3 ms·

I think that's a sore underestimation of the support costs Google would have to cover. It looks like support costs average around $4/call. (https://www.google.c

I think that's a sore underestimation of the support costs Google would have to cover. It looks like support costs average around $4/call. (https://www.google.com/search?q=how+much+does+the+average+support+call+cost https://www.google.com/search?q=how+much+does+the+average+su...). Google has approximately 4 billion users (https://www.google.com/search?q=how+many+users+does+Google+have https://www.google.com/search?q=how+many+users+does+Google+h...). So, if you allowed for one single support call per user per year, Google would be footing $16B annually. In general, I don't think you can expect free support for a free product. I do think businesses should let you buy support for any tier of product, both as part of a subscription and on a one-time basis.

Retric 4y ago
That’s at 98 cents per minute, I suspect at google’s scale they would have massive cost savings.
- foobazgt 4y ago
  Savings from scale usually comes from automation, and people are generally looking to talk to humans, not bots. This is the exact reason Google avoids human support in the first place - there are no massive cost savings, because it does not scale well.
  - Retric 4y ago
    There’s a lot to automate when doing phone support. The vast difference between the sub 20 cents per minute you need to pay people to answer phones and the 98 cents per minute each call costs are largely things that could be automated. And that’s ignoring trying to minimize call length or even the need for any calls in the first place.
    - foobazgt 4y ago
      There are definite cost reductions that can be made, but they pale in comparison to keeping humans out of the loop. For sake of argument, let's pretend you can shave half off the costs. So, now we're talking $8B annually for one support call / user / year. Does this undermine my argument about under-estimation of $1B costs?
      - Retric 4y ago
        That’s just the most obvious approach, adjusting the product and internal processes to avoid having people call is even more effective. Upselling people is another way to drop net costs even further, etc. I suspect they could get net costs below 800m/year.
        foobazgt 4y ago
        Ok, I agree to disagree then! :)

Retric 4y ago

That’s at 98 cents per minute, I suspect at google’s scale they would have massive cost savings.

foobazgt 4y ago
Savings from scale usually comes from automation, and people are generally looking to talk to humans, not bots. This is the exact reason Google avoids human support in the first place - there are no massive cost savings, because it does not scale well.
- Retric 4y ago
  There’s a lot to automate when doing phone support. The vast difference between the sub 20 cents per minute you need to pay people to answer phones and the 98 cents per minute each call costs are largely things that could be automated. And that’s ignoring trying to minimize call length or even the need for any calls in the first place.
  - foobazgt 4y ago
    There are definite cost reductions that can be made, but they pale in comparison to keeping humans out of the loop. For sake of argument, let's pretend you can shave half off the costs. So, now we're talking $8B annually for one support call / user / year. Does this undermine my argument about under-estimation of $1B costs?
    - Retric 4y ago
      That’s just the most obvious approach, adjusting the product and internal processes to avoid having people call is even more effective. Upselling people is another way to drop net costs even further, etc. I suspect they could get net costs below 800m/year.
      - foobazgt 4y ago
        Ok, I agree to disagree then! :)