sources
aysra is built on public filings submitted to the Internal Revenue Service.
what we ingest
aysra ingests the IRS Exempt Organizations Business Master File and bulk XML releases. these include:
- Form 990
- Form 990-PF
- Form 990-EZ
Form 990-N (the e-postcard, used by organizations with less than $50,000 in annual gross receipts) is excluded. the IRS does not release structured XML for those filings.
each filing is parsed as filed. values are not modified or estimated at the source layer.
the same files are publicly available through the IRS Form 990 Series Downloads page.
release cadence
IRS Form 990 filings are due annually — typically May 15 for calendar-year organizations. the IRS then processes and publishes the data in bulk periodically. public availability typically lags 2–6 months after filing, or 12–18 months after a nonprofit's fiscal year end.
the IRS publishes through the Tax Exempt Organization Search (TEOS) bulk data downloads tool, which is updated frequently. aysra ingests new bulk releases as they become available.
coverage is uneven in the most recent tax year and becomes more complete over time.
what this means
- values reflect what was filed
- missing filings are not inferred
- blank fields remain blank
aysra does not supplement IRS data with scraped or third party datasets at the source layer.
scope of sources
IRS filings are the system of record.
- data reflects what organizations report to the IRS
- disclosure varies by form and filing requirements
- differences across forms are preserved
aysra does not attempt to harmonize these differences at the source layer.
for how this data is processed once ingested, see methods.
methods →