6 ms·

> o Joseph Lampel and I studied the post-1990 records of all 19. How did they do? In a word, badly. A majority, 10, seemed clearly to have failed, meaning that

> o Joseph Lampel and I studied the post-1990 records of all 19. How did they do? In a word, badly. A majority, 10, seemed clearly to have failed, meaning that their company went bankrupt, they were forced out of the CEO chair, a major merger backfired, and so on. The performance of another 4 we found to be questionable. So 70% failed, 30% succeeded. Wasn't the average that 90% of businesses (perhaps more specifically startups) fail within the first 3 years? So if anything, wouldn't this show that Harvard MBA grads do better than the average CEO?

Chris2048 10y ago
Yep. motherfuckers act like they forgot about Bayes.
sacheendra 10y ago
I think the 2 datapoints are different. More than 90% of startups fail, yes. But, in the above scenario, these people ran established companies into the ground.
hyperpape 10y ago
I had the same concern about not having a clear comparison to non-MBA results, but the second study mentioned did have a clear comparison.
bluGill 10y ago
These were successful companies at one time. What is the percentage of Fortune 500 companies three years ago that have failed - I don't know, but it isn't 90% (it probably isn't hard to look this up if someone wants to fact check me - for extra credit find other measures companies of different sizes that are not startups). Of course bankruptcy is only part of failure. So really it is what percentages of Fortune 500 (or other measure of successful companies) have "their company went bankrupt, they were forced out of the CEO chair, a major merger backfired, and so on" Startups are in fact the wrong measure. A startup has nothing to lose, and often is trying something that will not succeed. Large companies often fund startups to limit their potential loss if it fails (if things seem not be going well they pull out without having to lay off all the people working for the startup which looks bad), and they buy the company if things go well.

Chris2048 10y ago

Yep. motherfuckers act like they forgot about Bayes.

sacheendra 10y ago

I think the 2 datapoints are different. More than 90% of startups fail, yes. But, in the above scenario, these people ran established companies into the ground.

hyperpape 10y ago

I had the same concern about not having a clear comparison to non-MBA results, but the second study mentioned did have a clear comparison.

bluGill 10y ago

These were successful companies at one time. What is the percentage of Fortune 500 companies three years ago that have failed - I don't know, but it isn't 90% (it probably isn't hard to look this up if someone wants to fact check me - for extra credit find other measures companies of different sizes that are not startups). Of course bankruptcy is only part of failure. So really it is what percentages of Fortune 500 (or other measure of successful companies) have "their company went bankrupt, they were forced out of the CEO chair, a major merger backfired, and so on" Startups are in fact the wrong measure. A startup has nothing to lose, and often is trying something that will not succeed. Large companies often fund startups to limit their potential loss if it fails (if things seem not be going well they pull out without having to lay off all the people working for the startup which looks bad), and they buy the company if things go well.