Most finance systems record what came in. Very few record what was promised.

An award letter is a commitment with conditions attached. Some of those conditions are about delivery, some about evidence, some about the timing of claim submissions. The money arrives only when someone completes the paperwork for each stage. If that person leaves, gets busy, or assumes a colleague has done it, the claim sits there quietly.

The bank statement gives you no help here. You cannot see an absence. A £40,000 award drawn in full and a £40,000 award never claimed look identical in a cashbook, because in both cases there is nothing further to reconcile against.

This is the reason we treat the funding register as part of the bookkeeping rather than a document the programme team keeps in a folder. It lists every award, the amount, the claim stages, the conditions and the cash received to date. Each month someone ticks the cash against the award. Any line where the two have drifted apart gets a name against it and a date by which it will be chased.

The routine takes minutes once the register exists. Building the register for the first time is the hard part, because it means reading every award letter you have signed.

In practice the money goes missing in a small number of recognisable ways.

• The claim was never submitted. Delivery happened, evidence existed, and the form was never filled in. This is the most common and the most painful, because the work was already paid for out of reserves.
• The award was treated as one payment when it had stages. The first instalment arrives, everyone relaxes, and the second and third stages fall off the calendar.
• The income was coded as something else. Software often offers a single catch-all heading for anything that is not a sale, so restricted funding, contract fees and donations land in the same place and stop being distinguishable.
• Grant and contract were confused. A grant is support towards your activity with conditions. A contract is payment for goods or services delivered. They are recognised differently and evidenced differently, and getting them the wrong way round distorts what you report to a funder and what you report to a board.

None of these require carelessness. They require a normal organisation where the person who signed the award is not the person who does the bookkeeping, and no routine connects the two.

The obvious cost is the cash. The less obvious costs do more damage.

Unclaimed funding means your reserves quietly paid for delivery the funder had agreed to cover. That shows up later as a cash squeeze nobody can explain, usually at the point where you are deciding whether to renew a post or commit to a lease.

Then there is the board conversation. If restricted funds are not tracked separately, trustees cannot tell whether spending in a programme was covered by its award or subsidised from general funds. They are being asked to approve decisions on figures that do not answer the question in front of them.

Then there is the funder relationship. Funders notice when claims arrive late, arrive incomplete, or do not arrive at all. A late claim reads as an organisation that cannot evidence its own delivery, whatever the truth is. That impression carries into the next application.

In the case we found, the claims were completed and submitted. The organisation now runs a funding register updated monthly and reviewed at each board meeting. The register was worth more than the recovered money, because it means the next award cannot slip the same way.

Here is what we put in place when a grant-funded organisation comes to us, and what you can run yourself without waiting for anyone.

1. List every live award. Funder, total awarded, claim stages, conditions, reporting dates, end date. Every award you have signed in the last three years, including ones you believe are finished.
2. Reconcile cash against each award line. Take the actual receipts in the bank and match them to award stages. Any award where cash received is less than cash claimable is your first call tomorrow.
3. Code restricted funds separately in the books. One heading per fund, not one heading for all funding. Then the question “where did that award go” has an answer you can print.
4. Put the claim dates in the same calendar as payroll and VAT. A claim deadline is as fixed as a filing deadline and deserves the same treatment.
5. Review the register at every board meeting. One page, awarded against claimed against spent. Trustees ask better questions when they can see the gap.

The first pass is uncomfortable. Every organisation that has done it has found at least one thing it did not expect.

Accounts prepared months after the year end will eventually show the deferred balances and the unspent restricted funds. By then the claim window has usually closed.

That is the argument for monthly management information rather than an annual filing exercise. Recognising grant income correctly matters for the accounts, and there is real technical detail underneath it: whether conditions are performance related, whether income should sit as a liability until delivery, how capital funding is released against the life of the asset. Get that wrong and you can create a tax charge on money you already spent, which is a very bad surprise for an organisation without reserves to meet it.

But the recognition question is downstream. The reason grant income gets missed is upstream, in the bookkeeping, where nobody was asked to check that a promised payment ever turned into an actual one.

We ask that question as a habit. Asking how a cost was being funded is exactly how the £200,000 was found. It was not a clever piece of analysis. It was reading the award letters and then reading the bank.