Data scientists are using the most annoying feature on your phones to save lives in Ukraine

The numbers discussed were grim. The United Nations had just declared Ukraine was facing the biggest humanitarian crisis to hit Europe since World War II as more than 4 million Ukrainians—roughly 10% of the population—had been forced to flee their homes to evade Russian President Vladimir Putin’s deadly and indiscriminate bombing campaign. That total has since swelled to 5.5 million, the UN estimates.

What the aid specialists on the call wanted to figure out was how many Ukrainian refugees still remained in the country (a population known as “internally displaced people”) and how many had crossed borders to seek asylum in the neighboring European Union countries of Poland, Slovakia, and Hungary, or south into Moldova. 

Key to an effective humanitarian response of this magnitude is getting accurate and timely data on the flow of displaced people traveling from a Point A danger zone to a Point B safe space. And nobody on the call, which was organized by CrisisReady, an A-team of policy experts and humanitarian emergency responders, had anything close to precise numbers.

But they did have a kind of secret weapon: mobility data.

“The importance of mobility data is often overstated,” Rohini Sampoornam Swaminathan, a crisis specialist at Unicef, told her colleagues on the call. Such anonymized data—pulled from social media feeds, geolocation apps like Google Maps, cell phone towers and the like—may not give the precise picture of what’s happening on the ground in a moment of extreme crisis, “but it’s valuable” as it can fill in points on a map. ”It’s important,” she added, “to get a picture for where people are moving, especially in the first days.”

Ukraine, a nation of relatively tech-savvy social media devotees and mobile phone users, is rich in mobility data, and that’s profoundly shaped the way the world sees and interprets the deadly conflict. The CrisisReady group believes the data has an even higher calling—that it can save lives.

Since the first days of Putin’s bombing campaign, various international teams have been tapping publicly available mobility data to map the refugee crisis and coordinate an effective response. They believe the data can reveal where war-torn Ukrainians are now, and even where they’re heading. In the right hands, the data can provide local authorities the intel they need to get essential aid—medical care, food, and shelter—to the right place at the right time.

‘The Big Bang moment for mobility data’

Thanks to the smartphones in our pockets, our Wi-Fi connections, and society’s addiction to social media, each one of us creates a vast stream of mobility data, or information that shows our movement from one location to another. Often when we opt onto a website or app we reveal—intentionally or not—our location through our connected devices.

Social networks, app providers, mobile phone companies, and adtech firms alike collect vast amounts of mobility data, often to serve up location-specific ads to potential consumers. If you’ve ever received an in-app advertisement on your mobile phone directing you to a nearby McDonald’s or Starbucks to buy a discounted Big Mac or Frappuccino, chances are the advertisers were using some form of geo-targeted mobility data. Whether you regard this as an annoyance or not, it’s become a huge business, with some projecting the global market for such targeted ads to top $300 billion in the next four years.

There’s a far less benign side to the technology, too. Big Tech has gotten into plenty of hot water for creating highly effective ways to track our comings and goings through websites, browsers, and apps, triggering pushback by privacy watchdogs on both sides of the Atlantic, urging them to cut out the 24/7 surveillance of our lives.

As a result of the war in Ukraine—and, before it, during the COVID outbreak—there’s now a third side of the argument. Crisis experts say mobility data has the power to play a big role in coordinating relief in a humanitarian crisis.

Andrew Schroeder, vice president of research and analysis for Direct Relief, a CrisisReady partner, has been using mobility data in recent years to address public health issues. For example, he and the data science team at Direct Relief and Harvard University’s School of Public Health used mobility data to map a cholera outbreak in Mozambique in 2019, and they used similar data earlier this year to show how and where winter-storm–related power outages adversely impacted some of America’s most vulnerable Medicare patients.

“The Big Bang moment for mobility data was the pandemic, from March 2020 to June 2020 when nobody knew what the hell was going on,” Schroeder says.

In the early days of the COVID outbreak, policymakers around the world enforced strict social distancing rules in hopes of keeping the coronavirus somewhat contained. Public health officials and researchers wanted to know if newly enacted rules designed to limit human contact were doing enough to suppress infection rates. Using publicly available mobility data, researchers mapped the disease vectors—in essence, the limited movements of you and me during that period—to visualize how people were moving about in a community. That data point could then be matched with localized infection data, for example, to determine if the public health measures were working, or if they needed to be tweaked to include, say, more buses on roads to thin out the crowds during rush hours, or obligatory-masking rules in public places. Incidentally, a year later, all manner of investment banks turned to mobility data to demonstrate that the reopening of the world’s biggest economies was in full swing.

Mapping the crisis

When Russia invaded Ukraine, the CrisisReady team and others turned to mobility data again in the hopes of building maps of refugee flows. Schroeder says they’ve sent their reports to international agencies such as Unicef, “which is making decisions on support for educational services to refugee children,” and to the World Bank “to help inform understanding of municipal investments for refugees.”

Schroeder shared with Fortune a map the group provided to the World Bank. In it, you can see where swells of displaced Ukrainians have temporarily resettled since the beginning of the war. The map, which uses phone-based location data supplied by Facebook, reinforces some general assumptions. For example, Poland’s biggest cities of Warsaw, Kraków, and Gdańsk as well as towns on the country’s southeast border with Ukraine are refugee hotspots, as shown by the dark green blotches. But the map also shows other migratory patterns that are less intuitive, such as the Polish communities that have depopulated since the start of war (see the beige blocks).

Using Facebook-supplied mobile phone data, the CrisisReady team was able to map changes in population density in the cities and towns marked on the map above. What’s important to aid workers on the ground though is where the displaced persons will go next. That’s where a second data set, also furnished by Facebook parent Meta, comes in. It’s called the Social Connectedness Index.

Launched two years ago, the index measures the strength of friendships among your Facebook contacts, and then maps those links geographically—between, say, users in New York and San Francisco or Kraków and Kyiv. The Social Connectedness Index, which contains fully anonymized data sets, has become a popular data source for researchers, including CrisisReady, interested in mapping crises, but other forms of mobility data are also being adapted in the Ukraine refugee crisis, including anonymized usage data from mobile phone companies.

The Social Connectedness Index helped the CrisisReady team model out a rough idea for how many Poland-based Ukrainian refugees—those, for instance, who happened to be Facebook users and were located in a certain geographic cluster on the map; say, Gdańsk—might move on again to another city or town, and where that might be.

“If you originated in Ukraine, where are you most likely to be connected, in terms of Facebook friends? Does that have a special pattern to it? It turns out, it really does,” Schroeder told ReliefWeb last month. “We can fill in things that you can’t find out by looking at the border patrol report.”

Problems during wartime

Data snapshots that show migratory patterns have become a vital tool to fight fast-spreading diseases, but so far applying it to wartime situations has been complicated. Countries that prefer to keep humanitarian groups in the dark during times of conflict may not want this kind of mobility data shared with the public, experts say.

And, sure enough, the list of countries not participating in Meta’s Social Connectedness Index reads like an almanac of countries at war. They include, according to Meta: Afghanistan, Western Sahara, China, Cuba, Iraq, Israel, Iran, North Korea, Russia, Syria, Somalia, South Sudan, Sudan, Venezuela, Yemen, Crimea, Jammu and Kashmir, Donetsk, Luhansk, Sevastopol, West Bank, and Gaza.

As Schroeder points out, researchers and civil society organizations still have a lot of work to do to persuade tech companies and countries to open up their mobility data vaults to help in times of crisis. (The anonymous nature of this kind mobility data means no single end user can be identified, and so it’s become—so far—of little to no concern for data privacy advocates.)

Despite the obstacles, Schroeder sees the war in Ukraine as an important proving ground for the potential of mobility data to save lives.

“Right at the Feb. 24 mark, when the invasion happened, day one, we could do an analysis using data from Meta,” Schroeder says. “It has its limits. Lots of other data would give a more rich picture, but we were able to get a valuable signal instantly.”

Each week, Fortune covers the world of innovation in Breakthrough. You can read previous Breakthrough columns here.