Case Study: Launching a Startup on the Market

For three years now we have been studying and experimenting with the Lean UX methodology on a sample of roughly one hundred clients. After three years we have solidified the method, and now we can say:

With that said, let’s take a very recent startup as an example to show the world our method:

Fabrizio has been a Cantiere Creativo client for a long time and is Co-Founder of Mediplus Srl, a company specializing in the sale and distribution of medical devices.

Recently Fabrizio, like many of us, had to deal with the healthcare system, and this time not as a business. Faced with the hassle of booking some diagnostic tests, he grew more and more convinced of the idea of simplifying the process with an app where you can find the best offers with the shortest possible waiting times.

As we said, we’re talking about Lean UX, a methodology born from the experience of Jeff Gothelf and Josh Seiden that is essentially the sum of Agile, Design Thinking and Lean Startup.

The process is fairly complex, but over the years we have managed to simplify it a lot. With clients, at least at the beginning, we try to keep things simple (and only later dig into aspects such as iteration). Here are the most important steps:

1. Discovery Meeting → We define the startup’s stage
2. Kick-off Workshop → We create the Lean UX Canvas
3. Design Studio Workshop → A creative session to design market experiments
4. Brand and product design → Brand research and creation
5. User research interviews → Distributing and processing surveys
6. Online page → We test the idea with real traffic
7. KPI monitoring → We analyze the results of the experiments

This process can change from client to client depending on the project.

This session helps us answer the following questions:

1. Have any market experiments already been run?
2. Do we have quantitative studies (by specialized analysts) on the industry and the startup idea we have in mind?
3. Has any qualitative research on our target users been carried out?
4. Does the client have previous experience as CEO of a (legally incorporated) startup focused mainly on the web market?

If the answer to all of the above questions is No, the startup is considered early-stage and the process follows the one listed above.

Tikeep is an early-stage startup. The last question, about previous experience, tells us how mature the client is, especially whether they are receptive to iterative methodologies like Lean UX. Fabrizio in fact already has a lot of experience in web entrepreneurship, having co-founded a well-funded startup that has been online for many years.

This session takes place in our workshop room with sticky notes, whiteboards, markers and a whole range of physical tools that spark creativity and help organize information.

As mentioned, what we define during the Kick-Off is the Lean UX Canvas. Here’s the one we came up with for Tikeep:

The logic behind this Canvas (and the table that follows) is explained during the workshop. Here’s an article that goes deeper for anyone who wants to know more.

Based on the Canvas, we generated our first solution hypothesis:

We can’t go into all the details we share with the client through short presentations and plenty of explanations along the way. Suffice it to say that with these two diagrams (and the proto-personas, which we’ll see in a moment) we justify our product idea by analyzing our target users’ needs as well as our own goals. We also lay the foundations for the experiments that, if positive, can tell us we’re ready to start a small version of our product in that market.

What I have called “target users” so far gets a more precise name: “proto-persona.” If the target users are the slice of users we want to go after, the proto-persona is their profile.

Here are the ones we identified for Tikeep: the non-digital patient and the digital-oriented patient.

Both proto-personas were developed by defining three reference scenarios, moments in their lives when they might interact with our product idea. Let’s take a look:

For each of these scenarios we mapped the mental model, meaning the set of needs, problems, desires, concerns and so on tied to that scenario.

Based on the mental models of each scenario, we designed the aspects of our product that (in our view) would address most of the proto-personas’ needs.

During the workshop we sketched our product idea on paper. Each participant did a quick competitor analysis and then put their own idea on paper. At the end of the session everyone presented their idea, then we voted on the aspects we liked best and settled on a solution that combined them all.

This workshop is based on Design Thinking, a user-centered approach to product design. We cover it in more depth in this article.

Over the following days, with all the paper sketches in hand, the UX Designer drew a first version of the flows as high-fidelity wireframes:

The screens above might look like the final design, but they are still just wireframes, with some visual design added to give the designers a clearer direction.

After some work on the brand:

We produced more polished screen designs and then animated the entire flow:

Based on the pain points identified through the proto-personas and the riskiest assumptions that came out of the Lean UX Canvas, we created a survey on Typeform:

It gathered important results for mapping and validating the pain points:

Finally, we met to review the results. Here are the most important takeaways:

1. Roughly 70% of people book a doctor’s visit or test only occasionally
2. 🔥 PAIN POINT: 70%+ of people are offered appointments too far in the future
3. 70% of people go private to get seen quickly
4. 🔥 PAIN POINT: 70%+ of people see private healthcare as expensive
5. 80%+ of people would like to know more medical facilities so they know where to go
6. 60% of people don’t feel lost when they often switch medical facilities
7. 🔥 PAIN POINT: 80% of people feel the need to be followed by a doctor
8. Satisfaction with private healthcare = 4/5
9. Satisfaction with public healthcare = 3/5
10. 85% of people would use a website that offers a varied choice of medical facilities and helps them spend less
11. 🔥 PAIN POINT: the final open-ended field revealed that the biggest perceived obstacle is waiting times

As with most early-stage startups, we designed a Coming Soon landing page for the product.

The landing page was designed around storytelling based on the biggest issues that emerged from the surveys. Here’s a screen recording of the page:

See it live here.

In part 2 of this article we’ll write about how we bought traffic and managed conversions to track some key KPIs.

Before that article comes out, though, we want to show you something about the chatbot. We decided to capture incoming leads through the Intercom chatbot. Here’s how:

In practice:

1. We buy traffic from Facebook
2. We bring the lead (user) to the landing page
3. We open the chatbot on the “Request an invite” button
4. We start a conversation, saving the email, name and a couple of notes on the user’s behavior regarding private healthcare
5. We connect the email to Mailchimp, where we save the lead already segmented by behavior
6. We immediately send a follow-up with a discount code and a short story about our mission.

Based on the interactions we collect over a couple of months, we’ll decide which first version of our product to build.

See it in action! 👉 www.tikeep.com

See you when part 2 comes out!