Sprint: How to Solve Big Problems and Test New Ideas in Just Five Days by Jake Knapp, John Zeratsky and Brad Kowitz
My rating: 4 of 5 stars
This book is really to take a challenge (high stakes, not enough time or just plain stuck) and learn the most that’s possible within one week. The book really focuses on two things firstly the benefits you can get by trying out risky ideas in the simplest way possible and secondly how to execute this with tips on facilitation and organisation.
Its key to have the right seven (would be ideal) people.
- The decision maker is key – without their buy-in, without which you are likely to not take the right challenge to push the product forward.
- Financial expert so you are clear on how this will impact the business
- Marketing expert so it fits in with the rest of the corporate message
- Customer expert so you have someone who understands what the customer wants or would use
- Tech/logistics expert who understands what the company can deliver
- Design expert so that you can make something people want to use
- Trouble makers these people will challenge what you are doing so the results are better
If you have extra experts have them attend on the Monday and decide on who will be the facilitator who will need to remain unbiased for the week.
No laptops, no phones – fully dedicated time from 10am till 5pm Monday to Thursday with a morning break at 11:30am, 1hr lunch at 1:00pm and another break at 3:30pm has been found to be the best. On Friday its a bit different so 9am till 5pm so you can squeeze in the real users.
Monday – Make a map and choose a target
Why are we doing this project? Where do we want to be in six months, a year, or five years from now?
- What questions do we want to answer in this sprint?
- To meet our long-term goal, what has to be true?
- Imagine we travel into the future and our project failed. What might have caused that?
Make a map of the way people currently complete the task – this will contain multiple actors (including the customer) and a clear ending, keep it simple, suitably high level and ask for more input to ensure that its accurate by welcoming experts to pop in for 30 minutes each during which time you can improve the accuracy of the map to better reflect real life.
Once you know the map you need to choose a target – a technique for this is for people to make “How might we …” cards independently before putting them on the wall together. Once on the wall group them and dot vote on the questions which people feel are the most valuable. The decider then needs to decide which step in the map will we focus on and which questions we will look to answer.
Tuesday – Sketch competing solutions
Start off by seeing other things which might be approaches to solving the problem, give three minute demos of things which you feel are relevant.
Each person independently should sketchup one or more ideas. Give it a title as it should be anonymous. Don’t share them with others at this stage.
While this is all going on someone not part of the team should be finding people for friday – real potential users, advertise on bulletin boards and offer a voucher reward but ask them questions to ensure they are your correct target market, if they are not then Friday will be a wastes. So when questioning be careful not to push them to the answer you want.
Wednesday – Decide on the best
- Put all of the sketches on the wall
- Look at each independently and put dots on features which are interesting
- Quickly discuss the highlights of each sketch and capture big ideas.
- Each person votes for one solution
- The decider makes the final decision – this could be to try two things if you want to compare different ideas (like A/B testing)
The next step is to plan things out in more detail, starting from how the user starts an interaction with your product (e.g. reading about it in a newspaper). By having a storyboard today it should be quicker to prototype tomorrow.
Thursday – Build a realistic prototype
Fake it, build a convincing facade this will give you the most learnings with the minium work. Things like slides etc can be great to present things quickly but it should be immersive, the person on Friday must feel they are using a product. You can prototype anything, they will be disposed of and build just enough to learn what you need.
Have a stitcher whose job it is to pull together all of the parts the rest of the team produce and make it into a coherent product fixing things like layout, the customers name to be consistent etc.
At the end of the day have time to trial it lead by the person who will run the trial on Friday.
Friday – Test with target customers
Five 1hr interviews a 30 min break after each and a debrief is the order for the day. From their research the more people you have you get diminishing returns and five is the sweet spot.
For each interview have it so that there is just the customer and an interviewer, others should watch remotely and taking notes is key to getting value from the day. The interviews follow the format:
- A friendly welcome to start the interview
- A series of general, open-ended context questions about the customer
- Introduction to the prototype(s)
- Detailed task to get the customers reaction to the prototype
- A quick debrief to capture the customer’s overarching thoughts and impressions
The onlookers should be noting down things which were good, things which were bad for each of the interviewers for each of the stages in the prototype to see if there is any correlation or new ideas which come out of it.