Lesson: Customer Discovery Primer with Steve Blank

Step #4 Customer Start: Start with customer development and agile engineering

What we now know is most startups fail from a lack of customers than from a failure of product development and that's really interesting. Even in Silicon Valley where we would probably take more technical risk per square inch or square foot than anywhere else in the world. We go out of business typically not because we didn't deliver a product. We go out of business, because we didn't find enough customers to pay us enough money. That's a big idea, because if you think about it, we have all these processes to manage the engineering process, why? Because that is the biggest risk in a large company, but we had no processes to manage the customer risk. In fact, the best we could do, in the old days, is hire a VP of sales who told you "Oh, I have a great Rolodex, great set of contacts." But we really have no formal process for searching for what customers needs are, and in fact, we've just come up with one and that's going to be a key part of this course. This search for the business model. This search for the truth is called customer development, and customer development really has two pieces, the search and its execution, and in this class, we're going to be talking about the first two steps, customer discovery, customer validation, the pivots that connect them, and then if you are so lucky, we actually get to execute your business model in customer creation and company building.

The other piece that goes with the process in a startup is not in a customer development process but in agile engineering process. One example is extreme programing or XP, which basically is built around this idea of iteration and incremental delivery of the product. This is a big idea; it says instead of building every possible feature on day one, we're actually going to incrementally and iteratively interact with customers, test each portion of the product, and see if what we're building actually has a home outside the building. So what this really means is that for process instead of just going to execution first, hiring product managers, doing waterfall development, or maybe even agile nowadays, we're actually going to start with a customer development process coupled with some kind of agile engineering process. Which one you pick? Iuse XP as an example, but it could be whatever your favorite process is, and by the way, don't only think that agile engineering and agile development is about software. Toyota and the Toyota production system actually had it for a hardware of decade before anybody in software ever thought about it. So you could be building products anything from microprocessors to medical equipment to hardware, software, etc. using an iterative and incremental engineering process. And again, once you're done with that, of course, you want to manage this in a formal process, but not before you do the planning.