Eric (32) is a brilliant and inspirational speaker. He had his own successful start-up and then spent a lot of time thinking about why it was successful and following a scientific process of hypothesis – experiment – measurement – analysis. He has published a book “The Lean Startup“.
A point that Eric made that resonated with me was that many British entrepreneurs’ “Minimum Viable Product” (MVP) was too big. The goal of the MVP should be to create a testable hypothesis about a feature and test it in the market.
When we created Keyapt the goal of our MVP was to create something that was competitive against rival technologies that users could use to execute the same use cases. This goal contained the hidden assumption that we were qualified to decide whether our solution was competitive. Eric’s point was that whenever you launch your MVP you will always find that user’s don’t like it. Therefore it is best to launch very early and get real feedback.
The risk of launching early:
- upsetting potential customers
- alerting competition (my point – not Ries’)
The risk of building a more feature complete MVP:
- time during which you lose market share
- money spent building features no-one wants
- building in wrong-think that then has to be taken out at great expense
Ries noted that the risk of upsetting potential customers can be mitigated by positive engagement with them.
The lean start-up believes a smaller MVP is better and that “Very few start-ups fail for lack of technology. They almost always fail for lack of customers.” Therefore concentrating your efforts on finding out what customers want should be the priority.
I note that technologists can take comfort in the fact that this can be done with analytics software and you don’t actually have to “talk” to people. This philosophy can be extended to daily builds and A/B testing. In this way the product can “walk uphill” in very small steps. Wired has a great article on this here.
Benchmark numbers (from Ries and others)
- Your first aim is to get real feedback from 10 real users
The highlight of the meeting was that Eric pointed out that there are three engines for growth
- paid growth
- You advertise and a customer costs you $ x but generates $ X revenue where X > x and so you can afford more advertising (eBay)
- viral growth
- The nature of the product means that each customer acquired will result in acquiring another customer (Facebook)
- Once acquired customers never leave (World of Warcraft – or your character will DIE!)
- A metric used by some is that 40% of your users should be “very upset” if you took the product away from them.
If your successful business needs help with a lean start-up, lean project management for growth or agile software development please contact me for a chat.