4 ms·

Well if you were to diagnose most companies there are probably 30-40 engineers that make the most value, like billions of dollars each. From there it starts dro

Well if you were to diagnose most companies there are probably 30-40 engineers that make the most value, like billions of dollars each. From there it starts dropping off like a long tail in search results. There's value, but a lot less than others. Cutting the long tail may still break huge areas of functionality - but the rest will patch and fix it over time. Heck there might be a team of 50 that has the sole responsibility to figure out the code/algorithms that puts something on the main page automatically. Instead that can temporarily be a small team of 3 people that rotate news articles or tweets. We tend to think "each employee generates $44B / 14000" but that's far from the truth.

mustacheemperor 4y ago
Following along your theory, do you expect that Twitter will achieve meteoric profit successes once this damage is patched, and that there will be an epic staffing contraction across tech afterwards? If you are correct then Twitter is about to demonstrate to the entire tech market that their staff headcounts are factors of ten beyond what is necessary to achieve equal success, and we should expect a massive course correction in the industry as a result. >there are probably 30-40 engineers that make the most value, like billions of dollars each. From there it starts dropping off like a long tail in search results. There's value, but a lot less than others. Cutting the long tail may still break huge areas of functionality - but the rest will patch and fix it over time. If this is true, then have companies like Microsoft, Apple, all just missed the big insight Elon has caught onto first? If the best engineers are generating a few billion in value each and can completely patch up the long tail, then Microsoft ought to be able to cut its staff down to about 200 such engineers, 300 if they want a lot of extra margin. 1.8T / 3bn per high performer. There's over 100k SWEs currently employed there, needless to say that's a big reduction in salary overhead.
- rhacker 4y ago
  I do believe the theory. Most people working are head nodders and fix some widget. It only takes them 10 minutes a week to fix or tweak. The only reason the widget exists is so that the company can claim X feature exists in their portfolio. The fear of firing them is that the feature will eventually break and that person is no longer there. But at some point they'll realize it's super simple and can fix it. - But the fear of cutting endless departments like that is so powerful. Any yes, the top performers don't want to work on widget X even if it's 10 minutes of their day. Here's the analogy - lots of startups with 50 employees have sold for billions right. Once the company sells the payout potential is OVER which at that point those 30 of those 50 leave (within a year) and are replaced with 1000 people to take over and slowly add new features. Now remember the codebase still exists that has revenue potential that can feed 1000 or more people, but they did it with 50. So the code is paying for the product, it's just stable and maybe making a little profit per year. That's enough to justify it. But startups prove all the value is in those 50 people. The other 1000? they're all replaceable but in a union. The union is that idea that ANY headcount reduction is bad press. Why'd you fire Zak, I'm leaving too! rant over

mustacheemperor 4y ago

Following along your theory, do you expect that Twitter will achieve meteoric profit successes once this damage is patched, and that there will be an epic staffing contraction across tech afterwards? If you are correct then Twitter is about to demonstrate to the entire tech market that their staff headcounts are factors of ten beyond what is necessary to achieve equal success, and we should expect a massive course correction in the industry as a result. >there are probably 30-40 engineers that make the most value, like billions of dollars each. From there it starts dropping off like a long tail in search results. There's value, but a lot less than others. Cutting the long tail may still break huge areas of functionality - but the rest will patch and fix it over time. If this is true, then have companies like Microsoft, Apple, all just missed the big insight Elon has caught onto first? If the best engineers are generating a few billion in value each and can completely patch up the long tail, then Microsoft ought to be able to cut its staff down to about 200 such engineers, 300 if they want a lot of extra margin. 1.8T / 3bn per high performer. There's over 100k SWEs currently employed there, needless to say that's a big reduction in salary overhead.

rhacker 4y ago
I do believe the theory. Most people working are head nodders and fix some widget. It only takes them 10 minutes a week to fix or tweak. The only reason the widget exists is so that the company can claim X feature exists in their portfolio. The fear of firing them is that the feature will eventually break and that person is no longer there. But at some point they'll realize it's super simple and can fix it. - But the fear of cutting endless departments like that is so powerful. Any yes, the top performers don't want to work on widget X even if it's 10 minutes of their day. Here's the analogy - lots of startups with 50 employees have sold for billions right. Once the company sells the payout potential is OVER which at that point those 30 of those 50 leave (within a year) and are replaced with 1000 people to take over and slowly add new features. Now remember the codebase still exists that has revenue potential that can feed 1000 or more people, but they did it with 50. So the code is paying for the product, it's just stable and maybe making a little profit per year. That's enough to justify it. But startups prove all the value is in those 50 people. The other 1000? they're all replaceable but in a union. The union is that idea that ANY headcount reduction is bad press. Why'd you fire Zak, I'm leaving too! rant over