Everyone knows what gentrification looks like.
Community gathering places are replaced by boutiques and bank branches. Corner stores are made obsolete by VC-funded vending machine startups. Blocks that spent decades in a state of disrepair sprout Michelin-starred bistros and cocktail bars—some even trading on the allure of formerly stigmatized neighborhoods with Instagram-ready touches like fake bullet holes and 40-ounce bottles of rosé wine.
These new businesses don’t just reflect gentrification, of course. They actively drive it. Take New York City’s SoHo neighborhood: sociologist Sharon Zukin has described how art galleries and live-work spaces displaced industry from Lower Manhattan and paved the way for luxury residences for non-artists. Or Dumbo: in the late 1970s, real estate developer David Walentas came up for the idea of redeveloping a quiet corner of the Brooklyn waterfront while eating in the area’s lone gourmet restaurant.
In recent years, however, technology has accelerated this dynamic. Upscale consumption spaces have always been engines of gentrification. What’s new is how these spaces increasingly attract customers beyond their immediate surroundings: through “location-based services” (LBS), a genre of smartphone and web applications that filter digital information about the world through a user’s spatial coordinates.
Apps like Yelp and Foursquare are not just tools for finding cafes and restaurants in gentrifying neighborhoods, in other words. By making new establishments more visible to upper-income patrons in other parts of the city, LBS are transforming those neighborhoods in very direct ways.
Everyone’s a Critic
Cities have evolved in tandem with text and image-based information systems designed to help overwhelmed residents choose among competing options for consumption. To take one example, annual publications of business information like the Yellow Pages have existed since the 1880s. But digital technology adds unprecedented speed and precision to this process. Today, companies like Foursquare, Yelp, and TripAdvisor offer continuously updated lists of businesses and attractions with extraordinary specificity.
You can find Polish restaurants that take American Express, dog-friendly hotels near Lake Tahoe, and bilingual Mandarin-English notaries. By speeding up the flow of information, LBS turn cities into what mainstream economists consider more “efficient” markets, tightening the connections between critical evaluation and real estate value in a process that can help accelerate gentrification.
Imagine moving to New York City or London in 1970 and deciding where to spend your free time. Most of your knowledge of the urban scene would come from personal experience, conversations with friends, family, and coworkers, or via articles, reviews, and ads in newspapers or magazines like New York or Time Out. If a new restaurant or art gallery opened elsewhere in the city, your likelihood of finding out about it would be limited by the publication schedule and capacity of the local media; many smaller establishments would never be “newsworthy.”
This is precisely the problem that Manhattan lawyers Tim and Nina Zagat wanted to solve in the late 1970s, when they asked their friends and colleagues in New York’s rapidly expanding professional services sector to rate various restaurants in the city. After being passed around among thousands of Citibank employees like samizdat literature for yuppies, the Zagats turned their survey into a wildly popular retail guidebook, ultimately expanding to over seventy cities. Rather than using professional dining critics, the Zagat Survey drew its judgments from hundreds of thousands of individual diners. This amateur contributor pool fed data into the company, which synthesized it into statistical averages that they sold in pocket-sized books by the millions.
Zagat’s crowdsourcing model represented a major innovation for urban information systems—but it would take digital technology to realize its full potential.
The Location Layer
In 2009, serial entrepreneur Dennis Crowley founded one of the most popular contemporary LBS: Foursquare. Early in his career, Crowley had worked as a software developer for a company that licensed Zagat’s reviews for PalmPilot mobile devices in the first dot-com boom, and he sold his first company, an early LBS prototype based on text messages called Dodgeball, to Google in 2005.
Working with data from Zagat and other publishers gave Crowley an idea. He saw the potential for dynamically updated reviews from mobile devices, contributed by users distributed throughout the city. While Zagat crowdsourced data from diners once a year, Crowley would let them weigh in on their phones in real time.
With smartphones, a new level of scale became possible. The number of users contributing data could be far greater than anything imagined by Tim and Nina Zagat—and the sheer volume of that data could be greater as well. Mobile devices also added a spatial element: users could constantly broadcast their location, enabling more geographically specific interactions.
The current culmination of Crowley’s efforts is Foursquare City Guide, a mobile app that provides personalized recommendations based on a user’s location and past interests. These recommendations are themselves sourced from other users, in the form of venue ratings and reviews. The company touts its use of machine learning to sort through all this information and match users with venues. Foursquare might recommend a particular coffee shop to you not just because it’s nearby and highly rated, in other words, but because the algorithm believes you will like it—based on your demographic profile, your clickstream of past likes, the time of day, and so on.
The scale is impressive. Formerly, Foursquare enabled users to “check in” to a particular location—essentially telling the app, “I am here.” Since Foursquare’s launch in 2009, users have performed 12 billion of these check-ins across over 8 million unique locations. But the company no longer needs users to announce their location—Foursquare has the technology to track users through GPS coordinates, cell towers, and Wi-Fi signals. Simply by walking around with a smartphone, the Foursquare user makes the company’s multidimensional map of urban space incrementally more useful.
The ultimate goal, Crowley says, is to create the “location layer of the internet.” He wants to “crawl the world with people in the same way that Google crawls web pages with machines.” Just as Google crawls the web in order to organize it for use, Foursquare crawls the city in order to organize it for consumption.
Serfs Up
But there’s a crucial distinction: whereas Google’s crawling is automated, Foursquare’s is performed by actual humans—a fact that LBS companies tend to downplay. A lot of the hype around urban information systems centers on “artificial intelligence”: the ability for companies like Foursquare to seamlessly guide you to relevant places and products. But as danah boyd and Kate Crawford, among others, have pointed out regarding “big data” more broadly, these apps actually derive their value from a vast stockpile of human labor.
And, as with any labor pool, some workers are more valuable than others. Certain Foursquare users are more passive in their interactions with the app; others actively contribute ratings and reviews, and check in to new businesses using Foursquare’s “lifelogging” app Swarm to keep the company’s data current. According to Foursquare’s own promotional material, this population of “superusers” is only 43,000. But this relatively small cohort, performing what Tiziana Terranova calls the “free labor” undergirding the online economy, produces a large portion of the company’s value.
The aggregate product of all this labor is extremely valuable for Foursquare’s advertisers and investors, at least in theory. The company is reportedly on track to hit $100 million in annual revenue in 2018, and its 2016 fundraising round gave it a valuation of $325 million. By industry standards, this isn’t even particularly high: its rival Yelp, which is publicly traded, has a market capitalization of over $3.5 billion. Google purchased Zagat in 2011 for $151 million, after infamously failing to acquire Yelp, and both Apple and Facebook have their own proprietary ratings of real-world businesses.
Schrödinger’s Latte
Power users and reviewers don’t just create value for companies like Foursquare, however. Their labor also enriches the developers, brokers, landlords, and small business owners that profit from the gentrification of urban space. That’s because LBS don’t just measure and map the city—they transform it.
By making it easier for urbanites to search for gluten-free pasta or private room karaoke, businesses can draw an audience from farther afield, especially in conjunction with the rise of ridesharing services like Lyft and Uber that efficiently ferry well-heeled patrons around the city. And by opening up a “long tail” of establishments for review, LBS enable even the humblest coffee shops to have thousands of reviews.
If one of these coffee shops becomes popular, it can have significant consequences for surrounding real estate values. In a capitalist real estate system in which property values reflect both scarcity and desirability, changes in a neighborhood’s amenities are quickly reflected in rising housing prices. Real estate brokers understand this dynamic, and actively cultivate it. In places like Harlem, they have even opened their own cafes in order to expand the gentrification frontier to areas where they have listings.
Brokers often enlist LBS directly to sell particular neighborhoods. Online real estate listings include “Walk Score” ratings calculating nearby amenities using LBS data. Real estate platforms like Trulia directly incorporate Yelp data into their listings pages. These digital urban information systems, then, don’t just gentrify neighborhoods by funneling customers to upscale establishments. They also provide a valuable source of data to the brokers who hope to convert those customers into residents.
Brunching While Rome Burns
Crawling the city is lucrative for technology companies, the real estate industry, and other beneficiaries of gentrification. But what happens when the users who crawl the city can no longer afford to live there? What happens when services like Foursquare and Yelp lose their labor supply?
The long-term profitability of these companies is threatened by the growing impoverishment of their key user base: urban millennials. Squeezed by stagnant wage growth and sky-high housing prices, both lambasted for their thrift and shamed for small indulgences like avocado toast, this cohort’s ability to generate new data and justify advertising campaigns is crucial to the location data economy. And while LBS founders and investors skew predictably straight, white, and male, prolific Yelp users in cities like New York and San Francisco are more likely to identify as female and Asian-American, facing a double bind of discrimination despite educational credentials well above average.
By willingly serving up ratings, reviews, and brunch photos to brutally efficient urban housing markets, millennials are turning prestige into price, collaborating in making their cities uninhabitable. A recent Twitter ad for Foursquare Swarm aims to appeal to users who are “constantly traveling.” To a user who responded, “I wish I could afford to travel,” the company replied, “exploring can be as close to home as walking a new way to work or checking out a neighborhood nearby!”
But even this attenuated form of travel may no longer be feasible in the near future. This is the final irony of LBS: the users who sustain these platforms are working hard to evict themselves from the cities they’re mapping for free. Just as their labor helped fuel the market dynamics that pushed people out of gentrifying neighborhoods, they will themselves be pushed out as values climb higher. The volunteer crawlers may eventually crawl themselves right out of town.