively in different situations, Milton points out.

“A mixture of physical infrastructure and wifi density are the key for location accuracy, so it makes sense that Boston — a city with the best combination of building density for allowing for Wi-Fi access, without the abundance of skyscrapers that would inhibit GPS signal reception — would perform the best in this test,” Milton says.

Going The Dynamic Distance

In conjunction with the study, PlaceIQ is introducing a tool meant to provide a deeper understanding of location accuracy: Dynamic Distancing.

PlaceIQ’s promise is that it can “dynamically” account for variability in location signals when determining visitation, which should give brands a greater sense of just how clear the picture of a consumer’s mobile behavior is when they go from place to place.

This technology uses a three stage process to attribute a device’s presence to a venue, both instantaneously for ad serving and retroactively for post campaign behavioral analysis and media measurement, says PlaceIQ CEO and co-founder Duncan McCall.

When post-processing data for behavioral analysis or visitation measurement, Dynamic Distancing evaluates streams of movement information to identify when a visit begins to take place. As a device slows and its locations cluster, Dynamic Distancing determines the outer boundary of the visit based on the accuracy of the data coming in and the size of the cluster being formed by the device’s movement. Then, Dynamic Distancing evaluates all mapping information contained within the computed boundary to determine the most likely venue being visited.

PlaceIQ’s newest product offering follows the evolution of its primary location data intelligence tool, the Enterprise Place Visit Rate, as well as its expanded work with auto manufacturer Audi, existing agency partner Starcom MediaVest Group and other allies like Acxiom. It’s all part of a strategy that emerged years before the use of smartphones and location-based advertising were such an intrinsic part of mainstream marketing.

“Location data is a critical part of every successful marketer’s arsenal,” says McCall, CEO at PlaceIQ. “It continues to act as a versatile resource for both informing marketing decisions across verticals and as a horizontal enabler of broader business decisions at every level. This report’s goal is to educate, cut through the marketing hype, and demystify the technology that brands are using to better understand, engage with, and drive sales. Only with a foundation grounded in truth and accurate data can the broader location industry thrive. The world’s most innovative brands and media agencies continue to turn to PlaceIQ’s innovations in location accuracy for this reason.”

Darwinian Approach To Location Evolution

In a conversation with GeoMarketing, McCall noted six-year-old PlaceIQ was “doing location before it was fashionable.” Referring to the past two years that have seen the use of geo-data grow in tandem with the rise in mobile ad spending. (According to eMarketer, mobile ad dollars have gone from $19 billion in 2014 to $42 billion this year, comprising just over half of all digital marketing spend.)

And as we’ve reported previously, geotargeted mobile ad revenues in the U.S. will grow from $6.8 billion in 2015 to $18.2 billion in 2019, according to BIA/Kelsey.

All of this activity, driven by the wide accessibility of real-time data, is still relatively new to marketers. In McCall’s view, that means the industry has only just begun to explore the possibilities geo-data is creating.

“It’s probably at some peak of the height cycle right now, which is hopefully good news, because now the real utility starts to come forward,” McCall says. “Initially, the view of using location to power marketing was that it wasn’t fashionable, and was considered by some to be very tacky. That’s changed, of course.”

For one thing, the sophistication of both consumers and marketers’ use of location has started to evolve rapidly.

“The use of location has gone from, ‘Well, I can just geofence my restaurants and add 4-mile accuracy, and that’s better than what I’m doing now’ to ‘Hey, I want to use location to understand visitation from advertising, so I can make really important business decisions,’” McCall says. “Location is now used to customize the creative, to analyze competitive behavior, to chart the path of consumer journeys to make incredibly important strategic business decisions across the organization.”

So with the proliferation of location-driven marketing, a bigger issue is starting to emerge as brands and agencies try to sort out the highest accuracy from the lowest.

To help separate the good geo-data from the bad, PlaceIQ rolling out a product dubbed Darwin. The new tool promises to detect and eliminate fraudulent location data signals. It does this by analyzing the location quality of any combination of location history data, including ad request logs, application location histories, sensor data archives, and more.

As McCall puts it, “Darwin allows brands to focus on legitimate data instead of false foot traffic.”

“What we’re trying to do is provide a platform and a path that allows our clients to take this journey towards this more sophisticated future,” McCall says. “There are new products we’ll be announcing along the way. If we can make our clients successful, we’re going to make our company successful. That’s the philosophy that we try and march forward with.”

About The Author
David Kaplan David Kaplan @davidakaplan

A New York City-based journalist for over 20 years, David Kaplan is managing editor of A former editor and reporter at AdExchanger, paidContent, Adweek and MediaPost.