Macro Releases & Vintages
What each economic release said at the time, and how it was revised.
Macro series straight from the publishing agencies (BLS, BEA, Census, Treasury, CFTC, EIA), captured as released, with first-print and revised vintages. Each release is a new version of the same fact, so both 'what did the release say then' and 'how was it revised' are simple as-of queries. Also the CFTC Commitments of Traders and the ECB's euro reference rates for 13 currencies since 1999.
Questions it answers
- What was the first print of payrolls or CPI, and how was it revised?
- How are speculators positioned in futures this week?
Use cases
- Macro backtests on first prints
- Nowcasting with release-time data
- Futures positioning (COT)
- Coverage
- 266,533 values; 8 publishing agencies
- History
- Back to January 1999
- Update frequency
- Hourly to weekly, by agency
- Key fields
- Series, release date, vintage, value, revision; COT positions; FX reference rates
- Point in time
- Vintages stored; release days stamped no earlier than the real release
- Sources
- BLS, BEA, Census and Treasury releases · CFTC Commitments of Traders · ALFRED vintages (St. Louis Fed) · EIA · ECB euro reference rates
- Licence
- US agencies: public domain. ECB, EIA and ALFRED: delivered with the source credited
- Delivery
- Data API · MCP · Python SDK · Excel · S3/Snowflake files · App
Data as of 2 November 2026
Each value carries the time it became public and the time Factget first saw it. Corrections and restatements are new versions; earlier versions stay queryable, so a backtest only sees what was known on each date. The source file behind every value is stored unaltered.
Build a datasetfrom your sources.
Tell us the sources and the fields you need. We’ll show you what factget can build, and keep current.