Google Flu Trends

Picture of Sophia Petrillo's frowning face from The Golden Girls.
Picture it! San Francisco, 2008
Google Chrome hadn’t yet come out, Web 2.0 was still a thing. Android was brand new.

Disease outbreaks are a major global threat.
Can technology provide an early warning, and facilitate speedier intervention?

Alt

Google.org’s Predict & Prevent team was focused on global threats from social and climate/ecological changes, initially disease outbreaks. I joined after several engineers took on the challenge of using aggregated search patterns to estimate the level of influenza in the population. Their efforts resulted in a novel technology that produces good estimates of official flu activity indicators in dozens of countries—and later, dengue activity.

Alt

Google Flu Trends has been cited, reported on, and analyzed tens of thousands of times by researchers, public health officials, media, computer scientists, and designers around the world. It was the first major demonstration of how anonymous aggregated patterns of routine online behavior could be used for a public health purpose like this.

Unfortunately, a lesson of the Covid-19 pandemic is that early warning isn’t enough. Society needs to be willing to respond.

Alt

Since flu is seasonal, what’s notable is how different the flu season is this year compared to previous years. The best view on that is to stack or layer the seasons—in this case, they recede into the past with lighter, thinner lines, like faded memories.

Alt

As usual, the design challenge was to do it in the Google way: Make something reasonably universal, in a language highly constrained by the Google voice.

Our website (looking dated here) was intended to serve as an adjunct for public health agencies (they often cited it), and to engage ordinary citizens in awareness of flu activity. Part of an effective public health response—as we’ve seen, or not seen, in 2020 and beyond—is a ready, educated public.

Alt