3 ms·
OKRs are bullshit because they rot the entire willpower-based culture of an organization and replace it with a mindless metric-optimizing machine. People should
by ajkjk 3y ago
OKRs are bullshit because they rot the entire willpower-based culture of an organization and replace it with a mindless metric-optimizing machine. People should be judged by their ability to do a good job at their jobs, not by their ability to hit stupid metrics at the expense of everything else.
As far as I can tell being "data-driven" is completely toxic to long-term organizational health.
- vkazanov 3y agoAnd how should one measure the "good at their jobs" metric?
- ajkjk 3y agoBy one's intelligent opinions. You hired them, gauge to the best of your ability if they're doing the job you hired them to do, at a level you deem acceptable. Your boss should measure you the same way.
- vkazanov 3y agoThis makes sense in a small company context when, say, a competent CTO/CEO personally knows everyone and can resolve conflicts, bottlenecks, disfunctions quickly. But what if a company in question is a relatively big, 1000s of employees, and something is clearly wrong in the org. Both managers and employees are friends, seemingly competent, have all the explanations at hand and all the right excuses. And nothing gets done.
- willsmith72 3y agoi don't even think it does make sense for small companies, they can benefit hugely from setting clear, public commitments about what they're working towards focus is one of the most important characteristics in a startup, because time is too short to waste.
- vkazanov 3y agoI've seen this direct CTO work scale to a billion-a-year online business with hundreds of engineers :-) Problems come when a company goes public and now there's no single personal that can accept responsibility for technical commitments. This when a decision-making system has to be put in place, and maybe a way to assess how teams are doing, etc.
- ajmurmann 3y agoHonestly, I think the best you can get is having a competent manager who understands the work and people well and gets the need autonomy and trust from the company. That's hard though and there is no magic wand for the CEO to make all managers "good managers". Maybe we keep introducing these approaches as an industry because it makes the worst cases less bad, even though it hurts teams and individuals doing well already because ultimately pushing the floor up makes a bigger difference. Not sure if that overall change is objectively better in total or only seems that way because the horrifically bad teams are what catches your attention.
- Marazan 3y ago"You can't manage what you can't measure" is the _fundamental_ fallacy of management. Management involves soft skills and subjective judgement. You cannot eliminate that and pretending that you can kills companies.
- intellectronica 3y agoYou are fighting a straw man. Of course data-driven management or any kind of planning, at the extremes, has some awful manifestations. That's just bad application of these tools. Bad management. In many successful orgs people use their "willpower" and motivation and skill and originality to do work that is informed by data and follows strategic plans, agreed on in OKRs or similar devices.
- ajkjk 3y agoWell I'm fighting the strawman of 'my last job' so it is definitely at least real. But my impression is that it's everywhere in tech.
- intellectronica 3y agoNo, definitely not everywhere. I had some bad experience that echo yours, as well as quite a few mostly positive ones.
- koliber 3y agoIf the metric is stupid, then you're right that it is a waste of time to hit it. Part of the OKR process is defining meaningful goals (Objectives) and ways of measuring progress towards them (Key Results). If you do this part decently, the metrics won't be stupid. Once you have metrics that are not stupid and your team agrees that they will advance you towards your goal, it's kind of smart to do the work to move the metric. Not at the expense of everything else, but if this goal is deemed a priority, then it's OK to put it ahead of many other things. Being blindly data-driven can be toxic. Not measuring anything means you might be flying blindly. I've seen it first hand. There is a lot of value in combining objective measurements with sensible management.