The Door Is Shut. The Demand Walked to the Phone

A channel closed, and casino marketing inherited the problem#

On Wednesday Flutter confirmed it is reviewing up to 100 Paddy Power shops, with about 400 roles at risk. The Racing Post put that at roughly a fifth of its United Kingdom and Ireland estate, and iGaming Times sized the estate at 506 shops.

Read that as a den closing rather than a business shrinking, because the customers do not close with it.

It is the third such announcement this year. Yogonet reported Evoke closing around 270 William Hill shops in April, with its chief executive specifying 230 in the closing programme itself, and Racing Better counts Betfred's 132 from the summer. Add them up and roughly five hundred doors close in one year.

The compliance fact behind it is real and, for this piece, only texture: remote gaming duty rose from 21 to 40 per cent in April 2026, with online sports betting duty due to rise to 25 per cent in 2027, per Bettors Insider and London Loves Business. Flutter told the Racing Post it expects a $320 million earnings impact in 2026 before $85 million of mitigation.

The number that matters is not the tax, it is the scent#

Every one of those shops had regulars. They are not lost customers; they are unaddressed ones, and reaching them again is now a casino marketing problem rather than an estate problem. That is the quiet inheritance, and it lands on whoever owns acquisition.

This is an acceleration, not a break in the trail#

The tempting story is that a tax killed the high street overnight. The regulator's own numbers do not support it, and a casino marketing strategy built on that story will misjudge the timing.

The Gambling Commission recorded 5,995 betting shops in Great Britain for April 2022 to March 2023, already a 3.9 per cent fall on the year before and a 22.0 per cent fall on the last pre-lockdown period. By the October to December 2025 quarter the count was 5,669.

Evoke has been blunt about the arithmetic ahead of it too, warning through London Loves Business that the announced changes could add up to £135 million a year to its duty costs from 2027. That is the pressure behind the pace, and it is not going to ease.

So the estate had shed roughly 326 shops before this year's announcements, and the 2026 wave stacks another five hundred on top of a decline running for more than a decade. Evoke said as much itself, telling Yogonet the closures were part of a broader shift towards a more efficient retail footprint as customer behaviour moves online.

Read that split carefully. Remote casino, betting and bingo produced £2.1 billion of gross gambling yield in the quarter against £1.2 billion across all land-based sectors. The hedgerow had already thinned. This year simply cut it back where everyone could see.

For anyone planning paid user acquisition, that distinction is the difference between a shock and a schedule. A shock demands emergency spend. A schedule can be planned for, and the operator that plans wins the quieter, cheaper quarter.

Why you cannot simply buy the quarry back#

The instinct when a channel closes is to buy the equivalent volume somewhere else. In this market that instinct is expensive, because the two channels acquire completely different people in completely different ways.

A shop customer arrived by walking past. There was no bid, no creative, no attribution window and no cost per acquisition. Replacing that with digital means entering an auction against every other operator doing exactly the same thing in exactly the same quarter, which is the least favourable moment to buy anything.

Operators cannot simply buy gambling traffic to replace a shop, either. Gambling advertising sits under tighter platform rules than almost any other category, so the inventory available is narrower, the approvals are slower and the creative is constrained before a single pound is spent. That is precisely the awkward, regulated ground folkfox works on.

Approvals are the hidden tax on speed here. A regulated creative that needs legal review does not move at auction pace, so an operator reacting in November is bidding with last quarter's assets while a competitor who prepared in September is bidding with the right ones. In this thicket, preparation beats budget more often than it should.

There is a second cost nobody puts in the plan. A retail regular is not a digital customer wearing a coat. The product is different, the session is different and the bonusing economics are different, so casino marketing that treats reacquisition as a media buy rather than an onboarding problem will pay twice: once for the click, again for the churn.

Casino marketing ideas that survive contact with the closure#

Strip out the noise and this week hands an operator a short list, none of which is glamorous and all of which is cheaper than an auction.

Note what the table does not settle. The Evoke figure is reported as around 270 announced with 230 in the closing programme itself, which is why two reputable outlets have printed two different numbers for the same event. Casino marketing plans should carry the range rather than pick the tidier figure.

First, capture the estate before it closes. In-shop signage, staff conversations and account linking convert a walk-in regular at a fraction of what the same person costs through an auction three months later. This is the single highest-return move available and it has a deadline attached.

Second, treat the closure towns as distinct markets rather than a national average. Third, budget retention against the reacquired cohort separately, because a player who joined under duress churns faster than one who chose you.

A closing shop is also the one moment when an operator has the customer physically present and undivided, which is a luxury no digital channel offers at any price. Squandering that window and then bidding for the same person in December is the most expensive sequence available, and it is the one most operators will run.

Fourth, and least comfortable, accept that some of that audience will not follow you at all. Racing itself is counting the same cost: the Racing Post reports Betfred's closures alone were estimated at about £4 million in lost Levy and media rights, and that one independent operator's cost of showing racing rose by approaching 75 per cent in five years, from £40,000 a shop to almost £70,000.

Flutter's own statement, given to Casino.org and others, called the estate a key part of the business in communities across the United Kingdom and Ireland, and the review an extremely difficult decision. Read commercially rather than sentimentally, that is an operator telling its customers the door is closing and not yet telling them where to go instead.

The opening this leaves for a patient prowl#

Every competitor faces the same auction at the same time, which is usually a reason to avoid it. It is also, for once, a reason to be early.

The brush is thick and the light is going, which is exactly when a careful operator gains ground on a hurried one. Casino marketing rewards the team that moved in September over the team that panicked in December, and this year the difference is measurable.

The operators that come out of this well will be the ones that did the unglamorous work before the shutters came down: postcode mapping, in-shop capture, a separate retention plan for the reacquired, and a measurement window honest enough to show which of those actually worked. That is a casino marketing strategy. Buying more impressions in December is not.

The trail from here is well lit, if nobody enjoys walking it. Roughly five hundred doors close, the regulator's own series shows the decline predates the tax, and the money moved online years before the estate did. Casino marketing ideas that assume a sudden crisis will be timed wrong; the ones that treat this as the visible end of a long, quiet drift will land.

More each morning in the newsroom, the vertical on iGaming marketing, the auction side on PPC, the social side on paid social, positioning on brand strategy and rates on pricing. The quarry did not vanish. It moved to a smaller screen, and it is still waiting to be found.

Frequently asked questions#