Gambling Commission bank data shows 45% of people who paid illegal gambling sites scored as problem gamblers, against 6% of licensed customers. What it means.
Last updated: 7 November 2026
What bank transaction data reveals about illegal gambling and harm
On 28 October 2026 the Gambling Commission published something it had not done before: an analysis of real bank transactions between British residents and illegal gambling operators. The open banking report used anonymised data from 38,000 people who had agreed to share their banking records for research, and linked it to their answers on the Problem Gambling Severity Index (PGSI).
The headline finding is stark. Among people with a detected payment to an illegal operator, 45% scored 8 or more on the PGSI, the threshold for problem gambling. Among all licensed gambling customers in the same dataset, the figure was 6%. This article explains how the study worked, what it found, what it cannot tell us, and what it means if you are tempted by a site that is not licensed in Britain.
How the study worked
The Commission’s research team explained the approach in a blog post published the same day. Rather than relying on surveys or web traffic estimates, the team looked for actual payments to known illegal operators inside a large panel dataset.
| Element | Detail |
|---|---|
| Data source | Open banking transactions from a YouGov panel, collected with consent and anonymised |
| People | 38,000 individuals |
| Period | July 2018 to July 2026 |
| Transactions overall | 157.2 million |
| Transactions with licensed non-lottery gambling firms | 4.8 million |
| Transactions matched to illegal operators | 3,426, made by 179 people |
| People with a PGSI score available | 146 of the 179 |
One detail in the methods section is telling. The team first tried to find illegal operators by fuzzy-matching their names against transaction descriptions, and that failed. Unlike most legitimate businesses, illegal operators rarely or never put their own name in the payment details that reach your bank. The matches that worked came from exact descriptions taken from real bank statements known to involve illegal sites, and from merchant names already linked to them.
Who the 179 people were
Their average age was 44, and 115 of the 179 were men. The report is careful to say the panel is not nationally representative: it slightly over-represents men and under-represents women, especially women over 55, and it only includes people who chose to share banking data for a small incentive. The numbers describe this sample, not the country.
Finding one: illegal gambling is added on top of licensed gambling
A common argument is that people who use illegal sites have “left” the regulated market. The transaction data does not support that. Among people with repeated detected illegal transactions, 91% also gambled with licensed operators. More importantly, their licensed activity did not fall away after they first used an illegal site.
The report tracks average monthly transactions with licensed operators for the 122 people with at least two illegal transactions, before and after their first detected illegal payment:
| Month relative to first illegal transaction | Licensed transactions per person | Illegal transactions per person |
|---|---|---|
| 3 months before | 23.1 | 0 |
| 1 month before | 24.8 | 0 |
| 1 month after | 23.7 | 4.9 |
| 2 months after | 23.5 | 1.8 |
| 3 months after | 25.3 | 1.4 |
| 6 months after | 23.6 | 1.5 |
Licensed play stayed at roughly the same level. The illegal payments sat alongside it, often in short clusters of consecutive transactions. The Commission’s reading is that illegal gambling is embedded in existing gambling behaviour rather than replacing it. For an individual, that means an illegal site is usually extra spending, not a substitute.
Finding two: much higher rates of problem gambling
Because the panel had already answered the PGSI screening questions, the Commission could compare harm scores within the same dataset.
| PGSI group | People with an illegal transaction | All licensed gambling customers (excluding lottery-only) |
|---|---|---|
| Score 0 (no problems) | 26% | 68% |
| Score 1 to 2 (low risk) | 12% | 16% |
| Score 3 to 7 (moderate risk) | 17% | 9% |
| Score 8 or more (problem gambling) | 45% | 6% |
The pattern is reversed between the two groups. Among licensed-only customers, most people show no problems and a small minority score highly. Among people who paid an illegal operator, the largest single group is the one in the highest-harm category. The report stresses these figures can only be compared within this sample, not with the national Gambling Survey for Great Britain.
What this does not prove
The Commission is explicit that the study design does not allow causal claims. It cannot say whether illegal sites cause harm, whether people already experiencing harm seek out illegal sites, or both. What it shows is association: the people who ended up paying unlicensed operators were, as a group, far more likely to be experiencing serious harm.
Finding three: credit cards were rare
Gambling businesses licensed in Britain have been banned from accepting credit cards since 14 April 2020. Illegal operators face no such rule, so one worry has been that they become a back door to gambling on borrowed money. The data offers limited comfort: 4% of detected illegal transactions came from credit card accounts, and 15 of the 179 people had used a credit card with an illegal operator at least once. Most payments came from ordinary current accounts.
Two caveats apply. The dataset only sees payments from accounts that support open banking, so it misses crypto entirely, which is a significant route to offshore sites. And the Commission says many illegal transactions in the data were probably not identified at all. For a wider view of how these sites operate and why protections disappear once you leave the licensed market, see our guide to the real risks of crypto casinos not on GAMSTOP.
How this fits with the Commission’s other illegal market research
This report is one piece of a growing evidence base. A week later, on 5 November, the Commission published survey estimates suggesting 1% of adults in Great Britain, and 4% of people who gamble online, had used an illegal gambling website in the previous 12 months. In her CEO Briefing speech on 5 November 2026, Acting Chief Executive Sarah Gardner described the open banking work as adding to the picture, and set out the scale of disruption activity: 1,005 cease and desist notices and 2,201 websites referred to search engines for delisting between April and September 2026, both already above the totals for the whole of the previous financial year.
The Commission also said the open banking study may be the first of its kind by any regulator. The method cannot size the illegal market, and it cannot find previously unknown operators, but it can follow what happens to real people’s money over time in a way surveys cannot.
What this means in practice
If you are thinking about an unlicensed site
- Check the licence first. Our step-by-step guide shows how to check a UK Gambling Commission licence in a couple of minutes. If a site is not on the public register, it is not legally allowed to take your bets in Britain.
- Expect no backstop. Deposit limits, the ban on credit cards, affordability checks and the complaints route to an independent dispute service all stop at the border of the licensed market.
- Be honest about why. The data suggests illegal sites are most used by people who are already gambling heavily on licensed ones. If a site appeals because it lets you keep going after a licensed operator has slowed you down, that is the warning sign, not the solution.
If any of this sounds familiar
A PGSI score of 8 or more is not a label, it is a measure of harm happening now. Free, confidential support exists and does not require you to stop gambling before you call. The tools licensed operators must offer, including deposit limits, time-outs and self-exclusion, are explained in our guide to responsible gaming, and GAMSTOP self-exclusion is covered in our piece on voluntary self-exclusion. The National Gambling Helpline is free and open around the clock on 0808 8020 133.
Frequently asked questions
What did the Gambling Commission’s open banking study find?
It found 3,426 payments to illegal gambling operators by 179 people in a panel of 38,000. Of those with a PGSI score, 45% were in the problem gambling category, against 6% of licensed gambling customers in the same data. Illegal payments sat alongside continued licensed gambling rather than replacing it.
Does using an illegal gambling site cause gambling harm?
The study cannot say. It shows a strong association between paying illegal operators and high PGSI scores, but the design does not allow causal conclusions in either direction.
Can I use a credit card on an illegal gambling site?
Illegal operators are not bound by the British ban on credit card gambling, and the data shows a small number of people did use one. Gambling with borrowed money is one of the clearest routes to financial harm, which is why licensed operators have not been allowed to accept credit cards since April 2020.
How many people in Britain use illegal gambling websites?
Separate Commission survey estimates published on 5 November 2026 put it at about 1% of adults, and 4% of people who gamble online, over the previous 12 months. The open banking study was not designed to measure the size of the market.
Where can I get help if gambling is a problem?
Call the National Gambling Helpline free on 0808 8020 133, or visit GambleAware. You can also register with GAMSTOP to block yourself from all licensed British sites at once.
18+. Gamble responsibly. If gambling is affecting you, visit GambleAware or call the National Gambling Helpline on 0808 8020 133.
