5 ms·

When your market and customer base is _that_ big, maybe you have scaled past what you are reasonably capable of supporting with actual humans.

GoOnThenDoTell 6y ago
that sounds like ‘scaled too far’ then
- ben_w 6y ago
  Or “not actually scaled”
jakub_g 6y ago
Alphabet had $34B net income in 2019, FB $18B. They could hire a legion of developers in any place which is not California/Switzerland, whose only responsibility would be bugfixing, and it would barely make a dent in their numbers.
- georgeecollins 6y ago
  There are those "mythical man month" problems along the lines of you can't hire legions of programmers and know that you are actually reducing bugs more than you are creating them. I am sure Facebook and Google could do a better job of support if it was there top priority, but it probably isn't just an issue of cost. It also requires attention, prioritization, engineering-- which each company probably has a finite amount of.
  - tpxl 6y ago
    You can instead hire a million support personnel for that kind of money.
    - creddit 6y ago
      That only cost you $18K/yr??? Not so sure.
      - tpxl 6y ago
        Average wage in most of the developed world is < 18K/year.
        georgeecollins 6y ago
        Again, my point is it is not just the number of people that make software work. It takes vision, the focus of the organization, prioritization.
        creddit 6y ago
        Wage != cost. It is just one component. Not sure of your definition of “developed world” but the median wage (less than average) of the US and other OECD countries is typically much closer to $30K.
  - jakub_g 6y ago
    For sure, when creating core technology it's better to have 2 people do it than to have 20 people. But when you have stable core and need to polish the edges, additional headcount with proper incentives should be able to make a difference. Part of the problem is (possibly biggest) lack of incentives all over the chain and focusing too much (solely) on a small set of business metrics.

GoOnThenDoTell 6y ago

that sounds like ‘scaled too far’ then

ben_w 6y ago
Or “not actually scaled”

jakub_g 6y ago

Alphabet had $34B net income in 2019, FB $18B. They could hire a legion of developers in any place which is not California/Switzerland, whose only responsibility would be bugfixing, and it would barely make a dent in their numbers.

georgeecollins 6y ago
There are those "mythical man month" problems along the lines of you can't hire legions of programmers and know that you are actually reducing bugs more than you are creating them. I am sure Facebook and Google could do a better job of support if it was there top priority, but it probably isn't just an issue of cost. It also requires attention, prioritization, engineering-- which each company probably has a finite amount of.
- tpxl 6y ago
  You can instead hire a million support personnel for that kind of money.
  - creddit 6y ago
    That only cost you $18K/yr??? Not so sure.
    - tpxl 6y ago
      Average wage in most of the developed world is < 18K/year.
      - georgeecollins 6y ago
        Again, my point is it is not just the number of people that make software work. It takes vision, the focus of the organization, prioritization.
      - creddit 6y ago
        Wage != cost. It is just one component. Not sure of your definition of “developed world” but the median wage (less than average) of the US and other OECD countries is typically much closer to $30K.
- jakub_g 6y ago
  For sure, when creating core technology it's better to have 2 people do it than to have 20 people. But when you have stable core and need to polish the edges, additional headcount with proper incentives should be able to make a difference. Part of the problem is (possibly biggest) lack of incentives all over the chain and focusing too much (solely) on a small set of business metrics.