SXSW 2023: Making maps and location tech is really hard, but not if we do it together
On March 12, 2023, in Austin, Texas, mapping experts from the likes of Meta, Lyft and TomTom, working at the heart of this industry, gathered on the Industries Expo – Next Stage at SXSW to discuss the future of mapmaking and location-powered tech. If there’s anything we need to take away from this talk it’s that working with maps and location data is hard, whatever part of the industry you operate in. But it doesn’t need to be, if we work together.
There are two sides to working with location data: the software side and the data side, Marc Prioleau, Head of Location Partnerships at Meta, says. The software is the easier part, it’s the data that’s hard.
“Maps are a digital representation of the physical world, and it has this annoying habit of changing; that’s problem one,” Prioleau says. “Problem two is, y’all want more and more out of your maps. 10 years ago, expectations were low, now your expectations are skyrocketing.”
There’s more nuance to it than that, though; creating or sourcing data is not the difficult part. There are sensors and people generating this data everywhere — cars, phones, roadside traffic monitors, user feedback, they all send data to inform the map. According to Mike Harrell, SVP Engineering Maps at TomTom, there’s “an incredible amount of data, a tsunami of data.” In fact, there are some 6 billion plus data points gathered every day.
The difficult part is taking that data and turning it into something useful and meaningful.From left to right: Corinne Vigreux, TomTom CMO and Co-founder; Benjamin Schrom, Director of Product Management - Mapping at Lyft; Michael Harrell, VP Engineering Maps at TomTom; and Marc Prioleau, Head of Location Partnerships at Meta.
Maps are globally pervasive
Over the past few decades there’s been huge growth in location-aware tech. In the early days of digital maps, there was one core use case: turn-by-turn navigation. But now, maps and location tech are used in hundreds of commercial applications, including online dating, food delivery, ride hailing, shared mobility, motor insurance and fleet asset management. There are use cases for the common good that rely on accurate maps too, things like climate change monitoring, sea level rise monitoring, traffic management and urban planning.
Want to learn more about the panel discussion? Check it out.
In modern use cases, map data isn’t just about routing, it’s about being able to make smart decisions and provide useful insights. When the map is not good, those insights become close to useless.
“There is a layer of fact, the map data. But almost every map experience that you build on top of it [map data], all the things we need to do to deliver trustworthy rides to our customers, is opinion,” Benjamin Schrom, Director of Product Management Mapping at Lyft, says.
“You need to form an opinion about the best driver/rider match, an opinion about the best route or an opinion about the right price. Those are things that your modelling does, but all those opinions become terrible if the underlying map data is terrible,” he adds.
The accuracy, freshness and relevance of map data is constantly under the microscope for people like Prioleau and Schrom, because it’s become the crucial ingredient for their products and services.
Corinne VigreuxTomTom CMO and Co-founder
“It pervades everything we do, even if users don’t realize it,” Schrom says. “Every promise we make to our customers is informed by the map.”
Maps are taking on new meaning and value
Maps are becoming so central to modern products and services they’re taking on entirely new meaning.
“The interesting thing with ride sharing, and many other apps now, is the map becomes the organizing feature, it becomes the thing people use as the basis of understanding what’s going to happen,” he adds.
With apps today, people are pressing buttons and having the world orient around them. The map is central to how they make sense of the world, it’s not something they use simply to get from A to B. As Prioleau says, maps are becoming a central organizing factor in people’s lives, and that’s what’s moving mapping and the underlying data to the forefront of tech innovation.
Marc PrioleauHead of Location Partnerships at Meta
The challenge of building with map data
The challenge of building good location-based tech is not about having enough map data. The challenge is taking the mass of location data and turning it into something meaningful and useful.
Location data is everywhere, but it’s inconsistent, sometimes messy, comes in many different formats and is constantly changing, Harrell says. Similar sensor types that make similar observations might all communicate their data in different formats. They may all log the same place attribute, but they might all tag it in different ways and in reference to a different base map, even though it’s the same thing — creating five formats and five entries for the same real-world thing.
“The industry has been stuck because of this. All devices, as they communicate their location information, they do it in a different way, to a different map. There’s no consolidated map that we all refer to,” Harrell says. “Your ability to innovate and advance in the space was held back because you were always trying to advance already solved problems.”
Companies and developers of location tech spend a lot of time, and money, resolving these kinds of differences and inconsistencies between geospatial data and base maps — so that they have a singular reference of truth. Everyone that builds with location data and wants to make apps like Lyft and Meta do, has to become a mapmaker to some degree. But, in a perfect world, that shouldn’t have to be the case. So, what do we do?
“The world needs a better map,” Harrell says. The world needs a single record of truth about the world.
Improving the world's map
If location-tech companies are all spending time solving the same problems over and over, that becomes a problem in itself. One which is inhibiting the development of features, service improvements and new location tech.
But it’s a problem that can be solved with an industry shift and refocus.
Today, companies in the location-tech industry differentiate themselves mostly in the higher levels of map data. That’s to say the advanced proprietary data and services built on top of basic forms of map data. Even so, they still spend a lot of time ‘fixing’ low level data and their base maps to ensure their higher-level systems work. But that work doesn’t add value and in most cases is not a differentiating factor to other companies in their part of the industry.
All this could be solved if, at a very basic level, the location tech industry oriented itself around one point of reference, one map.
This is easier said than done. But what needs to be done is clear — there’s only one way the commercial world is going to solve the above problem and that’s if it works together to collectively agree on base level map data and standards to ensure location-tech systems works together from the ground up. This is where the Overture Foundation comes in.
“The collaboration (Overture Foundation) is to create a digital representation of the entire world of its most basic features, road network, POIs, building features and admin areas,” Harrell says. “If you create that base understanding, everyone else can communicate on that same base and interact and innovate.”
Or as Prioleau says, “The map needs to be a common map.” The industry needs a new era of collaboration.
If you want to find out how the industry might achieve this and learn more about what it takes to build the map of the future, you’ll be able to watch the panel discussion in full soon over on the SXSW website.In fact, go and watch it, because what I’ve shared above just scratches the surface on what the future holds for mapmaking, geospatial data and location tech.Editor’s note: Quotes have been edited for clarity and brevity.
People also read
The world needs a better map: TomTom is making it with its new Orbis Maps and ecosystem
TomTom, Amazon Web Services, Meta and Microsoft join forces and launch open map data foundation
More than 20 years of mobile mapping, here’s how it began
Meet the devs that coded their way out of the coronavirus chip shortage