We don't know if they're charging less than it costs to run. They probably simply have to make investments this year in order to acquire revenue next year. Or a

We don't know if they're charging less than it costs to run. They probably simply have to make investments this year in order to acquire revenue next year. Or at the very least it's a large factor. They could choose to not grow, and instead realize it as profit, but it's not a long term good move. It's probably a good idea to spend $4B (made up number) this year in order to be able to grow revenues next year by $11B. (some simplified arithmetic based on Q3's $6.868B revenue times 4 for the year) And then next year $5.6B to acquire $15B the year after. The nature of exponential growth (anyone who's started a growing company would know this) is that the investment that was expensive this year could easily be paid by the same company next year. Assuming Google can avoid investing the $4B with current customer base, that would take them over a billion into profitable territory. Again, have a look at how Amazon did this the last ~20 years. You can't spin up a DC the same month you find out you need the capacity. It just cannot be done. You need to take into account the time axis. I'm looking forward to the Q4 earnings report. We live in interesting times, though. The USD exchange rate due to the wider economy has significantly affected the earnings of multinational conglomorates as expressed in USD, lately.