Removing saved card details at 9pm cut slot top-ups by 27% among a sample of UK players, according to a behavioural trial run across three mid-sized operators in late 2024. The intervention was deliberately blunt: at a fixed evening hour, stored payment credentials were deleted from accounts that had deposited more than £150 in the preceding seven days, forcing a manual re-entry of card details before any further play. The headline figure is an average across operators, and it hides as much as it reveals — but it is one of the cleaner pieces of evidence that friction, not persuasion, moves deposit behaviour.

The trial design matters more than the number. Roughly 41,000 accounts were split into matched cohorts by deposit frequency, average stake size and game type, with a control group that saw no change. The 27% reduction refers to the count of top-up transactions in the fourteen days after the first 9pm removal, not to total spend. Spend per remaining depositor fell by a smaller but still meaningful 11%, which suggests the intervention worked mainly by interrupting repeat small top-ups rather than by stopping a handful of large deposits. Median top-up value in the treatment cohort actually rose slightly, from £20 to £25 — consistent with players who did push through consolidating into fewer, more deliberate transactions.

Why 9pm and why saved cards

The choice of hour was not arbitrary. Operator data consistently shows a deposit peak between 8pm and 10.30pm, driven by post-dinner play on mobile. Saved credentials collapse the gap between intent and transaction to a couple of taps, and it is precisely that gap that most top-up behaviour occupies. A player who has to find a card, read sixteen digits off it and wait for a bank verification step is a player with roughly forty extra seconds to reconsider — and forty seconds is a long time at 9.15pm on a Tuesday.

There is a second mechanism worth naming. Removing a stored card does not just add friction; it breaks the visual cue. Several operators in the trial reported that re-deposit prompts and one-tap top-up buttons lost their anchor when no card was on file, so the interface stopped advertising the action. That is a design change disguised as a compliance step, and it is probably doing more work than the manual entry itself.

What the operators actually changed

• Deletion of tokenised card data from the player's wallet at 21:00 local time for flagged accounts
• Removal of the "top up" shortcut from the lobby for accounts with no saved method
• A single, non-promotional prompt explaining that card details had been removed and could be re-entered in account settings
• No bonus, no free spin, no retention message attached to the removal

That last point is the one most likely to be copied badly. Several operators have since layered a "re-add your card for a £5 bonus" message onto similar friction tests, which converts a behavioural intervention into a re-acquisition campaign and, predictably, erodes the effect.

The compliance backdrop

The Gambling Commission has been circling stored payment methods for two years without landing on a rule. The 2023 remote customer interaction guidance requires operators to monitor and intervene on indicators of harm, and rapid successive deposits sit squarely in that territory. Saved cards make those deposits faster and less visible to the player's own bank, which is why several compliance teams have started treating wallet persistence as a risk control rather than a convenience feature.

The trial results give those teams a defensible number. A 27% cut in top-up frequency is the kind of figure that survives a regulatory conversation, whereas "we added a session timer" rarely does. It also fits the Commission's stated preference for interventions that are proportionate and testable over blanket restrictions — the same logic behind the affordability checks that came into force for online slots in 2024, where £150 a month net loss triggers light-touch checks and £500 triggers enhanced ones.

Worth noting: the sample skewed toward players already flagged by those thresholds. Roughly 62% of the treatment cohort had triggered at least one affordability check in the previous quarter. The intervention is therefore best read as a second-line measure for players showing early friction signals, not as a universal default.

Where the evidence gets thin

Fourteen days is short. Behavioural friction effects decay, and the trial did not run long enough to say whether the 27% holds at eight weeks or whether players simply migrate to another payment method. Early indications from two of the three operators suggest a partial rebound: by day 28, top-up frequency in the treatment group had recovered to about 84% of control. That is still a real reduction, but it is not 27%.

There is also a substitution question the data cannot answer. Some players re-entered the same card and carried on. Others switched to bank transfer, which is slower. A smaller group — the operators estimate 3% to 5% — moved play to a second brand entirely. If friction pushes players between operators rather than out of the market, the harm reduction is real for the individual but the market-level effect is close to zero, and nobody has published cross-operator churn data to settle it.

The confound nobody controlled for

The trial ran in November and December. Slot play is seasonal, and the control group's deposit frequency rose over the same period in line with normal Q4 patterns. The 27% figure is measured against that rising baseline, which makes it a relative reduction — the absolute gap in top-up counts was closer to 19%. Both numbers are honest; they answer different questions. Anyone quoting 27% as an absolute fall is overstating the case, and anyone quoting 19% as the headline is underselling a result that held across three separate operator datasets.

What it means for product design

The interesting conclusion is not that friction works — that was never in doubt. It is that the most effective friction was almost free to implement. No new licensing, no third-party vendor, no change to the deposit flow itself. A scheduled deletion job and a UI tweak. Compare that to the cost of the average safer gambling intervention, which tends to involve either a vendor contract or a customer support team.

That has an uncomfortable implication for the direction of travel. If the cheapest intervention is also among the most effective, operators have a commercial incentive to adopt it — but only up to the point where it starts costing deposits. The 27% is, from a revenue standpoint, a 27% problem. Whether the industry treats this as a harm-reduction tool or a conversion leak to be plugged with a re-add bonus will tell you more about its priorities than any responsible gambling statement.

The open question is who blinks first on making it standard. A single operator removing saved cards at 9pm loses deposits to the brand next door that does not. Until the Commission writes wallet persistence into the guidance — and there is no timetable for that — the intervention will keep appearing as a trial result rather than a default, which is a strange place for a measure with this much supporting evidence to sit.

If you are playing slots and notice your saved card has quietly disappeared, that is probably why. Deposits are the point at which most harm escalates, and anything that adds a pause there is worth more than a banner about setting limits.