Steve Blank: New Lessons Learned From Berkeley and Stanford Lean LaunchPad Classes

What the entrepreneur and his team gleaned from introducing students to the lean startup model

Our Stanford and Berkeley Lean LaunchPad classes are over for the year, and as usual we learned as much from teaching the teams as the teams did from us.

Here are a few of the lessons learned from the two classes.

Have each team talk to 10 customers before the class starts

Each year, we learn how to move more of the Lean LaunchPad class logistics outside our classroom so teams have more time for in-class learning.

A few years ago, we moved the formation of teams to before the class started, and in doing so, saved a week of what normally would have been class time. To make this happen, we held three information sessions two weeks apart before the class started. In these info sessions, we described the purpose of the class and then let students mix, meet, and form teams. During this preclass time, we shared a Google Doc in which students who had ideas could find other team members. Students without an idea could find a team that matched their skills and interests. Application and admission into the class was by interview with a fully formed team.

Info session announcement
Info session announcement (click to enlarge)


The next thing we learned was to make applying to the class an integral part of the learning process. Teams applied by filling out both a business model canvas and a competitive petal slide. Having the teams do this accomplished three things. First, it forced the students to read and understand "what's a business model canvas" before they even came to class.

Freewire application

Team application: business model canvas

Second, the competitive slide enforced a modicum of due diligence on the product and market. We got tired of knowing more about a team's market by doing a Google search as the team presented--so we made it a team's job.

Farmsense competive slide

Team application: competitive petal slide (click to enlarge)

Finally, having teams spend time on the canvas and competition as part of the application process saved weeks of what would have been class time (and as a bonus gave the team a heads-up about the difficulty of the class and showed whether the team members were serious about the class or just shopping).

This year, we learned to raise the bar once again. Could we get the teams to come into class having already talked to 10 customers? Instead of using the first class to have teams just present their business model canvas, this time the teams' first presentation was about what they learned outside the building about their value proposition. We pointed them to our tutorials on customer discovery and how to conduct customer interviews but didn't expect them to be experts in Week One.

SignUP week 1

1st-week team title slide--11 interviews before class started (click to enlarge)

We did an A/B test when we required our teams in one school to do this while we didn't require it for the teams in the other school. The result? Teams that had to talk to customers before the class hit the ground running. There was a substantive difference in team trajectory and velocity that continued throughout the quarter. The amount of learning between the two felt quite different. Though there may have been other factors (team selection bias, team makeup, etc.), we'll now make this an integral part of all the classes.

Have each team put the number of mentor interactions on its weekly title slide

The second innovation this year involved mentors. Each team was assigned a mentor as a coach. We've been trying to figure out how to make mentor engagements with their teams a regular rather an ad hoc activity. Though we have required the teams to add a summary of any mentor interaction to their LaunchPad Central narrative, we felt we didn't have sufficient high-level visibility for these essential interactions.

GiveModo Class 8

(Click to enlarge)

But this year, a seemingly minor change to the teams' weekly cover slide had an important impact. As teams presented each week, their cover slides showed the number of customers interviewed for that week (>10) along with the cumulative customers interviewed. This year, we added one more metric for their cover slides--the number of mentor interactions for that week (>1) along with the cumulative number of mentor interactions.

This enhanced the visibility of the teams' interactions (or lack of) with their mentors and allowed us to intervene early if there wasn't sufficient interaction.

Here are a few of the Final Presentations (see here for all of them)

If you can't see the presentation above, click here

If you can't see the video above, click here

If you can't see the presentation above, click here

If you can't see the video above, click here

If you can't see the presentation above, click here

If you can't see the video above, click here

If you can't see the presentation above, click here

If you can't see the video above, click here

If you can't see the presentation above, click here

If you can't see the video above, click here

If you can't see the presentation above, click here

IMAGE: Corbis
Last updated: Apr 17, 2014

STEVE BLANK | Columnist | Founder, E.piphany, Convergent & Zilog

Steve Blank is a retired Silicon Valley serial entrepreneur turned educator who developed the Customer Development methodology that changes the way startups are built. His book The Four Steps to the Epiphany launched the Lean Startup movement.

The opinions expressed here by columnists are their own, not those of

Register on today to get full access to:
All articles  |  Magazine archives | Livestream events | Comments

Or sign up using: