Jake Knapp, the design partner at Google Ventures, first developed his design sprint process working on Chrome, Google+, and Gmail for Google proper.

The process was inspired by similar processes at Ideo and Stanford’s Institute of Design. They squeeze a maximum amount of design education into a necessarily short amount of time.

The process isn’t at all similar to brainstorming, despite it’s outward appearances.

The process revolves around a series of forced constraints. For example, participants must propose eight solutions to eight problems, with just 40 seconds apiece.

Participants come up with ideas silently and independently; only after everyone is finished do they get to post them up. In the process, the design sprints avoid two of the toxic outcomes of brainstorming: Unreasonable solutions and groupthink.

Another key to avoiding silly solutions: Getting test users into the room as early as possible, to vet the work even though it’s necessarily very rough.

Ideas get voted on with stickers once they’re posted. Decision makers get "super votes"--thus bowing to the reality that companies are not purely democratic.

Co.Design

From Google Ventures: 4 Steps For Combining The Hacker Way With Design Thinking

With 150 companies to mentor, Google Ventures’ team of seven designers has time only for a sprint—and that’s one reason why it’s effective.

Jake Knapp is like the middle-school teacher everyone should have had. Tall, with clear blue plastic glasses, he holds a classroom-clock-sized timer like a football while explaining the first activity of the day: an exercise called "Crazy 8s" that involves drawing eight different solutions, with 40 seconds for each, to address one design challenge.

On a whiteboard nearby, he’s sketched out a five-step game plan for redesigning the Blue Bottle Coffee website. Below the first box in the schedule, labeled "Understand," there’s an unsure emoticon. It turns to a smiley face by the next step, "Diverge," sticks its tongue out for "Decide," sports a toothy grin under "Prototype," and alternates between despair and glee under "Validate." Blue Bottle, I am told, is currently somewhere between the number two and number three emotional states.

A handful of the company’s employees—including both its COO and the person who handles inquiries from the website—sit scattered on cushions and couches around Knapp at Google Ventures’ San Francisco headquarters. They scribble silently on printer paper as he calls off 40-second intervals from his clock. "We can get good stuff out of ugly drawings," he reminds them. Knapp, who joined Google Ventures’ design team last March, first started running what he calls "design sprints" when he worked at Google proper. After leading projects that became Hangouts for Google+ and Priority Inbox during his 20% time, and inspired by design exercises from agencies such as Ideo and the Institute of Design at Stanford, he asked to run workshops for projects across the company. Google being Google, he got the job. And when he joined Google Ventures, he brought the sprints with him.

Google Ventures’ design team works as a resource for its 150 or so portfolio companies—doling out advice during office hours, weighing in on design hires, and teaching companies that may not employ a single designer a framework for thinking about design. But the team has only five members. Sprints help squeeze a maximum amount of design education into a necessarily short amount of time.

"It’s not like we’re just sitting around saying, 'Ah, there’s one solution, there’s another,'" Braden Kowitz, the team’s founding member, explains before the start of the second day of the Blue Bottle sprint. "It’s very, very structured."

Daniel Burka, who recently joined the team through Google’s acquisition of his startup, Milk, looks to Knapp, "I assume you tried [brainstorming] early on and pretty quickly figured out it was crap."

"Yeah, yeah, that was years ago," Knapp responds.

"But if someone outside looked in," Kowitz says, nodding toward the rows of yellow Post-It notes on the conference room’s floor-to-ceiling windows, "they must think we’re brainstorming."

They’re not. All of the Post-It notes, whiteboards, and doodling might at first glance look like agency hogwash, but they’re actually part of a constantly evolving recipe for baking good design into Google’s portfolio companies. Here are its most important ingredients.

1. Forced Constraints

"You can talk about, ‘Ooh, should we do a scrolling page or a step-by-step?’ for hours," Kowitz says. But with no data about which is actually better, it makes a lot more sense to just pick one. That’s easier said than done. Which is why Michael Margolis, the team’s only Seattle-based member, might influence the productivity of Blue Bottle’s design sprint in San Francisco more than anyone who is actually in the room. Using Craigslist and a decade of user research experience, he has scheduled a group of strangers to test the startup’s prototype on the last day of the sprint.

"Any of the steps," Knapp says, "you could probably spend a week on them. But knowing that by Friday morning we have to have something done, makes us find ways to make decisions faster."

2. Separate Processes for Pinpointing Problems and Solutions

One of many Post-It note groupings Blue Bottle has created by its second day of the design sprint is devoted to filling in the blank of the phrase "How About We." It’s the opportunity-focused version of putting all the problems on the table.

The design team uses the same method for collecting potential challenges as it does for most of its idea-gathering activities. Each member writes on separate notes without speaking, campaigning, or merging her ideas with others. Only after all ideas are stuck to the glass do participants share them, silently, voting on those they like by sticking tiny round stickers next to them. Decision makers get bigger stickers, or "super votes," because, while democracy is nice, it’s not always realistic.

Blue Bottle’s website isn’t terrible—in fact, online purchases account for about 10% of its business—but it’s a far cry from the customer service and obsessive quality of the company’s cafés. "When we’re at our best, we’re our best in our stores," Blue Bottle CEO James Freeman says. "People like the experience they have in our shops for whatever reason, and a lot of yesterday was figuring out how to get them not a similar experience but at least an analogous experience when they become a guest on our website. … How do we host people in a non-Internet way while they’re on the Internet?"

Narrowing down the wall of Post-It notes to a handful of design challenges such as how to help customers decide which coffee to order or find a nearby café took almost half of a day, but, says Kowitz, it’s worth it: "If you don’t do this, I’ll come up with a solution, Jake will come up for a solution, and we’ll battle. I want mine to win, you want yours to win, and after about a half an hour of arguing, we realize we were trying to solve different problems and we have to unwind all that and discuss this anyways."

3. Story Boarding The User Experience

Another round of Post-It note doodling later, and Blue Bottle had produced a series of comic strips depicting potential solutions to the challenges they had selected. Those notes, grouped into storylines on pieces of copy paper, line another glass wall of the room. Most of the artwork consists of different-sized squares, with blurbs of text and arrows explaining how the experience would work, though the occasional stick figure does make an appearance.

As with Crazy 8s, artistic prowess isn’t really important. By thinking of potential solutions in the context of customer stories, the hope is that participants will start to think of their product not as code on a website but as an experience (Airbnb uses a similar process; read more about theirs here). "The way that I explain it to engineers is that you write code and it just sits there—it’s static," Kowitz says. "It’s not until the code actually runs that you get to see if it worked or not. That’s what our stories are like. …The story is this exercise that can help us see what the design is going to be like to actually use."

4. Independence

You’d be hard-pressed to find a more talented group of designers than the team Google Ventures has put together, but the prototypes they create while Blue Bottle takes a break on the fourth day of the sprint look nothing like Photoshop gems; they’re hasty keynote presentations. Not only does Keynote work well within time constraints but it avoids the problem of, as Kowitz puts it, "polishing a turd." Margolis fakes functionality during usability tests by stealthily moving slides forward and backward as users click on fake buttons.

Blue Bottle and the design team watch a highlight reel of the tests on the last day. Sometimes, Knapp says, companies cheer on their videotaped testers. 'It’s right there,' they’ll yell, 'why can’t you see it? Just cliiiick it!' But most of the time it’s more of an "oh" situation. "I think the word for the folks from the company is 'clarifying,'" he says. Often they’ll put this clarification to work in another design sprint. And then, possibly, another. But the idea is for companies to be able to eventually run their own sprints.

Knapp describes his role in baseball terms. At first, he’s like the coach, running drills and teaching swing techniques. Then he becomes more like a manager and eventually as hands off as an owner. Startups leave the sprints with what looks like a band instrument case full of art supplies—all the Post-It notes, stickers and pens they need to run sprints in their own offices.

"If you were just to bust in the door and say you need to work differently and hire a bunch of people, they’d kick you out," Kowitz says. "What we do is design work with them in places like this, where we show them how it’s done, and then usually, they say, 'That was magic. How do we do that?'"

Photos by Christophe Wu/Google with Permission.

Add New Comment

9 Comments

  • GKVanPatter1

    Sarah: What you saw st google is a great idea in action...not a new idea but a sound one. These are well known rewrapped conventional level 1 innovation techniques. Sixty years after brainstorming was created there is not one way to do multiple participant divergence but rather many variations including what you saw. Wooo Hooo....Lets also keep in perspective that for most organizational leaders driving meaningful change and effective innovation capacity building involves much more  than redesigning web sites. If you are interested in the history/Origins of How Might We feel free to see Humantific. Thanks to Sidney Parnes it is a phrase construction that first appeared in applied creativity literature in 1965-66. Today it is what we do with How Might Wes that makes a difference not simply their generation. 

  • DMC

    'Oh it's not brainstorming, we've added constraints and a card sorting exercise.'

    Sounds like a lot of 'agency hogwash' to me.

  • Simone Oltolina

    As much as I like designing/doing workshops whose ultimate goal is to come up with ideas, I'm afraid the hard truth is that the end result is less about the process and more about the quality of the people you get onboard.

  • Oysterjim

    Interesting how orthodoxy creeps into these discussions of the "best" way to come up with new ideas. Twenty plus years have taught me that there are many useful approaches, thhe success of which depends on the goal, the group, etc. for example, working alone and voting might be less effective when consensus is needed. No need to view different approaches as inferior.

  • Matt K.

    Sounds like a valid and effective process for working with time constraints.  Just as brainstorming will remain an effective process when working with the luxury of  a little more time.  Branding brainstorming *results* (as opposed to "solutions," which they are not) as potentially "toxic," "unreasonable" and "silly" misses the point.  Brainstorming is not about solutions.  (Those start to be defined later, at the meeting that sounds more similar to where these people start.)  Brainstorming is about getting those neurons firing without constraint.  As a designer, I've been in many situations where the final, serious solution sprouted from a completely silly seed.

  • Jag

    The phrase for pinpointing problems is "How might we..." (the link in the article actually shows this.) "How About We" is an online dating startup in Brooklyn.

  • rfairfax

    Can we see the results of this process? Without seeing actual design work, we're left with the feeling that this is - as the author put it - 'agency hogwash'.

  • John Zeratsky

    Hey Russell, the Blue Bottle team continues to work on the redesign we started in our sprint with them. It can be hard to judge whether a design is successful without knowing all the context, goals, constraints, etc, but when the new site launches it'll give you a chance to see an example of where this process leads. Let us know what you think!