Eric Ries rejects the notion that the job of a startup is to execute a business plan. He proposes a different mindset and some tactics to implement it.

The basic point of the book is that when a new venture is launched, it is highly unlikely that the people behind it really know what their customers want or how they would perceive and use the service or product.

Therefore the aim of a startup is first to figure out what the right things to do are and only then roll out full scale. In order to do this, the startup needs to develop validated learning.

Validated learning comes from testing each hypothesis

Every startup or business plan has certain assumptions which no one really knows whether they hold true or not. Two types of assumptions are essential for the success of a business: value hypothesis and growth hypothesis.

The value hypothesis refers to what the clients will perceive to be valuable and want to pay for versus what they find useless and is only a drain of resources for the company. The growth hypothesis refers to how people will discover the new offer of the company and how much effort this discovery will require from the company.

The startup needs to test these assumptions as quickly as possible because in most cases it will need to modify them and tweak them so that resources can be spent wisely.

MVPs are an efficient way to test a hypothesis

In order to test these, Eric Ries proposes that only a minimum viable product is built initially and then tested in real market situations if possible. Then the product is refined and tested again, feature by feature.

How complex this minimum viable product should be, depends on the situation and entrepreneurs should use their common sense to judge that.

However, in order to perform a valuable test and achieve validated learning, it is important to formulate the hypothesis that needs to be tested clearly. This will also show what needs to be measured in order to ascertain whether the hypothesis has been falsified or not.

Validated learning is the basis for decisions

The purpose of the MVP is not only to test the assumptions of a startup, but to test them as early as possible and at the lowest cost. Anything that is included in the MVP that goes beyond hypothesis testing is waste.

Therefore the key activity of a startup is to apply the build-measure-learn loop as quickly as possible to develop the insights that will allow for a scaling of the business and the growth that is dreamt of.

However, this build-measure-learn loop can challenge the startup and may show that it needs to change its approach completely, in other words pivot. An important metric for a startup according to Ries is the number of pivots a startup can still perform before the money runs out, because this says something about the chances of the startup being successful even if the initial plan was not viable.

But in order to be able to pivot quickly and correctly validated learning is necessary and therefore this should be the key measure of progress for a startup, Ries argues.

summary by Denes Lenard – Webfoo