OKRs just seem like another attempt in a long line of failed attempts to make development costs more predictable. I applaud the effort, but at any company I've

OKRs just seem like another attempt in a long line of failed attempts to make development costs more predictable. I applaud the effort, but at any company I've worked, they've always turned into a stick used to beat on the underlings. I think I get how they could work in an ideal scenario, but the implementation of them has always failed, because the adopted process doesn't allow them to shift even as things arise outside your control, because they got tied to raises/bonuses in unfair ways, etc. My only "successful" encounter with OKRs was at a company that really bought into them hard, and so over the course of about half a year I got the teams in my group to slowly get ahead of the curve, such that 90+% of every next-quarter OKR items were things we were done with already, so that we could spend pretty much the whole quarter dealing with fires or secretly getting new stuff done for the next-next quarter.