product market fit with biased data
One of the toughest things to understand when creating a product is the Product-Market fit. In the problem-solution cycle we have many tools to understand if the problem we identify is common to other people and if they see a value in our proposed solution. However, when building a product, we have to make choices that depart from the ideal scenario. Product-market fit is therefore when we achieve a status in which what we have developed is what the market wants, even if not ideal.
The lean methodology suggests that we should learn from experiments and iterate quickly in order to improve our product. However, this experiments can be done on biased samples, for example testing a product with managers and not with the employees who will actually use it.
This can be even worse if we have gender biased data (see: 202011110957 and 202011110941). Therefore we can think that we have a product that satisfies the demand, but based an skewed insights and prejudices.
Therefore it is important to develop design procedures and business analysis tools that take into account not only gender as a random sample of the population, but actively considering that roles and physiology of women are different (see: 202011110929 and 202011110920).
These are the other notes that link to this one.