Simon Wardley’s Value Chain Mapping method has mostly colonized my brain for build vs. buy decisions. The advice in this article doesn’t much address the most important consideration: whether a thing is known or unknown, to-be-invented or a commodity. If you’re familiar with Wardley’s maps, a decision tends to be pretty easy. Build stuff for the upper-left corner, buy stuff for the lower-right. Avoid the lower-left and upper-right corners entirely.
Sounds a bit too optimistic. While admitting that there should be reevaluation, the list of recommended criteria doesn’t include estimating the cost of migration away from the solution. A pessimist would also evaluate the risk of a forced migration, for example when a product suddenly stops being available. And there is not only the question of the product you choose supporting the future change you need; there is also a question whether a product is likely to force changes onto you due to its own evolution.