Google News Lab: 5 Insights From Visualizing The News

We talk to Simon Rogers, data editor at Google News Lab, about the biggest ideas to come out of this year’s Year In Search.

For the past 15 years, Google has presented the biggest stories of the year according to Google search data in its annual Year In Search report. This year, Google handed the job over to Google News Lab, a recently formed group of data scientists and designers who have been making Google data accessible with projects like Google Trends. The result is a kind of magnum opus of data visualizations that gives insight into not only the biggest stories of the year, but also how we use and consume news.


Click through this year’s Year in Search, and you’ll find something different from the straightforward end of year lists of previous years. Visitors are greeted by an interactive timeline of the year’s biggest news events broken out by months; within those events, a slew of charts, graphs, datasets, and other visual tools distill the information further. The search data is updated in real time, so that the same visualization will give different information if viewed in December, for example, than if viewed in January.

“We wanted to do something different, transparent, and open. The insight we have into how people react to the news–nobody else really has that,” says Simon Rogers, the News Lab’s data editor, who joined Google in March after working for both the Guardian and Twitter. “We wanted to make a useful tool for exploring the news and seeing how the world worked for the last year. We wanted to provide something beyond the social media echo chamber and show what people really care about. ”

Google is uniquely poised to do that. Search queries not only show what people are interested in and when, but also people’s concerns, curiosities, and positions on public issues. As Rogers puts it, “there is an honesty to it because you are implicitly interested in what you’re searching for.”

There’s also an immediacy: as soon as something happens Google can see people start to search for it. That information has provided some valuable insight into how people are accessing the news (hint: it’s not always mainstream media) and how it spreads. In tracking search results for the Paris attacks of November, for example, Rogers and his team could see when the news began to hit certain areas of the world based on when people began searching it.

Ultimately, Google’s Year in Search demonstrates how powerful a tool data visualizations can be for revealing macro trends. The News Lab’s hope is that by making this data available and understandable, journalists, policy makers, and entrepreneurs will be able to both expand on it and put it to good use. Below are some of the biggest insights into visualizing the news from Rogers and his team at Google News Lab.


Questions Can Be A Powerful Storytelling Tool
Google gets over 3 billion searches a day from people all across the world. For the Year In Review alone, the News Lab pulled from trillions of searches over the past year. That volume offers unparalleled insight into global and local trends, but Google also has another significant advantage: the searches are voluntary. The data shows people’s interests but also what their concerns and curiosities are, from movies and music to national and international news.

When you’re armed with this much data, of course, you have to find a way to present it that’s accessible to the general public. “We’re thinking what’s the most useful information around the story what will make it more powerful or interesting,” says Rogers. “I’ve said this about data journalism before. It’s not about throwing everything in a visualization, it’s about telling a story. You have to narrow it down–take big data and make it small. Take all of this information and say ‘what is the biggest insight we get from this?’ That’s a journalistic skill.”


People Break The News Before The Media
For an excellent example of that philosophy in practice, look no further than the News Lab’s interactive data viz, which shows how news of the Paris attacks spread around the world. “The attack in Paris was something everyone was searching,” says Rogers. “We picked 20 or 30 cities. In those cities, we looked at ‘when was the spike?’ People in Berlin were searching for it within a few minutes of that happening. It was incredible that news spread within minutes, and that was something we wanted someone to make clear.”

To do so, they turned to designer Anna Vital, who had created a similar visualization for the terror attacks on Charlie Hebdo. While her Hebdo infographic was static, Vital designed this one to be interactive so users could see the news reach certain parts of the world and what type of questions people in those places were asking. An arched time line shows the minute that news of the attacks hit that part of the world. Remarkably, people as far as Sydney, Australia, were searching the attacks almost a full hour before the mainstream news had picked up the story.

“This data says something about how the world works and how we interact with each other,” says Rogers. “For something like Paris, we needed a designer to really think about how can we make the story more powerful, more useful, clearer and easier to understand.”


Interests Vary In Different Parts Of The World (And It’s Important To Show That)
One of the key things that Rogers’s team wanted to do differently with this year’s Year In Search was to make the data adaptable to specific locations. A drop down menu on the top right of the page allows users to see a global overview of top searches along with timelines for 13 different countries–countries that “had ‘super stories’ this year,” according to Rogers. “We wanted to make it more of a global experience because Google Search is being used differently in different places and we wanted to reflect that.”


Certain news stories–like the Paris attacks, for instance–were big across all countries. But location-specific events like the Labour Party leadership election in the U.K., the reversal of Japan’s military legislation, and the retirement of cricketer Virednder Sehwag in India shows the major interests and concerns in those countries specifically. Within the pages of those local searches, maps demonstrate how interest unfolded across the country and in other parts of the world. If the News Team had merely translated the global trends into different languages for different countries, viewers would have missed these types of insights, Rogers argues.


We’re Just As Interested In Adele As We Are In World News
Unsurprisingly, major news events like the terrorist attacks in Paris and the Volkswagen emissions scandal were popular search items both globally and locally, but in some months pop culture dominated. In December, for example, Star Wars took the crown, and in February The Oscars were the hottest search item. “The world is made of lots of things we care about music and movies and this shows how these things fit together. That we care about things that happen in the world, that’s reassuring to me. But then you also might be searching for something like The Dress,” says Rogers. “The way that these things coalesce into this mixed picture of the world is really interesting.”


Data Could Help Predict The Future (In The Future)
Visualizing the news isn’t just valuable for analyzing past events, it could also be useful for predicting future ones. Rogers gives an example of a visualization they did earlier in the year that focused on location-based search queries during general elections in the U.K. “We thought it would be fun to see in constituencies who the most searched for candidates were,” he says. “The results were so far removed from the polls we said ‘this is silly, but it’s fun.’” Turns out, Rogers and his team got closer to the final election results just looking at search queries than the polls.

When asked if he thinks Google Trends could be a reliable tool for predicting future outcomes, Rogers says that the Lab still has a ways to go before he could say that with any certainty. He points toward researcher Nikos Askitas, who claims, with Google search data he could have predicted the results of the Irish same sex referendum to a T. “I don’t know if that’s true, but we’re at the beginning of the process, we’re starting to lift the lid on data and see what it can tell us [about the future].”

About the author

Meg Miller is an associate editor at Co.Design covering art, technology, and design.