Even 20% growth gets AWS to $1T revenue in 10 years
Amazon Web Services (AWS) reports a $184B annual run rate (Q3 revenue × 4).
I started from a more conservative base: $161.4B, actual trailing revenue in the 12 months to September.
Time to $1T under three growth scenarios:
40% a year continues: ~5.4 years
40% for two years, then moderates to 20%: ~8.3 years
20% a year: ~10 years
Andy Jassy's outlook on the Thursday call:
"we believe that AWS can grow north of 40% in each of 2027 and in 2028, and be a trillion dollar revenue business in the next 5 to 10 years."
Growth today is 39.3%. Halve it, and compounding on this base still gets AWS to $1T within 10 years.
So how realistic is this?
$578B of demand is already under contract
AWS customer backlog hit $578B, up $82B in one quarter.
The next day, AWS CEO Matt Garman said in a Bloomberg interview:
"We have more customers that have committed to billion dollar spends with us this year than we ever have before."
Those commitments give AWS years of visibility.
And customers' AI journey is still early: many already get value from AI transforming "one, two, three, four, 5%" of their business, and he expects "3% is gonna grow to 10% is gonna grow to 50%."
Even model spend is spread across customers
Garman said AWS, unlike some rivals, isn't concentrated in a couple of AI labs.
I wrote that Anthropic's S-1 shows 48% of its revenue now comes through cloud marketplaces: roughly $7.8B in H1 2026 alone (my math).
Garman's framing: "a lot of that spend is in Bedrock. And so it's actually our customers that are spending on Anthropic in Bedrock." That's "tens to hundreds of thousands of customers" running inference.
With OpenAI and other models now on Bedrock, AWS is both a partner to the labs and the platform where enterprises buy their models.
$1T AWS revenue = multi-trillion-dollar partner opportunity
AWS-commissioned research by Omdia puts partner revenue at up to $7.13 for every $1 customers spend on AWS. Even a fraction of that puts the ecosystem around a $1T AWS revenue in the trillions.
At today's ratio, a $1T AWS will mean $3T+ in customer commits.
Current $578B backlog is 3.14× of AWS's annualized revenue. If that ratio holds at 1T, it implies 3T+ in cloud commits.
If 25% can go toward eligible Marketplace purchases, that's ~$800B of potential spend across multi-year contracts.
What this means for alliance leaders
AWS has a realistic path to a trillion-dollar platform. If you partner with AWS or sell through AWS Marketplace, your opportunity compounds with it.
Is your 2027 AWS plan built for the next quarter, or the next five years?