$AMZN · Schwab Network · 2026-10-30

We saw Amazon shares fall more than 6 percent. We have restored some of it. We are now down less than 2 percent. Even though AWS has grown by almost 40 percent, we are still under pressure.

You say that Wall Street is overreacting to this spending. And why should they look beyond their $42.4 billion negative free cash flow? I believe this share selling spree is mainly a reaction to capital expenditure or CAPEX.

But the main point is, according to your points, AWS had their best quarter in the last 5 years. Cloud revenue grew more than 39 percent, enterprise backlog reached $578 billion, and the company's operating margin reached a record high.

So, I think if you're a short-term trader, it's natural to be worried about infrastructure costs, but Amazon isn't spending this money for short-term investors. They are doing this for the long term, and in my opinion their current range of services with AI will bring the enterprise AI sector under their control for the next decade.

So, if we think long term, management is saying that they expect significant growth in spending in 2027, which is higher than what it is now. And we were certainly well above expectations in terms of capex.

That is, our costs have exceeded $77 billion, while expectations were below $60 billion.

Yes, I think it's important for investors to keep an eye on capex spending and AWS' growth rate. If you see capex increasing and AWS growth slowing, then as an investor you should be worried.

I think we're in the very early stages of AI here. We talked a lot about physical AI, edge inferencing, and things like that. And Amazon is in a pretty good position compared to its other hyperscaler competitors.

So from an investor's perspective, I think you have to give it a few more quarters, but the product portfolio that Amazon has now for enterprise AI is really unique. And I think they are in a very good position.

In fact, you could say that in the next wave of AI, which will be largely CPU-driven, their decision to sell their own CPUs directly, even to competitors, will put them in a more advantageous position to compete with Nvidia or perhaps even this new wave of AI.

So let's talk about AWS and specifically the backlog. That means it has come to $578 billion. In terms of visibility, how important do you think it is? How much lead time or advance time can we see through this number?

And how quickly do you think Amazon will be able to convert these promises into actual sales and profits?

Yes, that's a good question. I think they can convert these pretty quickly. And that means, within a few quarters. We are at the end of 2027... sorry, we are at the end of 2026.

If we don't see this transition by mid-2027, investors should be concerned.

I think when it comes to AI, there's a bit of a time lag between what the technology is capable of doing and what organizations are actually doing. I've talked to a lot of CIOs and they all have AI on their roadmaps, and they're wondering what to do with it.

I think we'll start seeing this from the middle of next year.

And in terms of technology, we think of taking the typical 'hockey stick', in terms of AI, the blade will be longer and then there will be a much steeper curve towards the top.

And that's when we'll see that Amazon backlog start to decrease.