4 ms·

When things are growing, any nonsense is thought to be "the way". OKRs, matrixes, oh they must be so smart with all those "systems thinking". Somehow none of th

When things are growing, any nonsense is thought to be "the way". OKRs, matrixes, oh they must be so smart with all those "systems thinking". Somehow none of those systems work when things are going down. Look at Google and Intel today, none of those OKRs saved them did it? It's the same thing with CEOs. I don't remember which book it was, but it showed how the CEO changing didn't matter at all but was just natural market cycles.

mjburgess 3y ago
In many cases leadership are best when they're effect-neutral on the performance of a company/project/etc. Consider coaches of two sports team: if they're both equally good, their effects wont be noticeable from the performance of either team. In a competitive environment, leadership ought wash-out if its any good. It's only bad leadership which will be noticed.
apienx 3y ago
> When things are growing, any nonsense is thought to be "the way". Modeling reality properly is hard. Cargo cults are less cognitively expensive to establish.
- joquarky 3y ago
  As I get into my late 40s, all I see is cargo cults anymore. Not just in tech, but everywhere. I thought we had people who knew what they are doing, but it feels like everything is being run by silver-tongued conmen now.
roenxi 3y ago
I think "OKRs" exemplify this best because after reading the wiki page I'm not sure that they exist. As pointed out in the article, it seems to be a term that describes ... nothing in particular. Every serious industry since at least the days of Toyota's rise has used metrics. I don't think the tech industry has ever shown me something that has improved on what traditional industries use. The only change is the underlying processes of the tech industry are too chaotic to harness (what management process gives us Google? Given that we have literally 1 Google do we have enough evidence to reject it being built with a different management style? No. On the other hand Toyota could only exist with Toyota's management style.) and so any amount of management quickly becomes futile. Time will pass, we'll hit the boring part of the S curve in processor improvement and then management culture will start to be a factor. And the industry will be a lot more boring, less growth focused and care a bit more about consistency and reliability.

mjburgess 3y ago

In many cases leadership are best when they're effect-neutral on the performance of a company/project/etc. Consider coaches of two sports team: if they're both equally good, their effects wont be noticeable from the performance of either team. In a competitive environment, leadership ought wash-out if its any good. It's only bad leadership which will be noticed.

apienx 3y ago

> When things are growing, any nonsense is thought to be "the way". Modeling reality properly is hard. Cargo cults are less cognitively expensive to establish.

joquarky 3y ago
As I get into my late 40s, all I see is cargo cults anymore. Not just in tech, but everywhere. I thought we had people who knew what they are doing, but it feels like everything is being run by silver-tongued conmen now.

roenxi 3y ago

I think "OKRs" exemplify this best because after reading the wiki page I'm not sure that they exist. As pointed out in the article, it seems to be a term that describes ... nothing in particular. Every serious industry since at least the days of Toyota's rise has used metrics. I don't think the tech industry has ever shown me something that has improved on what traditional industries use. The only change is the underlying processes of the tech industry are too chaotic to harness (what management process gives us Google? Given that we have literally 1 Google do we have enough evidence to reject it being built with a different management style? No. On the other hand Toyota could only exist with Toyota's management style.) and so any amount of management quickly becomes futile. Time will pass, we'll hit the boring part of the S curve in processor improvement and then management culture will start to be a factor. And the industry will be a lot more boring, less growth focused and care a bit more about consistency and reliability.