This is the fourth blog in my anticipated 13 blog series dedicated to each chapter of Eric Ries' seminal book The Lean Startup (2011).

The the third chapter is titled "Learn" and the main focus is to identify "validated learning" as the key dimension that startups have to be productive at in the beginning.

Imagine that you have come up with a brilliant vision for a new software product, given yourself a 6 month timeline for launch, financed the development through investors, and after 6 months worth of long days you are ready to launch. It is then that you realize no one is willing to download your product or pay for it. You then make lots of changes to see if you can get customers to download and pay for it but your earnings peak at a measly $500 per month for many months until you start pivoting from a major aspect of your initial vision and start focusing on what your customers really value. That is the story of the eventually highly successful company, IMVU, that Eric helped found in the capacity of product designer and software engineer.

Eric was schooled in the bible of lean manufacturing, lean production, agile development and lean thinking so wasting so much time and effort in developing features and software no one wanted was deeply embarrassing. The concept of the "lean startup" emerged from this deep sense of frustration:

Anything we had done during those months that did not contribute to our learning was a form of waste. Would it have been possible to learn the same things with less effort? Clearly, the answer is yes .... As the head of product development, I though my job was to ensure the timely delivery of high-quality products and features. But if many of those features were a waste of time, what should I be doing instead? How could we avoid this waste? ~p. 48-49

Fortunately, Eric had a scientific mindset and studied his users and conducted many experiments to figure out what was working and what wasn't. He also had enough investor-financed runway to recover from his initial misjudgement where many might have gone under. Perhaps setting a 6 month launch date gave him enough time to recover and figure things out for the next 6 months. I suspect the investors weren't in it just for a 6 month experiment.

When the company started firing on all cylinders was when they starting determining what was of value to their customers and what was not. This was done through numerous changes to the product, to their website, to the value proposition to customers, to branding and then measuring whether the existing versus changed versions produced different quantitatively (or qualitatively) different results. The product that a customer eventually consumes is not just the product itself but also includes the way it is branded, marketed, and distributed into the marketplace. These have to be tested as well. If, however, the "product" is not something that customers value then they won't remain customers so validated learning should be first and foremost focused on defining what product customers will value. It should be noted, however, that when Eric changed the name of their product from "avatar chat" to "3D instant chat" signups and paying customers increased so what defines your "product" is not just the software coding and UI design.

Eric doesn't explicitly define what validated learning is in this chapter but it is easy to get a sense of what it is from this passage:

Positive changes in metrics became the quantitative validation that our learning was real. This was critically important because we could show our stakeholders - employees, investors, and ourselves - that we were making genuine progress, not deluding ourselves. It is also the right way to think about productivity in a startup: not in terms of how much stuff we are building but in terms of how much validated learning we're getting for our efforts .... This is true startup productivity, systematically figuring out the right things to build. ~ p. 51-52

We see here that the definition of validated learning is linked to "quantitative validation" using "metrics" that inform us about "the right things to build". It involves forming hypothesis about what might improve the value of your product, thinking about changes that might be made to test whether it does, and then measuring whether the difference makes a difference. This is where the learn startup should focus most of its initial time and resources, not on extensive planning exercises based on preconceptions about the expected value the product.

Keep in mind that we know quite a bit about the expected value of some products in the marketplace (e.g., coffee, housing, oil wells, etc...) but startups are engaged in creating innovative products that don't currently exist in the marketplace so the best investment of time and resources early on is to validate assumptions about product value as early as you can in whatever form you can.