Mobile Proxy Nonprofit Data
The charitable sector is unusually well documented, because regulators require it. Annual returns, accounts and trustee records are public by obligation in most jurisdictions — the work is that every jurisdiction does it differently and all of it lags.
- Regulator filings are the backbone — annual returns and accounts are public by statutory obligation.
- Each jurisdiction has its own regulator — and its own forms, thresholds and definitions of a charity.
- Everything lags by a year or more — design the analysis around it rather than fighting it.
- Small charities report far less — reporting thresholds create a systematic gap in every dataset.
Reach each jurisdiction’s charity register directly.
Track returns and accounts as a multi-year series.
Nonprofit Data and Where It Lives
The charitable sector is unusually well documented, and for a straightforward reason: regulators require it. Organisations above a size threshold must file annual returns and accounts, and those filings are public by obligation rather than by choice. Income, expenditure, trustee names, stated activities and filing history are all there.
What makes it a collection problem is that every jurisdiction does it differently. The United States publishes annual information returns for tax-exempt organisations, with several services republishing the whole corpus in structured form. England and Wales have the Charity Commission register, Scotland has OSCR, Australia has the ACNC, and Canada publishes charity returns through its revenue agency. Different forms, different thresholds, different definitions of what a charity even is.
| Jurisdiction | Regulator or source | Character |
|---|---|---|
| United States | Federal annual information returns, plus structured republishers | Very complete, heavily lagged, bulk available |
| England and Wales | Charity Commission register | Open, searchable, with accounts attached |
| Scotland | OSCR | Separate regulator with its own register and rules |
| Australia | ACNC | Open register with an annual information statement |
US Filing Thresholds: 990-N, 990-EZ, 990 and 990-PF
The United States return itself splits by size. Organisations under $50,000 in annual gross receipts file Form 990-N, an electronic postcard that discloses little beyond the organisation's existence and address. Mid-sized organisations file the short-form 990-EZ or the full Form 990, both of which include a statement of revenue, functional expenses and a list of officers. Private foundations file Form 990-PF regardless of size, which additionally discloses grants paid, investment income and compliance with self-dealing rules — a foundation writing one grant a year still files the long form.
Most e-filed returns since 2012 are published as machine-readable XML through a public data release, and structured search tools rebuild the underlying Business Master File and filing extracts into a searchable database rather than requiring anyone to parse XML directly. Coverage is not universal: only around 60 to 65 per cent of 990 and 990-EZ filers submit electronically rather than on paper, so the machine-readable corpus understates the smaller end of the sector even before the threshold gap described below is added on top.
Regulator Filings by Jurisdiction
Two properties of this data shape every analysis built on it, and both need stating up front rather than discovering later.
The first is lag. Annual returns are filed after a financial year ends, usually near the deadline, and the regulator then takes time to publish. Twelve to eighteen months between activity and its appearance in the register is entirely normal. Nonprofit data answers questions about trajectory and structure; it cannot answer what is happening this quarter, and treating it as though it can produces confident nonsense.
The second is the threshold gap. Every regulator sets a size below which an organisation files a much shorter return or none at all, so the smallest charities are systematically underdescribed. Any sector analysis built from filings therefore over-represents large organisations. That is not a collection failure and no amount of extra crawling fixes it — but it has to be stated, because otherwise the dataset quietly misdescribes the sector it claims to measure.
Sector Classification and NTEE Codes
Filings are usually tagged with a National Taxonomy of Exempt Entities code, a three-character classification that sorts organisations into major groups such as arts, education, health, human services and religion, then into finer subcategories within each. The taxonomy is self-reported at registration and rarely revisited, so an organisation whose activities have shifted over a decade often still carries its original code. Any sector breakdown built from NTEE codes is a breakdown of stated purpose at founding, not of current activity, and that distinction matters most for older, larger organisations that have had the most time to drift.
Grants, Funding and Programme Data
Beyond the statutory filings, the richer material is in what organisations publish themselves: grant awards, programme reports, impact publications and funder listings. Grant-making foundations in particular frequently publish award lists that name recipient, amount, purpose and period — which, aggregated, describes funding flows through a sector far better than any individual return does.
This material is also where a local exit starts to matter, because national regulators and funders build for domestic users. Several degrade or restrict access from foreign address ranges, apply harder bot mitigation to them, or render search behaviour differently. Reaching each register from inside its own jurisdiction is the most reliable way to see what a domestic researcher sees.
Ratings Services and Third-Party Aggregators
A layer of services sits on top of the primary filings and adds its own value: rating agencies score organisations on financial efficiency and transparency, and republishing platforms combine years of historical returns into a single organisation profile with trend lines the raw filings do not show on their own. These aggregators are useful starting points for due diligence and donor research, but they inherit every property of the underlying filings — the same lag, the same threshold gap and the same self-reported classification — so treat a rating as a summary of old filings, not as a live signal.
Collecting Nonprofit Data Responsibly
These are small organisations and public regulators, frequently running on modest infrastructure, and the data changes annually. Collecting it aggressively is both pointless and rude — an annual dataset does not reward a daily crawl, and a monthly sweep catches everything that moves.
- Take the bulk file where one exists — Several regulators publish the whole register for download. Crawling instead is slower, less complete and worse for both parties.
- Key on the registration number — Charities rename and merge; the regulator’s identifier is the only stable key.
- Record the financial year, not the filing date — Filing dates cluster around deadlines and say nothing about the period described.
- State the threshold in your analysis — Every conclusion about the sector is conditional on which organisations had to report.
- Treat trustees as personal data — Names are published for regulatory transparency, which is not the same as unrestricted reuse. Take your own legal advice before aggregating individuals across registers.
Matching Organisations Across Registers
A nonprofit operating in more than one country registers separately in each jurisdiction, usually under a different name and always under a different identifier — an EIN in the United States, a charity number in England and Wales, an ABN in Australia. There is no universal cross-register key, so matching the same organisation across two countries' data means matching on name, address and stated activity, and accepting a non-trivial false-match rate. Build that uncertainty into the analysis rather than presenting a fuzzy match as a confirmed link.
For the neighbouring public-sector sources see government data, and for statutory company records, startup data.
Reach Each Charity Regulator Locally
Live PXM2 locations — pick the jurisdictions whose charity sectors you research:
France
India
Poland
Frequently Asked Questions
Where does nonprofit data actually come from?
Regulators, overwhelmingly. In the United States, tax-exempt organisations file an annual information return that is public, and several services republish the whole corpus in structured form. England and Wales have the Charity Commission register, Scotland has OSCR, Australia has the ACNC, and Canada publishes charity returns through its revenue agency. Each holds income, expenditure, trustee names, activities and filing history for organisations above its thresholds.
Why is a local exit useful for regulator sites?
For the same reason it is useful across the public sector: these are national services built for domestic users, and several degrade or restrict access from foreign address ranges, apply harder bot mitigation to them, or render search behaviour differently. Reaching each register from inside its jurisdiction is simply the most reliable way to see what a domestic researcher sees.
How current is any of this?
It is not, and that has to shape the analysis. Annual returns are filed after a financial year ends, often near the deadline, and regulators then take time to publish. A twelve to eighteen month lag between activity and its appearance in the register is normal. Nonprofit data answers questions about trajectory and structure rather than about what is happening this quarter, and treating it otherwise produces confident nonsense.
What is the biggest gap in nonprofit datasets?
Small organisations. Every regulator sets thresholds below which a charity files a much shorter return or none at all, so the smallest organisations are systematically underdescribed. Any sector analysis built from filings therefore over-represents large charities. That is not a collection failure and cannot be fixed by collecting harder, but it must be stated, because otherwise the dataset quietly misdescribes the sector.
Are trustee names a problem?
They are personal data in a public register, which is the same position as company officers. Publication for regulatory transparency is not the same as unrestricted reuse, and several regulators say so explicitly in their terms. Purpose and retention matter, and it is worth taking your own legal advice before building anything that aggregates named individuals across registers.
Related Mobile Proxy Guides
Charity registers behave like the other statutory sources in this cluster and are usually researched alongside them.
Business use cases
Core mobile proxy guides
Reach Each Charity Regulator Locally
Dedicated 4G/5G modems with unlimited bandwidth and unlimited rotations — carrier IPs inside the jurisdictions whose registers you research.
Get a Mobile Proxy