The problem with these stories is that they assume that correlation is causation: they pick some subset of all the things that influenced Amazon and present the

The problem with these stories is that they assume that correlation is causation: they pick some subset of all the things that influenced Amazon and present them as the decisive influences, without a evidence to support that suggestion. I humbly suggest that many people that share this worldview and attempted to execute based on it have failed. This advice is as worthwhile as 'lean startup advice': there are no guarantees, it depends on many factors and most will fail.