What’s the role of designers in the age of big data?
Illustration by Erin Schell
Our lives are data saturated. Everyone seems bent on telling us so. Someone will insist that in November 2012 the typical cat was exposed to more raw information than 2.39 Eleanor Roosevelts experienced in their entire lives. Or that all of the tweets sent in three seconds, printed out in 5-point Cooper Black, would still be less awful to read than Fifty Shades of Grey.
And what are we to do with all this data? The verb of the moment is visualize. Take a bunch of numbers and make—a chart? An infographic? Everyone loves a choropleth map, right? Especially if there’s an election happening. Or a touchgraph, with all those lines and nodes to show you that the Stones are one degree of separation from Hendrix. Useless, but neat.
With the data glut, there are new jobs with hot new titles: data scientists, data journalists, interaction designers. These people are supposed to work together as a sort of cognitive oil refinery, absorbing the crude data and producing light, sweet charts. And in the course of my days as an Old Web Person, I sometimes help people assemble interactive teams. The other day, someone at a magazine asked: “How many interaction designers should I hire?” My first thought was, All of them! But then: Wait.
You have all this data. Rows, columns. Inside your database are millions of songs or thousands of movies (Spotify, iTunes, Netflix) or millions of products (Amazon). And as an interaction designer you need to make something out of that. Something interesting and engaging. Something that brings in traffic. Something with a BIG RED BUTTON saying buy me. That’s the fundamental problem of interaction design in the age of big data: How do you convert your data into traffic and purchases?
Think about Spotify for a moment. If you haven’t used Spotify, it’s like iTunes, but all of the music is already there—millions of songs. You subscribe to the database and listen to whatever you want, whether that’s Bach or a Katy Perry song. The Spotify interface has a lot of “social” stuff that tells you what your Facebook friends are listening to, but underneath that, the core interface is just a table, with Track, Artist, Time, and so forth running across the top. Much of my time in Spotify is spent searching and clicking, then dealing with the resulting table of results.
It’s a pretty ugly table, crabbed, with small type and minimal padding. It’s a huge step back from a bill of lading from 1780 or a chart of sunrises in the Old Farmer’s Almanac. In truth, Spotify at its core is a spreadsheet of music. And yet I use it all day, without thinking, while I work on websites that encourage people to click and buy.
I’d be sad if Spotify took away my spreadsheet of music. I’d be happy, however, if they made the colors a little less goth. And I also wish they’d let me play with the data. I’d love a pie chart of the music genres I prefer. I’d like a bar chart of the dates that the songs I listen to were recorded. I’d like to know when I listen to music the most. I’d like to know how much money I give to individual artists, as well. A personal music-consumption “dashboard” would be excellent.
I wish for the same from all the big, database-driven websites. Amazon and Netflix come to mind. They’re so busy pasting in recommendations on every web page that I doubt they’d want to risk a single bland table in the place of all the pictures and yellow stars. But imagine if you had the freedom to make your own recommendations! The ways you could explore their world and the weird decisions you could make. A single scroll would bring forth rankings and connections hitherto available only after a thousand clicks. What was the first vampire book published in 2005? What were the most popular words used in the top 50 cookbooks of 1998? Nothing beats a tabular list. That’s why people keep opening spreadsheets at work. Spreadsheets kick-started the computer revolution with Visicalc and they’ve been with us ever since.
All of this changes the role of the designer. Because instead of making charts and graphs, we’d be developing the tools that let people make their own charts and graphs. In this new tabular world that looks more like Excel and less like a web browser, a designer becomes the art director to the masses, the creator of visual systems that can be used and reused by millions of people per minute.
So how many interaction designers should my friend hire? Maybe she needs art directors instead, people who can invent a language (tweet, like, poke, fave, wishlist), use it to organize a world (track, director, related artists), and then teach the users to speak that language.
As a class of humans, interaction designers have worked out how to get people to click on things. They’ve done everything they can to make it effortless (hence the title of a classic in the field, Don’t Make Me Think, by Steve Krug). But the future is different: The transaction is over, the user has been obtained. Now what? It’s like a horror movie: The click is coming from inside the paywall. People actually want to think. But they need to know the thoughts available to them. It’s up to us to speak first, to find the common language and then to share it so that people can make sense of all this data for themselves. Start with a table . . .