My manager (a director) at a FAANG used to give me (and only me, justified with my strong background - though he used similar tactics on others) tasks that couldn't be solved and didn't have any business justification so he could keep me from getting promoted. The first few promotion cycles I still tried to find alternatives (X doesn't make sense, but Y would reach similar goals in a reasonable way), but the motives were quite clear.
At first I was looking for ways to move within the company (I was hired in the wrong part of the organization and on the wrong level, making this a bit difficult) and started working with other teams to be able to switch, but I realized it wasn't really worth it because most teams were building things that were very strongly driven by VPs and directors own motives rather than users or even what is good for the business. [1]
My impression was that this is partially driven by several factors:
- the very strong market position means customers don't really have a choice. It's actually rather hard to make good decisions when the market can't give you good signals. Situations would arise where one customer wants something, another wants the opposite and both may have a negative short-term impact on the bottom line. The lack of competition to really verify which decision would be successful gives too much power to the decision maker with really bad incentives, leading to some suboptimal decisions (unless there's a strong long-term strategy that is followed through).
- the drive to make "data driven" decisions, without understanding what that really means. Depending on how the data was presented, completely different conclusions were possible and instead of using common sense with the data, this was treated as an absolute truth to end many discussions. What quite frequently would happen was optimizing towards the wrong metric or underestimating bias.
- Silos created by people who were hired at a different stage of the company. Some of the managers just hadn't grown with the company - people who were around for 10+ years were hired for much smaller revenue und user numbers and this further increased the problems with the two points above. My impression was that the incentives weren't well aligned with overall goals of the specific product.
I think the solution in the article above would be a good way to avoid some of the problems I've encountered, by creating a more direct feedback loop between the product and the sales side and aligning incentives better. I don't think it's the best way how many companies grow with very diverse sets of offerings to have huge sales and engineerings orgs, rather than smaller groups that cover the product end-to-end. It's still possible to have many of the benefits of being a large enterprise (shared talent pool, operational processes, costs etc.), while avoiding the drawbacks.
[1]: I can't share evidence, but this particular company is well known for discontinuing popular services.