Reducing Month-End Safeguarding Pressure

How to reduce month-end safeguarding reporting pressure

Month-end pressure is not really a reporting problem. It is a symptom of the work that should have happened earlier in the month, still needing to happen at month-end.

There is a specific week, or in demanding months a specific weekend, when the safeguarding team of a payments or e-money firm is quieter than usual, and not in a good way. Screens are open longer. Coffee is bad. Someone in finance is asking whether the numbers from the customer ledger match the numbers from the safeguarding accounts, and someone else is trying to explain why a break from the 12th of the month was corrected but is no longer easy to trace, and someone else is emailing the bank about an acknowledgement letter they thought was filed but cannot immediately find. Somewhere, the head of compliance is waiting to sign off, and the CFO is asking, gently, whether they can have the figures by Wednesday.

That is the monthly safeguarding return. Or rather, it is the two days that most firms mistake for the monthly safeguarding return, when what they are actually doing is trying to reconstruct a month of activity in a short enough window to hit a fixed deadline. The return itself, the form, the numbers, the attestation, is comparatively straightforward. It is the assembly work behind it that generates the pressure, and the assembly work is a symptom, not a cause.

This article is about how to reduce that pressure by treating it as what it is. Not a reporting problem, but a design problem. Month-end pressure is the visible surface of four upstream weaknesses that show up in the same week every month: data assembled too late, breaks resolved too late, evidence gathered too late, and sign-off compressed into a window that is too narrow for the review it is meant to provide. Each of those weaknesses can be pushed upstream. The return, when it comes, then becomes a summary of controlled activity, not the moment control gets applied.

What the monthly safeguarding return actually is

Before the diagnostic, a short grounding in what the return itself asks for. Under the safeguarding regime introduced by Policy Statement PS25/12 and in force since 7 May 2026, payments and e-money firms are expected to produce a monthly safeguarding return that captures the firm’s position on relevant funds, its reconciliation activity, its discrepancies and their status, its arrangements with banks and providers, and the attestation of a senior individual that the return is accurate. The return then goes to the FCA.

The important thing about the return is that it is a summary. It does not produce information the firm should not already have. Every number on the return corresponds to something the firm should have been able to produce on demand at any point during the month: the safeguarding balances, the customer positions, the reconciliation outcomes, the discrepancies identified, the corrections made, the evidence supporting them. In principle, producing the return should be an act of retrieval and aggregation. In practice, it is often the moment those numbers get assembled into a coherent picture for the first time, because during the month they lived across ledgers, bank portals, PSP exports, spreadsheets and the working memory of the people who produced them.

That gap is what generates month-end pressure. It is closable.

Data assembled too late

The first source of month-end pressure is that the picture of what the firm owes customers, and what it holds to protect them, has not been kept together during the month. The data exists. It has always existed. It is scattered across the customer ledger, the bank portals for each safeguarding account, the payment processor exports, internal cashbooks, FX conversion records and any number of spreadsheets that have accreted around the process. When the return is due, someone has to bring all of that together, in one place, in a form that reconciles internally.

The upstream fix is that the picture is kept together every business day rather than assembled at month-end. When the internal safeguarding position and the external bank and third-party records are compared daily, month-end no longer involves that comparison. It involves confirming that the last day of the month reconciles like every other day of the month, and producing the aggregate from data that has been kept current throughout. The two-day exercise becomes a two-hour one.

There is a secondary benefit. When data is assembled daily, the person who most recently touched a discrepancy is still available to explain it. When it is assembled at month-end, that person is often on the wrong shift, in the wrong week or, on occasion, in the wrong quarter. Institutional memory is a poor substitute for a record that was captured while the events were fresh.

Breaks resolved too late

The second source of pressure is that reconciliation breaks identified during the month were not resolved cleanly at the time, and now need to be traced, corrected and evidenced under time pressure. Some were flagged, worked on and closed properly. Others were flagged and left in a shared inbox. Others were noticed on a Wednesday and then, in the way of small operational discrepancies, quietly aged into the background as more urgent work displaced them.

At month-end, all of those breaks re-emerge. The unresolved ones need to be corrected before the return can be attested to. The partially resolved ones need to be traced, and the trace usually involves reconstructing an activity chain that has since been overwritten by later transactions. The properly resolved ones need their evidence checked, and often the evidence is thinner than the firm remembers.

The upstream fix is that break resolution happens as breaks are identified, not as month-end approaches. Every break carries an owner, a severity, a deadline and a closure test, and moves through investigation, correction, evidence and sign-off within the window the firm’s own policy requires. At month-end, the review of break activity becomes an exercise in reading closed items rather than closing open ones. The pressure of resolving under time constraint disappears, and with it the tendency for corrections to be made hastily rather than properly.

Evidence gathered too late

The third source of pressure is that the evidence supporting reconciliation activity, break resolution and sign-off has not been captured as the work happened. It has to be assembled now, from memory, email threads, screenshots taken after the fact and summaries written in the days before the return is due.

This is where a lot of the credibility risk lives. Evidence assembled late is evidence that depends on the person assembling it. A reconciliation output produced from a screenshot taken this morning does not prove what the firm’s records looked like on the 8th of the month. An email confirming that a break was reviewed does not by itself prove who authorised the correction or when. A summary written a week after the event fills in the gaps in the record with the assembler’s best recollection. Each of those substitutions is small in isolation and, over a full month of activity, adds up to a picture whose evidentiary support is thinner than it needs to be.

The upstream fix is that evidence is captured by the systems that do the work, at the time the work happens. Reconciliation runs are logged when they run. Break resolution stages produce their own records as each stage completes. Sign-offs are timestamped and attributable to the person who made them. Month-end evidence work becomes a matter of retrieval, not creation. What the auditor and the FCA see is what happened, not what the firm remembers happening.

Sign-off compressed into a window

The fourth source of pressure is that the reviews and approvals which should have flowed through the month get compressed into the final days. A senior manager who should have signed off ten decisions across four weeks signs off ten decisions in one afternoon. A head of compliance who should have reviewed the resolution pack progressively reviews it in one sitting. A CFO who should have attested to numbers they have been watching all month attests to numbers they are seeing in aggregate for the first time.

None of that is inherently wrong. It is inherently high risk. Compressed sign-off is the moment when reviewers are least able to spot the specific thing they are meant to be spotting, because they are reviewing volume under time pressure. It is also the moment when the appearance of oversight is easiest to produce and the substance of oversight is hardest to demonstrate. An auditor who sees ten approvals dated within a two-hour window on the 28th will have questions about the quality of each.

The upstream fix is that sign-off happens at the point of the action, not the point of the report. When a break is resolved, its closure is signed off then and there, by a person with the right authority, against the closure test the firm has defined. When a reconciliation is completed, its acceptance is recorded at the time. At month-end, senior management attests to the aggregate, having seen the components as they were produced. The review is real, and the record shows it.

Why this matters at leadership level

Reducing month-end pressure is a leadership question, not just an operational one. The design of the operating model, and the discipline with which controls actually operate through the month, sits with the CFO, COO and head of compliance in most firms. When the safeguarding team is under pressure at month-end, that is not primarily a story about a stretched team. It is a story about how the operating model has been designed and what senior management is signing off.

There is a specific accountability dimension here. The monthly return is attested to by a senior individual. Under the safeguarding regime, that attestation is not a formality. It is a personal statement that the return reflects the firm’s actual position on relevant funds, its reconciliation activity, its discrepancies and their resolution. Attesting under time pressure, to numbers seen in aggregate for the first time, is a different exercise from attesting to a summary of activity the attester has had visibility over throughout the month. Both are formally the same act. Only one of them is a defensible act.

The other leadership dimension is what the pattern says about safeguarding culture. Firms whose operating model produces a smooth month-end tend to be firms in which safeguarding is treated as a continuous discipline rather than a periodic exercise. Firms whose operating model produces a difficult month-end tend to be firms in which control is applied episodically. That distinction is invisible on the return itself and becomes obvious in an audit or a regulatory review. Leadership designs the operating model that produces one or the other.

What this changes in practice

Read together, the four upstream fixes describe a specific operational pattern. Data is assembled daily. Breaks are resolved as they arise, with defined ownership and a closure test. Evidence is captured by the systems that do the work, at the time the work happens. Sign-off happens at the point of the action, not the point of the report. Month-end then becomes what the regime intends it to be: a summary of controlled activity, produced from data already assembled, evidence already captured, breaks already closed and sign-offs already recorded.

This is the work Imperium(L) Prism is built for. Prism keeps the picture together daily, wraps every reconciliation break in a structured workflow with timestamped sign-off, captures evidence automatically as work happens, and generates the data behind the monthly safeguarding return from that controlled activity. The return is produced from the platform, reviewed inside it, and exported for the senior individual to attest to, rather than reconstructed from spreadsheets and email threads in the final week. The technology behind Prism has supported businesses in live, high-volume environments for over 12 years, so the operational logic has already been tested against real reconciliations and real audits.

The monthly safeguarding return is a summary. It is a summary of a month’s controlled activity, or it is a summary produced under time pressure by people trying to reconstruct that activity in the last week. The difference is not in the return. It is in what the firm did with the twenty-nine days before it. Firms that design their operating model around the return tend to find that month-end is a long week. Firms that design their operating model around continuous control tend to find that the return, when it comes, more or less produces itself.

Frequently Asked Questions

What is the monthly safeguarding return?

The monthly safeguarding return is a return that payments and e-money firms produce under the safeguarding regime introduced by Policy Statement PS25/12, in force since 7 May 2026. It captures the firm’s position on relevant funds, its reconciliation activity, its discrepancies and their status, its arrangements with banks and providers, and the attestation of a senior individual that the return is accurate. It goes to the FCA.

Does Prism submit the monthly safeguarding return to the FCA?

No. Prism generates the data behind the monthly safeguarding return from controlled reconciliation activity, supports review and approval workflows, and exports the return in a structured format for the firm’s senior individual to attest to. The firm remains responsible for submitting the return itself, unless direct submission is separately configured and confirmed.

Who attests to the monthly safeguarding return?

The attestation is made by an appropriate senior individual in the firm, confirming that the return reflects the firm’s actual position and activity. Attestation under the safeguarding regime is a personal statement, not a formality, and is one of the reasons continuous oversight through the month, rather than one-off review at month-end, matters.

How can a firm reduce the time pressure of month-end safeguarding reporting?

By pushing four upstream weaknesses earlier in the month: assembling the safeguarding picture daily rather than at month-end; resolving reconciliation breaks as they arise, with defined ownership and a closure test; capturing evidence by the systems that do the work, at the time the work happens; and running sign-off at the point of each action, not the point of the report. Month-end then becomes a summary of controlled activity rather than the moment control is applied.