DRAFT. This document has no template — it is ours. It is also the one that actually protects us on accuracy: it turns "the field is empty" from a defect into a described property of the product, disclosed before purchase.

Where our data comes from

Version 1.0. Effective 23 August 2026.

This statement forms part of our Terms of Use. It describes how iGaming Finder compiles its information, and — more importantly — what that method cannot do. Read it before you rely on anything you find here.

1. Our sources

We compile information from three kinds of source.

1.1 Official regulator publications. Gambling regulators publish registers of the licences they have issued: licensee names, licence numbers, statuses, and in some jurisdictions the address, filed contact details and named directors of the licensee. We collect these publications from the regulator's own website or published files, on a recurring schedule.

1.2 Publicly accessible websites. We visit websites that are publicly reachable without an account and record what a visitor can observe: that a domain resolves and responds, what it presents, which third-party services it loads, and how that changes over time.

1.3 Third-party data providers. For some company and people information we use commercial data providers. Where a record comes from such a provider, we are bound by that provider's terms as well as our own.

We do not buy our core licence data from a data broker. The register information in the Service is collected by us from the regulator's own publication. We say this because it determines who you should ask when a record looks wrong: see section 5.

2. What we do to it

Raw publications are not comparable across regulators. Between collection and what you see, we:

• parse each source's own format, which each regulator chooses and changes without notice;
• normalise names, statuses, dates and jurisdictions into one shared shape;
• match records to each other — a domain to a company, a brand to an operator, a licence to a licensee.

Matching is inference, not fact. It is our best judgement from the evidence available, it is not confirmed by the regulator or by the company, and it is wrong some of the time. Anywhere the Service joins two things together, treat the join as a lead to verify, not as a verified relationship.

3. Known limitations

These are not hypothetical. They are properties of the method, and we would rather you learn them here than from a bad decision.

3.1 Coverage is partial and uneven. We cover the jurisdictions and sources listed on our coverage page, not the whole market. A jurisdiction may be absent because its regulator publishes nothing usable, publishes in a form we have not yet handled, or restricts reuse. Coverage changes in both directions.

3.2 Absence is not evidence of absence. If a company, domain, licence or relationship does not appear, the correct reading is "we do not have it", never "it does not exist". Do not use the Service to establish a negative.

3.3 Fields are unevenly populated. Regulators publish different fields, and some publish a field for some records only. Dates in particular — when a licence was issued, changed or ended — are frequently not published at all. An empty field means we have no value, not that the value is empty.

3.4 Website-derived information is an observation at a point in time. What we record about a site is what our collection observed on the date shown. Sites change, block automated visitors, serve different content by country or device, or fail to load. Some observations do not reproduce on a later visit. Every such record carries the time it was observed; that time is part of the information.

3.5 Signals inferred from a website are probabilistic. Where we report that a site appears to use a particular platform, provider or service, that is an inference drawn from publicly visible evidence such as the resources the page loads. It can be wrong in both directions: an integrator, a reseller, a shared host, a copied template or a logo on a marketing page can all produce a misleading signal. Where we can express confidence, we do. Confidence is not proof.

3.6 Freshness depends on the source. We re-collect on a schedule, but a regulator may publish daily, monthly, on no schedule at all, or stop. The Service shows when a record was last observed. Between that time and now, anything may have changed. Nothing in the Service is real-time.

3.7 We do not verify what the regulator published. If a register contains an error, a stale entry or a duplicate, that error will appear in the Service. We reproduce what was published; we cannot confirm it and cannot correct it at source.

3.8 Identity is hard and we sometimes get it wrong. Company names repeat, change, and differ between registers. Two records we treat as one company may be two companies, and one company may appear as two records.

4. What we do not do

To be unambiguous, the Service does not:

• verify the identity, ownership, solvency or good standing of any company or individual;
• confirm that a licence is valid, current or applies to a given activity or market;
• screen anyone against sanctions, PEP or adverse-media lists;
• make any compliance, regulatory, credit or eligibility determination;
• provide legal or regulatory advice.

If you need any of those, use a service licensed and warranted for it. This one is not.

5. Corrections

If a record is wrong at the source, tell the regulator. We cannot change what a register publishes, and we will keep reproducing it until the source changes.

If a record is wrong in our reproduction of it — a parsing error, a bad match, a value we invented by mistake — tell us at legal@igamingfinder.com. Include the record and what you believe is correct. We will investigate and, if we agree, correct it and re-collect from source. We do not commit to a response time, and we do not accept liability for a record that was correct at source and wrong here, beyond correcting it.

If you are an individual and the record is about you, see our Privacy Policy. You do not need to justify the request.

6. Changes to sources

A source may change its format, restrict reuse, move behind an account, or stop publishing. When that happens the affected coverage may become stale, be frozen, or be withdrawn from the Service. Where a source is held or degraded and we know it, we say so on the coverage page rather than serving stale data as if it were current.

Version 1.0 — 23 August 2026. This statement forms part of the Terms of Use and is subject to the same versioning and acceptance process.

Cookie settings