Here is the first thing I said to my group of 5 teams when I started the first of 20 workshops for the summer.
You are CEOs. I’m a guide. You are not working for me. You are working for yourself. If you don’t feel like something is helping your business succeed, tell me. This accelerator is for you and I want to make it as useful and effective as it can be
Getting into startup mode
The goal of the first class is to get the teams into “startup” mode. That means out of school mode, where they are supposed to be giving the right answer. That means out of planning mode, where they focus on writing business plans and whiteboarding ideas. And it means focusing exclusively on testing their product ideas with real customers. In short, customer development.
Programming for startup accelerators
My goal in writing this blog over the next ten weeks is to have a blueprint that anyone can follow to run a startup accelerator. I’m still working through a lot of the ideas myself, so we will see how it turns out. A big challenge I’m working on is how to present all the information I want to the teams to learn as exercises rather than lectures.
The teams are going to submit a weekly blog entry, that will cover:
Hypothesis: Here’s what we thought about our value prop
Experiment: Here’s what we did to test that thought
Results: Here’s what we found out
Action: Here’s what we want to do next
Here’s the link to the Week 0 entries
Setting the table
I started with a discussion of the Steve Blank udacity videos. I asked – what did you get from the videos?
The key topics I wanted to cover were:
- search vs execution
- model vs plan
- business model canvas
And the following terminology:
- customer development
Drawing your business model
After a brief discussion, we launched into our first exercise, drawing your business model. A key challenge in drafting a business model, that Jim Wheeler and I have discussed, is that the business model canvas is not intuitive. We suggest having teams draw out their business model, thinking about what they need to buy to make their product and who they will sell it to. Here’s an example of what a drawn business model would look like for the All-American lemonade stand.
After each team drew their business mode, I went over how to translate the drawing to the business model canvas. Key is to follow the money!
The primary purpose, for me, of the business model canvas at this point is to list out the startup’s assumptions. Common one for early stage startups are 1) people want to buy/use what we are selling and 2) we can actually make/produce the product. Once teams can identify critical assumptions, they know what to tackle first. I emphasized testing/validating the critical stuff, not the easy-to-test stuff.
Presenting the business model
Then each group presented their business model, in words, to the class. And the students offered feedback. The goal of this exercise is to get teams in the habit of practicing their pitching/explaining of their business – can’t do that enough! And to get the other people in room thinking critically about other teams’ business models. It is often much simpler to look at someone else’s work with a critical eye, and through the critique, build up the analytical muscles that will serve them well throughout the startup process.
Then we dived into my favorite topic – how will we test!? Who do we need to talk to? The teams mapped out their strategy for week 1, with the understanding that Class 2 of Week 1 would focus on lean startup methodology and developing the experiment they would run in week 1.
I had the teams choose a tool to share their results with me – ie: dropbox, evernote, or google drive. Steve Blank offers lean launchpad software but I didn’t want to add the overhead. I prefer the teams use whatever tool they are comfortable with to track their progress and I’ll adapt. I chose this route because I heard from one student that they didn’t choose to continue using Lean Launchpad after they didn’t have to for class, so I figured it wasn’t an effective tool for actually managing a startup’s growth.
I’m following the Y Combinator model of requiring scheduling of office hours to talk with me. I shared my calendar with this bit.ly link – http://bit.ly/e2mcalendar .