I started mentoring a business called CoursAvenue (previously LeBonCours) at Le Camping. Their goal is to make it easier to find local, well-reviewed classes on anything from yoga to dance to cooking. I think of it as Yelp that can deep dive through class listings to find the right class for you based on criteria like location, time, cost, teacher and reviews. Today we had a meeting about how to get their lean process started.
Your company exists to solve a problem, not to push a product
First thing I’ve noticed about startups calling themselves lean is that they often aren’t. They use the terminology but they have trouble applying the principles. My goal for mentoring (and in a new class I’m putting together for Republikken and Paris Factory) is to dramatically simplify going lean. In other words, less theory and calling yourself lean, and more actionable ideas and actually being lean.
CoursAvenue has already created a beautiful web application to access course listings in Paris. They used a “Wizard of Oz” test (see chart below) to manually populate the back end of the course catalog by combing through class listings for thousands of schools in Paris.
The first tough realization – the web application isn’t CoursAvenue, the defined customer problem or a value hypothesis! It is a MVP (admittedly a very elaborate one). It is probably more elaborate than was needed to test the critical assumption that people will search for courses online (see coursehorse.com for a much simpler iteration and of course Yelp.com for searching for reviews more generally).
CoursAvenue is about solving the problem of the “space”, in this case how people find courses to attend. There is room in this space for CoursAvenue, no matter what their product turns out to be. Detaching from the original MVP can be a challenge for startups. That’s why there is so much value in keeping the MVPs simple, so there isn’t so much emotional investment in them, making them easier to discard or keep as tests are validated or not.
Once we accept that premise (shifting from MVP as the reason for the company to solving a customer problem as the reason for a company) , then we are moving! We began by discussing the five principle methods of testing assumptions.
Specific tools (MVPs) to validate and test assumptions
I believe the core of Lean Startup, and where young startups can start getting the most impact with the least amount of work, is to think in terms of assumptions and tests. In my experience I’ve found five main ways of testing assumptions.
|Customer interviews||Best for canvasing early ideas||Place ad and interview responders|
|Landing pages (smoke test), Kickstarter, high hurdle (cash)||Offer product for sale before it exists||Use mockup of product to test customer interest|
|Models and prototypes (popup store, food carts)||Low-investment shop to sell real products||Test dim sum taste before opening a full restaurant|
|Concierge – manually do work before automation||Do all the work by hand to see if solution solves customer problem||Food on the Table - http://bit.ly/10kpd6Z|
|Wizard of Oz – create front end, but do back end manually||Customers think a computer is doing all the work but in reality a human is||Zappos – pretend to be a fully-formed web app, but do backend by hand|
List out your assumptions (Critical, Important, All of them – you choose)Each of these tools is the start of a MVP. I’ll discuss each in more detail as CoursAvenue uses the different tools to test their own assumptions.
Another common problem I’ve found is that what many startups think is the assumption they want to test is in fact a bundle of separate assumptions. In order to best test the assumptions with the MVP tools listed above, we need to get the assumptions into working order by unpacking, unbundling, and separating out the ideas.
Off the bat for CoursAvenue, we had to decide if we are going to go after schools or students first. CoursAvenue wants to be a marketplace, bringing together classes and students. Marketplaces are everywhere today, in almost every industry – for example, eBay, Amazon, Etsy, The Lending Club, Streeteasy. A traditional business creates a product and sells it directly to a customer. A marketplace brings together two parties to make a transaction.
We chose to go after students, because we feel there is a more of a valuable prop there (finding the right class right now involves hunting through Google and different company websites). And if we get the students, the schools will follow naturally. So we are going to continue with the “Wizard of Oz” approach to populating the classes manually and focus our tests on improving the student-customer experience.
We are going to organize our tests according to Lean Startup principles, by putting them into the following categories.
|Idea we want to test||Key Assumption||Test applied||Method of Testing||Key Metrics (what is success?)||Results of Test|
|Reserving a course on LBC with credit card||Visitors will use LBC to directly book the class, instead of booking with the school||Wizard of Oz||Create form to input cc, forward reservation to school to process manually||% of users who submit form > 10%|
(Note: For inspiration, consider Hotwire or Hotels, where customers book the hotel with site who then passes on the reservation to the hotel. The customer only needs to enter credit card info once to book at all hotels)
This is the first chance I’ve had to go really direct on the lean process with a really cool startup, so I’m looking forward to working with CoursAvenue. Here we go!
Assortment of sites I find inspiring
I think the concept of Lean Startup, at the heart testing ideas with cheap models, makes a lot of sense. However, Lean can’t tell you what to test or how to test, only that you should test. The creativity and vision of the founders is needed to figure out what to do next. Here are the websites we examined to find inspiration.
37signals – video of home page iterations, simple signup
Meetup – how they ask for more information step-by-step during the signup process so the customer isn’t overwhelmed by blank boxes
Proflowers – let you buy as a guest (without signing in) and then ask you if you want to save info for next time and with a password. More on this
Google Analytics – Visualize what a funnel looks like (especially visitors falling out of one)
ColSpace – Quick, entertaining, and explanatory homepage intro video
Screenflow – To make simple explanatory videos (and examples of those videos)
Classtivity – Startup in the same space, 10 pass innovation, need to make going to a class sticky to be valuable to the instructor, reserving class on the site needs to be sticky to bring people back to LBC
Unbounce – for thinking about landing pages