Cloud backlog (remaining performance obligations)
1What this measures
Remaining performance obligations reported by Oracle, Microsoft and CoreWeave at each quarter end, from SEC filings. Contracted, not yet recognised revenue; the headline series is Oracle's.
Why it matters. RPO is the forward view of whether compute demand is contracted ahead of revenue. Its growth is the clearest filed evidence of capex-ahead-of-revenue, and of how much of it is one customer.
- Proxy types
- capital_flow, market_structure
- Unit
- USD
- Cadence
- quarterly
2How we track this
- series
sec.orcl.rpo.q - series
sec.msft.rpo.q - series
sec.crwv.rpo.q
Direction over 4 periods, dead-band $10.0B; higher = concentrating.
Four quarters; moves under $10B are noise at this scale. Growing backlog concentrates future cloud revenue in a few contracts.
Applied to sec.orcl.rpo.q.
3Tracker interpretation
Backlog is growing far faster than recognised revenue; a large share is a handful of AI-lab contracts, which is the circular-financing question in filed form.
4Evidence
Direction over the last 4 periods; the band is ±$10.0B around the window's first point. Higher reads concentrating.
73 observations. Hollow points are disputed (see counterevidence). Every point links to its observation.
5Status and reasoning
Oracle remaining performance obligations at quarter end (10-Q/10-K XBRL): May 2025 $138B, Aug 2025 $455B, Nov 2025 $523B, Feb 2026 $553B, May 2026 $638B. Rising every quarter, far above the $10B dead band; Microsoft ($684B at Jun 2026) and CoreWeave ($104B) move the same way. Initial seed.
6Timeline notes
7Counterevidence
What cuts against this reading
RPO includes non-cancellable commitments that may never convert; Oracle's ~$300B OpenAI contract is reported, not itemised in the filing. Definitions differ across filers.
8Update history
- 2026-09-10unmeasured to concentratingconf — → 80 · claude-initial-seed
Oracle remaining performance obligations at quarter end (10-Q/10-K XBRL): May 2025 $138B, Aug 2025 $455B, Nov 2025 $523B, Feb 2026 $553B, May 2026 $638B. Rising every quarter, far above the $10B dead band; Microsoft ($684B at Jun 2026) and CoreWeave ($104B) move the same way. Initial seed.
9Confidence
80 / 95 — good evidence, some ambiguity
Confidence is independent of status: 90–95 multiple strong independent sources; 70–89 good evidence, some ambiguity; 50–69 mixed or hard to operationalise; below 50 limited or vague.
10Related
- Rent concentration in semiconductors concentrating
- Bottlenecks #61 (Narayanan & Kapoor's list)