A Product in Five Days: The Design Sprint

By Gilles Raymond

This series recounts the creation of Gilles Raymond’s new startup, Done, from the first idea to the final product with its best and worst moments (past and upcoming). No bullshit. Promised.

Episode 1, “The Two Foundations”
The sale of my previous startup, a news aggregator called News Republic, to a giant Chinese digital group.
Episode 2, “Three Chinese lessons”
What it really means to work for a Chinese company. The need to rethink how we interact digitally.
Episode 3, “Strategic Branding”
How Done drew inspiration from the hospitality industry.

If the outcome of a brainstorming session looked like an Americano coffee, a Design Sprint experience would be more like a ristretto: dense, tasty, reduced to the essentials. For the inception of Done, it had a tremendous impact on defining key functionalities.

In the Summer of 2018, I start to feel the pressure of

launch of my new startup. In about a month, everybody will be on deck, some will quit their job to jump from the cliff. One of them is Sylvain Chaussard, Chief Product Officer. Sylvain was one of the first people I called to join Done. At News Republic (the news aggregator I created in 2008), he was already our product manager; he stayed after the sale to Cheetah mobile, before switching to Done. Echoing my concern about securing the vision around the product, Sylvain raised the idea of organizing a Design Sprint.

His first idea was to go to the inventor of the concept, Jake Knapp, the author of the bestseller Sprint who coined the idea while at Google Ventures, along with Xander Pollock. But it turned out that working between Bordeaux and the West coast of the United States was not easy. Sylvain suggested going with another well-known practician of Design Sprint, a young Swiss guy named Stephane Cruchon, whose company is based in Lausanne.

It was one of the best decisions we took in the early stages of Done. Here is why.

The usual process for designing an application involves a well-known sequence: framing the problem, writing design and technical specs, coding, testing, and hoping for the best. Shortcomings are numerous: the guarantee of months of work, one or two thousand man-hours; endless meetings, gazillions of largely redundant sticky notes, in the best cases summed up in a vast list of requirements. In most cases, there are no prototypes; and more importantly no confrontation of the product by its end-users.

Design Sprint is a five-day boot camp, which starts with a blurry vision of the product and ends with a prototype — one that has actually been tested to a selected audience.

What I like is the purity of it, summed up in this chart:

It started one Monday morning at Done’s office in Bordeaux. As instructed by Stephane Cruchon, we had selected a core team of contributors: product, tech, two external designers with expertise in app developement, myself and Stephane as the facilitator. Five people, five days to go. No time to waste.

We had a vision, but it needed some refining. Our big idea was to address the need of 200 million monthly active users with a cool-to-use tool that would perform daily basic business tasks: communication, organization, and collaboration. We wanted an app so easy to use that it would trigger immediate and widespread adoption by users put off by the complexity of Slack, Microsoft Teams, and the like.

The first day of the Sprint was devoted to refining and validating the vision, reframing the question, and finding out what the product’s key functions must be. Each group member came with their own ideas. Some wanted to put ease-of-use first, others the Swiss Army knife aspect of it, etc.

The next day was about selecting the target group. Who will be our typical customers: what they look like, the primary issues in their daily tasks, how they currently collaborate with their team, how they communicate, share files and organize meetings with among themselves (or, better minimize them). We defined five or six personas, each in greater detail: where they live, their routine, their profile, etc. We simulated the key moments in which they would have to solve a particular problem. This process would prove essential for the final phase, three days later.

A key moment involved what is called the Crazy Eight: each of the participants must sketch eight screens in eight minutes (I found this 2-minute video a good explanation of the method). For each of the six personas, we had six scripts, six practical situations applicable to the product.

This notion of time and pace is critical to the process. As such, the key accessory to a Design Sprint is a timer compulsively activated for each phase of the process.

On Wednesday, we had to make a selection on the scenarios we were about to test. For instance, an important feature of the Done app is organizing a meeting in no more than three taps. We knew it is an important differentiator that will set the tone for other functionalities of the application. In order to fully test it, we needed to build it in the most realistic way.

That crucial part occurred on the fourth day. From 9:00 am to 2:30 am, the team transformed a set of sketches and wireframes into a working prototype, using InVision, a great prototyping tool for mobile that perfectly mimics mobile interactions. They built dozens of screens in the process:

Late on Thursday afternoon, Stephane turned to me and said with a smirk: “Now we need to recruit the testers. Tomorrow is the big day…”. Answering to my skeptical look, he added: “Yes… We know precisely the key functionalities we want to test and the kind of people we want to attract. So let’s find them, they must be here tomorrow by ten.” Stephane has built a process and a methodology for each step of the sprint. He knew how to find the personas we need. A moment later, a targeted classified was posted on Facebook. Candidates were incentivized with €100 Amazon coupons. We quickly got what we needed: a freelance worker, an executive assistant, a traveling salesperson, and a startup entrepreneur (Stephane managed to get a few back-ups in case of defections).

At ten sharp, the testers were on deck. Participants were asked to organize a Christmas party in their company; they had to find a date that fitted everyone, distribute various tasks, send multiple messages, put files in a folder, assign a to-do list, etc. Each of the five 30 to 40-minute sessions was recorded.

Honestly, I would not have imagined that possible: on Monday morning, we had a vision of our product, but it was wrapped in a haze of fuzzy ideas and concepts. By late afternoon on Friday, we had a prototype featuring the look and feel of the final product along with essential working functionalities. Above all: we had submitted the key features to the users, collected and mapped their feedback. All in five days.

One year into the development of Done, I cannot fathom how critical the Sprint Design process was. Because I have the chance to work with a great team, we were pretty much on target with the key functionalities of the app. After the test, we ended up keeping more than 80 percent of the essential features we had envisioned. More importantly, the process initiated a long-lasting dynamic in the making of the product.

Gilles Raymond

With Deepnews Digest, we select the best stories of the week thanks to the Deepnews.ai scoring system. Get it in your mailbox every Friday at 6 am EST, 12 noon CET Subscribe now!

A Product in Five Days: The Design Sprint was originally published in Monday Note on Medium, where people are continuing the conversation by highlighting and responding to this story.