I kind of agree with the initial precept. I have seen useless PMs more than I can count, and organizations turned into government bureaucracies on the altar of

I kind of agree with the initial precept. I have seen useless PMs more than I can count, and organizations turned into government bureaucracies on the altar of cargo cultish "best practice" processes. Retro-engineering okrs from a pre-existing list of features, looking at data only for vanity purposes (how good are we). I think it goes one step too far and throws the baby, the bath tub, and the whole house with the bathwater. > Do you know what else they had in common? None of them were motivated by OKRs, KPIs, data, or other people’s opinion. This is not true for Bezos for example. He was famously driving the company through data, metrics and process, and there's countless examples of "Jeff meetings" where he changed his mind after reading the data presented by others. All those examples are tools, that you can, and often shouldn't use. It makes no sense to have a heavy process when you're small, in direct touch with the customer, and are fully driven by an idea. Similarly, it makes no sense keeping on wet-fingering it and single-person driving once you have a set of established customers and want to increase your penetration. Same is true for all aspects of a company, from engineering to sales to marketing. As you grow you get more people working, and you need a minimum of coordination and standardization to avoid going in antagonistic positions all over. And same goes with seems between these functions. Let me tell you that I prefer having a good PM to handle these needs and that coordination rather than having to deal with synchronizing their needs as an engineer. I went in there hoping I would have a good resource to share about the topic, but it's too sour and mono-directed into some weirdly "aw common just be smart about it" to be credible. Ironically, the conclusion of the article (no structure, just smart people everywhere) contradicts one of its central argument (everything doesn't apply everywhere).