4 ms·

You customize your workforce to meet the current market demands. When the price of oil is high, you hire a lot of people to work the wells. When the price of o

You customize your workforce to meet the current market demands. When the price of oil is high, you hire a lot of people to work the wells. When the price of oil tanks, but the price of coal rises, you downsize on oil production and ramp up coal mining.

ethbro 11y ago
> you hire a lot of people to work the wells I think it's a valid question whether that's still a valid approach in a modern, high-value knowledge economy (and yes, I'm including oil well operators in there). Honestly interested from companies that have been through grow-shrink-grow. Is it as clear as that, re: being able to find new people when the time is righr? Or is there enough of a ramp-up time for new employees that the equation is more complex?
- Eridrus 11y ago
  The usual estimate of ramp-up time for employees at a startup is considered to be a month or two, but the business cycle is measured in years, so if you expect a secular downturn, firing people makes sense. I think oil is actually complicated and rare enough that it might actually be an example where you don't want to lay them off, since finding replacements may be quite hard if you lay off those positions for a few years, so you may want to try and weather the storm while the Saudis flood the market with oil. Having said that, companies make bad decisions all the time since making good decisions is a lot harder than everyone seems to think.
  - collyw 11y ago
    Sure you can be productive after a month or two, but having the same in depth knowledge of a product will take a lot longer. My last place claimed they couldn't afford to give me more money, yet they are now employing two people to do my job (and I doubt they will be as effective as me).
    - Inthenameofmine 11y ago
      This is very specific for people who work on systems, such as programmers. I have a devshop and have sent his over and over again. The actual cost per employee isn't really paying them salaries to do stuff, but rather the salaries you pay them to learn stuff and get acquainted with the system. We call it ROL (Return on Learning). We've learned that on average it takes 3-8 months for an intelligent person to become fully productive within any system. Since we started hiring for the capacity to learn and culture fit, rather than existing technical skill, it's changed our business for the better. That said, turns out even if you're in the Bay area with all that great talent pool, chances to find local people who just happen to want to work with you are practically nill. Once you truly internalize this concept of ROL, there is no other way but to work globally with remote people and teams. Without that, you can't access a sufficiently large talent pool. I think the ROL is actually the leverage factor people have in mind when they say "10x" engineer.
      - bcassedy 11y ago
        How do you select for these people with high capacity to learn?
      - abalashov 11y ago
        In my experience, the more specialised the work, the more one has to select for a balance of capacity to learn with a certain amount of objective skills and experience present today. I tried the strictly-capacity route in the past (it was all I could afford) and, while these people had all the potential in the world, the ramp-up time was simply more than the business could sustain when customers expected high-end knowledge and expertise now. It was aggravated by a lack of time and resources typically found in a woefully undercapitalised startup, of which a consequence was inability to properly structure and manage the learning process.
      - ethbro 11y ago
        I also think "having someone already in the company with X knowledge, who will be working closely enough with new hire to share it" is a huge multiplier in ramp-up rate. The amount of un- or poorly-documented things in decade+ legacy systems can be a minefield, but so easily solves with a quick "Oh, X, we always do Y because Z."

ethbro 11y ago

> you hire a lot of people to work the wells I think it's a valid question whether that's still a valid approach in a modern, high-value knowledge economy (and yes, I'm including oil well operators in there). Honestly interested from companies that have been through grow-shrink-grow. Is it as clear as that, re: being able to find new people when the time is righr? Or is there enough of a ramp-up time for new employees that the equation is more complex?

Eridrus 11y ago
The usual estimate of ramp-up time for employees at a startup is considered to be a month or two, but the business cycle is measured in years, so if you expect a secular downturn, firing people makes sense. I think oil is actually complicated and rare enough that it might actually be an example where you don't want to lay them off, since finding replacements may be quite hard if you lay off those positions for a few years, so you may want to try and weather the storm while the Saudis flood the market with oil. Having said that, companies make bad decisions all the time since making good decisions is a lot harder than everyone seems to think.
- collyw 11y ago
  Sure you can be productive after a month or two, but having the same in depth knowledge of a product will take a lot longer. My last place claimed they couldn't afford to give me more money, yet they are now employing two people to do my job (and I doubt they will be as effective as me).
  - Inthenameofmine 11y ago
    This is very specific for people who work on systems, such as programmers. I have a devshop and have sent his over and over again. The actual cost per employee isn't really paying them salaries to do stuff, but rather the salaries you pay them to learn stuff and get acquainted with the system. We call it ROL (Return on Learning). We've learned that on average it takes 3-8 months for an intelligent person to become fully productive within any system. Since we started hiring for the capacity to learn and culture fit, rather than existing technical skill, it's changed our business for the better. That said, turns out even if you're in the Bay area with all that great talent pool, chances to find local people who just happen to want to work with you are practically nill. Once you truly internalize this concept of ROL, there is no other way but to work globally with remote people and teams. Without that, you can't access a sufficiently large talent pool. I think the ROL is actually the leverage factor people have in mind when they say "10x" engineer.
    - bcassedy 11y ago
      How do you select for these people with high capacity to learn?
    - abalashov 11y ago
      In my experience, the more specialised the work, the more one has to select for a balance of capacity to learn with a certain amount of objective skills and experience present today. I tried the strictly-capacity route in the past (it was all I could afford) and, while these people had all the potential in the world, the ramp-up time was simply more than the business could sustain when customers expected high-end knowledge and expertise now. It was aggravated by a lack of time and resources typically found in a woefully undercapitalised startup, of which a consequence was inability to properly structure and manage the learning process.
    - ethbro 11y ago
      I also think "having someone already in the company with X knowledge, who will be working closely enough with new hire to share it" is a huge multiplier in ramp-up rate. The amount of un- or poorly-documented things in decade+ legacy systems can be a minefield, but so easily solves with a quick "Oh, X, we always do Y because Z."