Sprint: How to Solve Big Problems and Test New Ideas in Just Five Days

by Jake Knapp

  You should be able to make the first quick draft of your map in thirty to sixty minutes. Don’t be surprised if you continue to update and correct it throughout the day as you discuss the problem. We never get ours right the first time, but you have to start somewhere.

  * * *

  At this point, you will have reached an important milestone. You have a rough draft of your long-term goal, sprint questions, and map. You can already see the basic outline of your sprint: the unknowns you’ll try to answer in Friday’s test and the plotline of your solutions and prototype. The long-term goal is your motivation and your measuring stick.

  For the rest of the day, you’ll interview the experts on your team to gather more information about the problem space. As you go, you’ll add more questions, make updates to your map, and perhaps even adjust the phrasing of your long-term goal. And you’ll take notes as a team, to add more depth to the map on the whiteboard.

  Your job on Monday afternoon will be to assemble one cohesive picture from everyone’s pooled knowledge and expertise. In the next chapter, we’ll give you a recipe for learning from the experts at your company, and we’ll show a nearly magical way to take notes.

  * * *

  I. For more on the Misty Mountains, refer to Led Zeppelin IV.

  II. If you’re counting: Yes, there were more than seven people in Flatiron’s sprint. It’s a guideline, not an ironclad rule.

  6

  Ask the Experts

  Your team knows a lot about your challenge. But that knowledge is distributed. Somebody knows the most about your customers; somebody knows the most about the technology, the marketing, the business, and so on. In the normal course of business, teams don’t get the chance to join forces and use all of that knowledge. In the next set of exercises, you’ll do exactly that.

  Most of Monday afternoon is devoted to an exercise we call Ask the Experts: a series of one-at-a-time interviews with people from your sprint team, from around your company, and possibly even an outsider or two with special knowledge. As you go, each member of your team will take notes individually. You’ll be gathering the information you need to choose the target of your sprint, while gathering fuel for the solutions you sketch on Tuesday.

  Why go to all this trouble? As with many of the steps we do in sprints, we learned to do this one after making a big mistake. When we first started running sprints, we thought we could learn everything just by talking to the people in charge: usually the CEOs and managers. It makes sense. The Deciders should know the most about the project, right? Well, as it turns out, they don’t know everything—even when they think they do.

  We were running a sprint with WalrusCo (again, names and identifying details changed to protect the innocent). We’d already heard everything their CEO and their chief product officer had to tell us. We’d drawn our map on the whiteboard, and we were feeling confident about it. The CEO told us we “absolutely, one hundred percent” had it right.

  That’s when Wendy (again, name changed) stepped in the room. She was full of energy. Her shirtsleeves were rolled up to her elbows, and she rubbed her hands together and paced as she talked.

  Wendy ran WalrusCo’s sales team. And what she understood better than anyone was how customers reacted at different steps in the sales process. She pointed at our diagram. “Here,” she said, “they’re saying, ‘I’ve never heard of this WalrusCo. Why should I trust you suckers with my account number?’ ” She took a swig of water from a paper cup. “Here,”—she pointed to another spot—“we’re going to require their business tax ID. Nobody has that memorized. They’ve got to find the papers, and they’re digging in their filing cabinet. If I haven’t solved the trust issue by this point, game over.”

  Everyone jotted notes. Jake ran to the whiteboard, rubbed out a few lines with his thumb, and drew in Wendy’s corrections. “Like this?” he said. Wendy looked at her watch, then checked Jake’s work.

  “Yeah.” She crumpled her paper cup and tossed it in the trash. “About like that. Look, thanks for having me.” She gave an apologetic shrug. “I got this call.”

  At WalrusCo, the CEO was certain we had covered everything. But Wendy changed almost every part of our map. Now, before you start thinking that WalrusCo’s CEO was a goofball, we should explain that the map was accurate before Wendy came in. It was just more accurate afterward. Wendy put the basic facts into a real customer’s context.

  Nobody knows everything

  What Wendy taught us was that big challenges have a lot of nuance, and to understand it all, you need to incorporate information from many sources. Nobody knows everything, not even the CEO. Instead, the information is distributed asymmetrically across the team and across the company. In the sprint, you’ve got to gather it and make sense of it, and asking the experts is the best and fastest way to do that.

  Deciding who to talk to is a bit of an art. For your own team, you probably have a hunch about the right people already. We think it’s useful to have at least one expert who can talk about each of these topics:

  Strategy

  Start by talking to the Decider. If the Decider is not going to be in the sprint the whole time, be sure she joins you on Monday afternoon. Some useful questions to ask: “What will make this project a success?” “What’s our unique advantage or opportunity?” “What’s the biggest risk?”

  Voice of the Customer

  Who talks to your customers more than anyone else? Who can explain the world from their perspective? Wendy is a prime example of a customer expert. Whether this person is in sales, customer support, research, or whatever, his or her insights will likely be crucial.

  How Things Work

  Who understands the mechanics of your product? On your sprint team, you’ve got the people building your product or delivering your idea—the designer, the engineer, the marketer. Savioke interviewed roboticists, Blue Bottle interviewed baristas, and Flatiron interviewed oncologists. Think about bringing in the money expert, the tech/logistics expert, and the marketing expert as well. We frequently talk with two, three, or four “how things work” experts to help us understand how everything fits together.

  Previous Efforts

  Often, someone on the team has already thought about the problem in detail. That person might have an idea about the solution, a failed experiment, or maybe even some work in progress. You should examine those preexisting solutions. Many sprint teams get great results by fleshing out an unfinished idea or fixing a failed one. Savioke, for instance, had nearly all the pieces of their robot personality before the sprint, but hadn’t had the opportunity to assemble them.

  Talking to these experts reminds the team of things they knew but may have forgotten. It always yields a few surprising insights. And the process has another nice, long-term benefit. By asking people for their input early in the process, you help them feel invested in the outcome. Later, when you begin executing your successful solutions, the experts you brought in will probably be among your biggest supporters.

  Ask the Experts

  Allow half an hour for each conversation, although you likely won’t use all of that time. Once the expert is ready, we follow a simple script to keep things moving.

  1. Introduce the sprint

  If the expert isn’t part of the sprint team, tell her what the sprint is about.

  2. Review the whiteboards

  Give the expert a two-minute tour of the long-term goal, sprint questions, and map.

  3. Open the door

  Ask the expert to tell you everything she knows about the challenge at hand.

  4. Ask questions

  The sprint team should act like a bunch of reporters digging for a story. Ask the expert to fill in areas where she has extra expertise. Ask her to retell you what she thinks you already know. And most important, ask the expert to tell you where you’ve got it wrong. Can she find anything on your map that’s incomplete? Would she add any sprint questions to your list? What opportunities does she see? Useful phrase
s are “Why?” and “Tell me more about that.”

  5. Fix the whiteboards

  Add sprint questions. Change your map. If necessary, update your long-term goal. Your experts are here to tell you what you didn’t know (or forgot) in the morning, so don’t be shy about making revisions.

  That’s it. Your experts don’t have to prepare a slide deck. If they already have something to show, that’s fine, but off-the-cuff discussion about the map and the customers is often more efficient. This need for improvisation is a little unnerving, but it works. If they’re truly experts, they’ll tell you things you wouldn’t know to ask.

  * * *

  Your experts will provide a ton of information. So how are you going to keep track of it all? By tomorrow, when the team sketches solutions, a lot of the interesting details will have faded from your short-term memory. The whiteboards will be helpful, but they’re not enough. You’re going to need some additional notes.

  Imagine that every person on the team took his or her own notes. That would be nice, but if one person alone had an interesting observation, the rest of the group wouldn’t benefit from it. Each person’s notes would be trapped in his or her notebook.

  Now imagine that you are a wizard. You wave your magic wand. Sheets of paper fly out of everyone’s notebooks and organize themselves into one big collection. Then the pages tear themselves into scraps. Then—remember, this is magic—the most interesting scraps separate from the rest and stick themselves onto the wall for all to see. Nice job, wizard! You organized and prioritized the group’s notes, and it took no time at all.

  Unfortunately, we don’t know how to do any actual magic. But we do have a technique that results in organized, prioritized notes from the entire team. And it’s pretty fast.

  The method is called How Might We. It was developed at Procter & Gamble in the 1970s, but we learned about it from the design agency IDEO. It works this way: Each person writes his or her own notes, one at a time, on sticky notes. At the end of the day, you’ll merge the whole group’s notes, organize them, and choose a handful of the most interesting ones. These standout notes will help you make a decision about which part of the map to target, and on Tuesday, they’ll give you ideas for your sketches.

  With this technique, you take notes in the form of a question, beginning with the words “How might we . . . ?” For example, with Blue Bottle, we could ask, “How might we re-create the café experience?” or “How might we ensure coffee arrives fresh?”

  Some of Blue Bottle Coffee’s How Might We notes.

  Now, some folksI will bristle at the slightly unnatural “How might we” phrasing. After all, most people don’t talk like that in real life, and, combined with writing on sticky notes, it can feel a little silly. We had the same concerns ourselves when we first learned about the How Might We method.

  When we tried it, we came to appreciate how the open-ended, optimistic phrasing forced us to look for opportunities and challenges, rather than getting bogged down by problems or, almost worse, jumping to solutions too soon. And because every question shares the same format, it’s possible to read, understand, and evaluate a whole wall full of these notes at once (which is what you’ll do later in the afternoon).

  Take How Might We notes

  Every person on the team needs his or her own pad of sticky notes (plain yellow, three by five inches) and a thick black dry-erase marker.II Using thick markers on a small surface forces everyone to write succinct, easy-to-read headlines.

  To take notes, follow these steps:

  1. Put the letters “HMW” in the top left corner of your sticky note.

  2. Wait.

  3. When you hear something interesting, convert it into a question (quietly).

  4. Write the question on your sticky note.

  5. Peel off the note and set it aside.

  Each person will end up with a little stack of notes—you’ll organize them later.

  There’s no denying that this method is awkward at first, but every team we work with figures it out once they start writing. To better illustrate how all this Ask the Experts and How Might We stuff works, let’s look at part of an actual interview, and the notes that came out of it. In this scene from Flatiron’s sprint, we’re interviewing Dr. Bobby Green, their VP of clinical strategy. This is roughly the first two minutes of his fifteen-minute interview.

  • • •

  “All right, Bobby,” said Jake. “What’s missing on our map?”

  “Well, I can talk a little more about this part.” Bobby pointed to the whiteboard, where the diagram said Search for matching trials. “I’ll give you the doctor’s perspective here.”

  Bobby handed around a few copies of a three-page printout. “This is a typical list of criteria for a clinical trial,” he said. “When we’re trying to decide if a patient might be a match, we’re comparing what we know about the patient to lists like this.”

  The pages were filled with requirements. There were fifty-four in all, everything from “Age 18 or over” to “At least four weeks since prior sargramostim (GM-CSF), interferon alfa-2b, or interleukin-2.” For Jake, Braden, and John, it was tough to decipher. But the point was clear: It was a long list.

  Alex Ingram, Flatiron’s product manager, looked up from his printout. “The clinics don’t have all of this information about their patients, right?”

  Bobby nodded. “Some of these criteria are in the electronic medical record, but a lot aren’t.”

  “Remind us how it works when the info isn’t in the medical record,” Amy Abernethy, Flatiron’s chief medical officer, said. It was obvious that she already knew the answer, but she also knew the rest of us would benefit from hearing it.

  “Well, it depends,” said Bobby. “For example, many trials call for ‘no uncontrolled cardiac disease.’ That’s pretty vague, but it probably means the patient hasn’t had a recent heart attack. That kind of thing won’t be easy to find in the electronic medical record. So someone from the clinic has to talk to the patient or to the patient’s cardiologist. At the end of the day, the oncologist might have to make a judgment call.”

  Bobby set his own stack of papers on the table. “To match a patient to a trial, we’ve got to answer a dozen or two dozen open questions. Now multiply that by the number of new patients every week, and the number of trials at each clinic.” He gave a tired smile. “And as an oncologist, you were already busy anyway.”

  Around the room, people nodded. Then we all wrote furiously on our sticky notes.

  • • •

  A recap: First, Jake, as the sprint’s Facilitator, began the interview by asking Bobby about the map on the whiteboard. That gave us all context for how the new information would fit with what we’d already discussed.

  Next, the team asked a lot of questions. Amy’s phrase “Remind us . . .” is useful, because most interviews include content the team has heard before, at some point or another. That’s okay. Covering it again refreshes everyone’s memory and reveals new details. The “Remind us” phrase is also a nice way to make your expert feel comfortable. Bobby didn’t need that—he’s a confident public speaker—but by asking questions in this way, you can draw out great information from even the quietest person on your team.

  Let’s talk about note-taking. Here’s a basic outline of the problems Bobby presented:

  • The information required to screen patients is hard to find in their medical records.

  • Filling in missing information requires a lot of time and effort.

  • The number of patients, trials, and requirements is overwhelming.

  Ugh. That’s depressing, right? But the entire time Bobby was talking, the Flatiron team was turning those problems into How Might We opportunities. Here are some of the notes they took:

  Reading the How Might We list feels a lot better than reading the problem list. It was exciting when the interviews ended and we saw each other’s notes on the wall. Each How Might We note captured a problem and converted it int
o an opportunity.

  What’s more, each question could be answered in many different ways. They weren’t too broad (“How might we reinvent health care?”) or too narrow (“How might we put our logo in the top right corner?”) Instead, Flatiron’s How Might We notes were just specific enough to inspire multiple solutions. On Tuesday, they would provide the perfect inspiration for our sketches.

  Bobby’s interview illustrates the basic formula for Monday afternoon. You’ll interview experts, using your map as an outline. You’ll take notes as a team, turning each problem you hear into an opportunity. By the time you finish your interviews, your team will have generated a pile of notes. In most sprints, we end up with somewhere between thirty and a hundred. Unfortunately, you can’t make good use of that many How Might We questions. Once you turn your attention to sketching, it will be too many opportunities for the poor human brain to track. You’ve got to narrow them down.

  Organize How Might We notes

  As soon as the expert interviews are finished, everybody should gather his or her How Might We notes and stick them on the wall. Just put them up in any haphazard fashion, like this:

  First, put up the How Might We notes without any organization.

  Wow, what a mess! Now you’ll organize the notes into groups. Working together, find How Might We questions with similar themes and physically group them together on the wall.

  You won’t know what themes to use ahead of time. Instead, the themes will emerge as you go. For example, imagine you were working with Flatiron Health. You might look at the wall and notice a few notes about electronic medical records. You’d pick those notes up and put them near each other. Bingo. You’ve got a theme.