Gambling Commission Bank Data Finds Illegal Gamblers Keep Using Licensed Sites

The Gambling Commission has traced 3,426 payments to illegal operators through consenting people's bank records and found that those making them did not abandon the licensed market. Of those with a survey score, 45% scored 8 or more on the Problem Gambling Severity Index, against 6% of licensed gamblers in the same data.

• The Gambling Commission identified 3,426 transactions with illegal gambling operators, made by 179 people, in anonymised open banking data from a YouGov panel of 38,000 people covering July 2018 to July 2026
• Of the people with repeated detected illegal transactions, 91% also gambled with licensed operators, and the Commission saw no substantial shift away from the licensed British market after a first illegal payment
• Among those with a detected illegal transaction and a survey score, 45% scored 8 or more on the Problem Gambling Severity Index (PGSI), compared with 6% of licensed non-lottery gamblers in the dataset
• Only 4% of the illegal transactions were made on credit cards, which licensed operators have been barred from accepting since April 2020, while 96% came from current accounts
• The Commission says the data cannot size the illegal market or show that illegal gambling causes harm, and the study follows illegal-market figures acting chief executive Sarah Gardner set out last week

The Gambling Commission Finds Illegal Payments Inside Licensed Gambling Habits

The Gambling Commission published "Illegal gambling: findings from an open banking data analysis" on 28 October. According to the report, YouGov gathered the data from people who consented to their open banking transactions being used anonymously for research, and the Commission bought it after anonymisation. It spans 38,000 individuals between July 2018 and July 2026 and 157.2 million transactions, 4.8 million of them with licensed non-lottery gambling companies. Licensed gambling payments were classified with an algorithm developed by Yaspa, and Warwick Business School provided technical support.

Fuzzy matching of known illegal operators' names against transaction descriptions failed, which the report says may indicate that illegal operators rarely or never include their name in the information they give payment providers. Exact matching against bank statements known to contain illegal payments found 821 transactions, and matching the merchant name field against names strongly suspected of illegal use found 2,628, with 23 in both sets. The 3,426 unique transactions were made by 179 people with an average age of 44, 115 of them men.

Overlap With the Licensed Market and Higher Harm Scores

Of the people with repeated detected illegal transactions, 91% also used licensed operators. For the 122 with at least two, average monthly transactions with licensed operators ranged from 23.1 to 28.8 in the six months before the first illegal payment and from 19.4 to 25.3 in the six months after, while detected illegal transactions averaged 4.9 in the first month and 1.2 to 1.9 thereafter. The Commission concludes that illegal gambling is "embedded within existing gambling behaviour rather than triggering a sustained shift away from the licensed market".

PGSI scores were available for 146 of the 179 people. Of these, 45% scored 8 or more, compared with 6% of the 12,543 licensed non-lottery gamblers in the data, while 26% scored zero against 68%. Because the panel is not nationally representative, the Commission says, the figures support only a within-sample comparison and cannot be set against the Gambling Survey for Great Britain (GSGB).

Of the illegal transactions, 138 (4%) were made on credit cards, by 15 people, and 3,288 (96%) came from current accounts. The Commission banned gambling businesses from accepting credit cards from 14 April 2020.

What the Commission Says the Data Cannot Show

The panel slightly over-represents men and under-represents women, especially those over 55, and includes only people who shared their data for a small incentive, the report says. Cryptocurrency payments are not captured, and only a subset of illegal operators is covered, some not believed to be currently operating in Great Britain. The Commission therefore says the data cannot estimate the size of the illegal market, in absolute terms or relative to licensed gambling, cannot establish whether illegal gambling causes harm, and cannot identify previously unknown operators. No spending values for the illegal transactions are published.

On 20 October the Commission published Phase 3 of its Consumer Voice research by Yonder Consulting, which found that seeking sites not blocked by GamStop was among the strongest predictors of illegal website use. The same day, at the International Association of Gaming Regulators conference, acting chief executive Sarah Gardner said the Commission disrupted 1,134 websites in 2025/26 and that questions added to the GSGB would give "the best picture so far of the scale and trends of the illegal market", according to the published text of her speech.

A September 2024 study by Frontier Economics for the Betting and Gaming Council (BGC) estimated that £2.7 billion a year is staked online with black-market operators, 2.1% of the £128 billion staked with regulated remote operators. It found that 5.4% of gamblers surveyed used both regulated and black-market operators and 0.8% only the black market, and described the roughly £14.8 billion that dual users stake with regulated operators as potentially vulnerable if they shift their spending.

Both Datasets Show the Illegal Gambler Usually Multi-Homes

Frontier's survey and the Commission's bank data reach the same picture from opposite directions: most people who use illegal sites also use licensed ones. The BGC-commissioned work treats dual users' regulated stakes as money at risk of leaving, while the Commission saw no substantial move away from licensed operators in the six months after first illegal contact. The Commission's own caveats limit that rebuttal, because missed operators and cryptocurrency make its illegal counts a floor and the trend rests on 122 people. Channelisation claims in either direction still lack a measurement both sides accept, a problem far from unique to Britain.

The 45% Figure Is a Strong Signal on a Narrow Base

A gap of 45% against 6% is hard to dismiss, even on 146 people. The study cannot say whether illegal sites draw people already in difficulty or deepen their harm, and the Commission says so. The Phase 3 finding on GamStop points to the first explanation for at least part of the group. Either way, these people were still transacting with licensed operators, where protections such as financial risk assessments can reach them, which complicates the argument that friction in the licensed market simply pushes the most vulnerable gamblers out.

Payment Descriptors Are the Next Enforcement Front

If illegal operators keep their names out of payment data, banks and payment providers cannot easily block what they cannot see, and 96% of detected payments came from ordinary current accounts. That puts weight on intelligence sharing with payment firms, alongside police raids and the court powers to suspend IP addresses and domains that Gardner said are "coming soon".

The Commission now has direct transaction evidence on some of the people using the illegal market in Britain, and it undercuts the simplest version of the exodus argument. It has not measured the market's size, and on its own account this method cannot.