3 ms·

Thanks for the great rundown of some factors! I would like to take issue with just one part: > Very few people at that level would want to relocate for any am

Thanks for the great rundown of some factors! I would like to take issue with just one part: > Very few people at that level would want to relocate for any amount of pay, hence a shortage. I'm not sure if you're using hyperbole or not. But I'm guessing that if you paid $1 million USD / year, there would be no practical shortage of data scientists willing to live in Podunk, Iowa.

wenc 8y ago
There is a cost-benefit proposition to every hire. Unfortunately, the issue there is that the majority of data scientists don't generate $1 mil of value (exceptions exist of course), and so it's difficult to justify that level of compensation to management, which means the position may never get created in the first place. Absent an existential threat to their bottom lines, companies will just muddle on without hiring. This may be different in high growth companies, but in most traditional companies that aren't sloshing around in VC cash, to increase headcount you need to provide a value justification vis-a-vis salary (unless it's for cost center positions, but even then..) There's also an underlying assumption that talent flocks to the highest bidder.... but most HR folks will tell you that it's more complex than that. There are many quality of life issues that come into play, like weather, peer-group, spousal happiness, etc. Money doesn't buy everything, and humans aren't optimizers but satificers. I'm a sample of one, but for me, I would absolutely not move to Podunk, IA for a $1mil salary. I'm happy taking a lower salary living in a city and intellectual milieu that feeds me. The kind of person who would take the $1mil may likely not be the kind of talent you want.
rabidrat 8y ago
It does not matter how much you pay, there are only so many 'qualified' data scientists to go around. If the wages are so ridiculously high, then there will be all manner of hucksters pitching themselves as data scientists with minimal training (but good enough to pass an interview), and they will give data scientists a bad name. This will make paying high-wages a super risky move, which almost never pays off, putting downward pressure on demand and wages. This is already happening.

wenc 8y ago

There is a cost-benefit proposition to every hire. Unfortunately, the issue there is that the majority of data scientists don't generate $1 mil of value (exceptions exist of course), and so it's difficult to justify that level of compensation to management, which means the position may never get created in the first place. Absent an existential threat to their bottom lines, companies will just muddle on without hiring. This may be different in high growth companies, but in most traditional companies that aren't sloshing around in VC cash, to increase headcount you need to provide a value justification vis-a-vis salary (unless it's for cost center positions, but even then..) There's also an underlying assumption that talent flocks to the highest bidder.... but most HR folks will tell you that it's more complex than that. There are many quality of life issues that come into play, like weather, peer-group, spousal happiness, etc. Money doesn't buy everything, and humans aren't optimizers but satificers. I'm a sample of one, but for me, I would absolutely not move to Podunk, IA for a $1mil salary. I'm happy taking a lower salary living in a city and intellectual milieu that feeds me. The kind of person who would take the $1mil may likely not be the kind of talent you want.

rabidrat 8y ago

It does not matter how much you pay, there are only so many 'qualified' data scientists to go around. If the wages are so ridiculously high, then there will be all manner of hucksters pitching themselves as data scientists with minimal training (but good enough to pass an interview), and they will give data scientists a bad name. This will make paying high-wages a super risky move, which almost never pays off, putting downward pressure on demand and wages. This is already happening.