4 ms·

What I am not seeing in this discussion is any calculations showing how much of these outcomes are down to dumb luck, sample bias and correlation. For example:

What I am not seeing in this discussion is any calculations showing how much of these outcomes are down to dumb luck, sample bias and correlation. For example: > Furthermore, research using our CEO database found that the top decile of high performing CEOs are 35 percent more likely to dynamically reallocate capital than average performers. Maybe they stumbled into an opportunity that the rest didn't. That's luck. > To move “boldly” is to shift at least 30 percent more than the industry median. Making one or two bold moves more than doubles the likelihood of rising from the middle quintiles of economic profit to the top quintile, and making three or more bold moves makes such a rise six times more likely. So if your random walk takes you upwards, you wind up having walked upwards. Brilliant. > In the largest research effort of its kind, McKinsey found that CEOs who insist on rigorously measuring and managing all cultural elements that drive performance more than double the odds that their strategies will be executed. Measured number rises, unmeasured number remains unmeasured. Scientists baffled. Remember: these are the best and the brightest. They are taken extremely seriously. Edit: they do give a source for their 45%-explained-by-CEO claim ... and it's a pop-business book written by the same authors. Oh well.

jacobolus 7y ago
Considering McKinsey’s business model (“hey CEOs, if you hire us we’ll make you excellent”), there’s nothing surprising in the content of this article (“coincidentally, the vague list of things that we consider to make CEOs excellent is the same as the list of things that we have experience promoting”).
shantly 7y ago
I've seen nothing to convince me that most C-level management "experts" care about confounders, selection bias or, generally, the quality of any data they come up with. Nor have I seen evidence that others in a position to call them on it care to. They either do not care, or do not know, how to reasonably use data to support a point or to make a decision. To include folks from the big-3 management consultants. Though to be fair, it's not like people down the org chart are usually much better at it.
- bjornsing 7y ago
  I can second this. Everybody wants to be “data driven”, but hardly anybody wants to adhere to the rigorous standards necessary for sound data analysis. I think it’s a bit unfair to point the finger at “C-level management” in particular though. The same kind of thinking can be observed in government agencies, in the legal system, even in academia. Statistics is just hard, and a lot of people are simply not interested and cognitively capable enough to get it right, sadly.
  - gatherhunterer 7y ago
    > a lot of people are simply not interested and cognitively capable enough to get it right, sadly. This is how executives discredit engineers. Someone who believes themself to be intellectually superior to others and is socially inept enough to say so is an easy target to discredit. This is how you become a code monkey.
    - bjornsing 7y ago
      I don’t know, is it? I used to be a “code monkey”. To be honest I think the main reason I ended up in that position was that I saw myself as socially inept and thought that there was some kind of “big picture” that the mgmt/power focused people saw and I couldn’t see. I don’t any more, and my latest title is C-level (for a 100 person startup though - so nothing fancy). People love to set social norms that limit how much use others can make of their talents and abilities. But that’s usually not in the company’s best interest, and those norms seldom have the backing of senior management. With that said, of course you should be tactful and show your coworkers respect. People deserve respect not because they are smart, but because they are people.
  - zipwitch 7y ago
    One piece of data is that there's an inverse correlation between CEO pay and company performance: https://cooleypubco.com/2016/07/25/new-study-shows-inverse-correlation-between-ceo-pay-and-performance-over-the-long-term/ https://cooleypubco.com/2016/07/25/new-study-shows-inverse-c...

jacobolus 7y ago

Considering McKinsey’s business model (“hey CEOs, if you hire us we’ll make you excellent”), there’s nothing surprising in the content of this article (“coincidentally, the vague list of things that we consider to make CEOs excellent is the same as the list of things that we have experience promoting”).

shantly 7y ago

I've seen nothing to convince me that most C-level management "experts" care about confounders, selection bias or, generally, the quality of any data they come up with. Nor have I seen evidence that others in a position to call them on it care to. They either do not care, or do not know, how to reasonably use data to support a point or to make a decision. To include folks from the big-3 management consultants. Though to be fair, it's not like people down the org chart are usually much better at it.

bjornsing 7y ago
I can second this. Everybody wants to be “data driven”, but hardly anybody wants to adhere to the rigorous standards necessary for sound data analysis. I think it’s a bit unfair to point the finger at “C-level management” in particular though. The same kind of thinking can be observed in government agencies, in the legal system, even in academia. Statistics is just hard, and a lot of people are simply not interested and cognitively capable enough to get it right, sadly.
- gatherhunterer 7y ago
  > a lot of people are simply not interested and cognitively capable enough to get it right, sadly. This is how executives discredit engineers. Someone who believes themself to be intellectually superior to others and is socially inept enough to say so is an easy target to discredit. This is how you become a code monkey.
  - bjornsing 7y ago
    I don’t know, is it? I used to be a “code monkey”. To be honest I think the main reason I ended up in that position was that I saw myself as socially inept and thought that there was some kind of “big picture” that the mgmt/power focused people saw and I couldn’t see. I don’t any more, and my latest title is C-level (for a 100 person startup though - so nothing fancy). People love to set social norms that limit how much use others can make of their talents and abilities. But that’s usually not in the company’s best interest, and those norms seldom have the backing of senior management. With that said, of course you should be tactful and show your coworkers respect. People deserve respect not because they are smart, but because they are people.
- zipwitch 7y ago
  One piece of data is that there's an inverse correlation between CEO pay and company performance: https://cooleypubco.com/2016/07/25/new-study-shows-inverse-correlation-between-ceo-pay-and-performance-over-the-long-term/ https://cooleypubco.com/2016/07/25/new-study-shows-inverse-c...