Decision IntelAmazon
Amazon AWS: The Accidental $80B Business
Estimated impact: $80B+ annual revenue; created $1.5T cloud computing market
Amazon Web Services began as an internal infrastructure project in 2003, launched publicly in 2006, and grew into a $80+ billion annual revenue business by 2022. Jeff Bezos bet that Amazon's internal computing infrastructure could be productized for external developers. Wall Street analysts dismissed the project as a distraction from e-commerce. Andy Jassy led the effort with a small team, operating with minimal resources for years before reaching profitability.
Decision context
Whether to invest in building and selling cloud infrastructure as a service — a market that did not yet exist — while the core e-commerce business was still proving its own profitability to skeptical investors.
What the company’s own filing already carried
Not our reading of a memo. Amazon's own 10-K405, filed 2002-01-24 (12 months before the 2003 decision), was read by the audit engine's deterministic structural layer: the same detectors that run on every audit, with no language model involved. Everything below is drawn from the company's own disclosure, before any knowledge of what followed, and never from the decision itself.
Survivable as structured: if the flagged risk fires, the deal has the exits, cushions, or staging to absorb the hit and continue.
The structural layer flagged no unabsorbed condition in this filing.
- This reads the company's own 10-K405 (filed 12 months before the decision), not the decision memo itself, so it can only understate what a full audit of the memo would surface.
- It reads the structure of the situation, not the trigger: whether a shock to the flagged conditions would be absorbed, never that a shock arrives.
- A reading of a dated filing, not a prediction: every condition can be checked against the company's own 10-K405.
Decision anatomy
Red = risk factor present · Green = protective factor present
Biases present in the decision
★ Primary driver · Severity estimated from bias type and decision outcome
The curated library this case sits in
Across all 155 curated case studies in our library, by documented outcome:
The library is curated toward documented failures, so this split describes the library, not the base rate of deals. The case above is one row in it.
Lessons learned
- Overconfidence bias was present (Bezos believed Amazon could serve external developers better than anyone) but was mitigated by iterative development and customer feedback loops.
- Anchoring bias was managed by NOT anchoring to competitor pricing or existing market definitions — AWS created a new category.
- Survivorship risk is medium: the bet could have easily failed if the internal infrastructure wasn't genuinely world-class or if enterprise adoption had been slower.
Source: Brad Stone, "The Everything Store" (2013); Amazon 10-K filings (2006-2015); AWS re:Invent keynotes (2012-2015) (Annual Report)
These patterns were flaggable in Amazon's own record — before the outcome.
See the full reasoning audit we ran — no login, no card. Then run the same audit on a deal you have already closed.
Or leave your email, we'll run the audit on a deal of your choosing and send the readout within a business day.
Workflows that fire on decisions like Amazon’s
The same Recognition-Rigor Framework that documents this case audits memos in the same shape — before the outcome forces the lesson.