Everyone is late, which should tell you something
Talk to anyone who administers grants and you will hear the same thing: reports come in late, and they come in late from organisations that are otherwise well run.
When a failure is that widespread, individual discipline is not a satisfying explanation. Something about the shape of the task is producing the outcome, and if that is true then the fix is structural rather than motivational.
This is not a defence of lateness. Late reports have real costs, and this piece gets to them. It is an argument that the usual remedy — trying harder next quarter — does not work, and that a small number of specific changes do.
What follows is what funders are actually checking, five reasons the material arrives too late to check it with, and the four changes that address them.
What they are actually checking
Reporting requirements can run to several pages, which obscures how few questions are really being asked. Underneath the forms there are three.
| The real question | How it appears | What satisfies it |
|---|---|---|
| Did the money go where we agreed? | Expenditure by heading | A schedule that maps to the budget |
| Can you show me? | Evidence on request | The document behind any figure |
| Does the story match the money? | Narrative plus schedule | Two reports describing one quarter |
The third row is the one recipients underestimate. A funder reads the narrative and the schedule together, and the mismatches are obvious from the outside in a way they never are from the inside — an activity described in detail that carries no cost, a heading full of spend that the narrative never mentions.
Everything else in a reporting pack is administration around those three questions. Knowing that helps when a form asks for something that seems pointless: it is usually a proxy for one of the three, and answering the underlying question directly is often accepted.
Five structural reasons
None of these is about anyone being lazy, and each one has a different fix — which is why “we’ll start earlier next time” reliably fails.
The documents live with the people who spent the money, not with the person writing the report.
Costs belonging to the period keep arriving after the period ends.
Nobody owns the deadline, because everyone involved has a different main job.
The narrative and the numbers come from different places and different people.
The deadline was set before anyone knew how long the work actually takes.
Each gets a section. The first is the largest by a considerable margin.
The documents are in other people’s pockets
A funded programme spends money through the people doing the work. A project worker buys materials. A facilitator pays for a room. Someone drives somewhere and buys fuel. Each transaction produces a document, and each document lands wherever that person happens to be.
Three months later, someone else has to assemble those documents into a claim. They were not there when the money was spent, they do not know what half of it was for, and the documents are in inboxes, glove compartments and photo libraries.
This is the single largest consumer of grant reporting time and it is almost never budgeted for. The figures are usually available — the card statement shows what was spent. What is missing is the evidence and the explanation, and both are held by someone who has moved on to next week’s work.
Notice how badly the incentives sit. The person holding the receipt bears the cost of handing it over and gets none of the benefit. The person who needs it has no way to compel it and often no idea it exists. Nothing in that arrangement improves through exhortation.
The fix is not a policy. It is making the handover cost two seconds at the moment of spending — a photograph to one place, no form, no categorisation, no naming convention. Anything more elaborate than that will be complied with for a month and then quietly abandoned.
The costs keep arriving after the period ends
A quarter closes on your calendar. Your suppliers have not heard about it. The invoices for the last few weeks of the period arrive over the following month, some of them well after.
This creates a genuine dilemma rather than an oversight. Report on time and the claim is incomplete. Wait for the costs and the report is late. There is no third option that is simply better.
Most organisations resolve it by waiting, without ever deciding to. The report drifts a fortnight past the deadline because someone is waiting for two invoices, and nobody presents that as a choice because it does not feel like one.
It is a choice, and it is worth making explicitly. The alternative — report on time and carry late costs into the next claim — is permitted by most funders, is cleaner, and is almost always better than a fortnight of silence.
What makes it work is agreeing the convention with the funder at the start, when it is a procedural question, rather than at the first claim, when it looks like an excuse.
Nobody owns it
Grant reporting sits between roles. The finance person has the ledger and not the context. The programme lead has the context and not the documents. The director has the relationship and not the time.
Each of them assumes one of the others is on it, and each of them is right that it is not primarily their job. So the work starts when the deadline is close enough to generate anxiety, which is precisely when it is most expensive to do.
The fix is one name against the deadline. Not necessarily the person who does all of it — the person responsible for it happening, who chases the others and notices in week two rather than week eleven.
This is the cheapest of all the changes here and it is skipped most often, because it feels like it should not be necessary in a small team. It is most necessary in a small team, where everyone genuinely does have a different main job.
A useful test: ask three people in your organisation who is responsible for the next grant report. If you get three different answers, or three hesitations, the deadline has no owner.
Two reports, two sources, one quarter
The narrative usually comes from the programme lead, working from what was planned and what they remember. The schedule comes from finance, working from documents. Both are honest and they describe different things.
The plan says three workshops. The documents show two venue hires. The narrative, written from the plan, mentions three; the schedule, built from documents, shows two. Nobody has done anything wrong, and a funder reading both notices immediately.
What makes this expensive is when it is caught. Found before submission, it is a sentence explaining that a workshop moved. Found by the funder, it is an email, a reply, a clarification, and a note on your file that the reports need checking.
Ten minutes of comparison before submission removes it. Does every activity in the narrative carry cost somewhere? Does every heading with significant spend appear in the narrative? Do both describe the same period, including whatever slipped?
And where they legitimately diverge, say so. The divergence is usually the most interesting thing in the report, and a funder much prefers reading about it from you.
The deadline was a guess
Reporting dates are set in the agreement, before the work starts, by people who do not know your supplier payment terms or when your programme lead is on leave. They are usually a round number of weeks after period end, chosen because it sounds reasonable.
Sometimes it is reasonable. Sometimes it lands two weeks after a period whose costs take five weeks to arrive, and the arrangement was impossible from the day it was signed.
Almost nobody says so. The deadline is in the agreement, the agreement is signed, and raising it feels like admitting weakness before any work has been done.
It is worth raising anyway, and the best moment is the start. “Our suppliers typically invoice a month in arrears; would a four-week reporting window work rather than two?” asked in month one is a scheduling conversation. The same point made after two late reports is an excuse.
Funders are more flexible about this than recipients expect, because a report that arrives complete and on a workable schedule is worth more to them than one that arrives fast and gets amended.
What lateness actually costs
Recipients tend to hold two contradictory beliefs: that a late report is a catastrophe, and that everybody is late so it does not matter. Both are wrong in useful ways.
| Situation | Actual consequence |
|---|---|
| One report a fortnight late | Usually nothing, if you said so in advance |
| One report a fortnight late, unannounced | A chasing email and a note on the file |
| Three consecutive late reports | Everything read more closely from then on |
| Late and incomplete | A round of queries, and weeks of delay to payment |
| Late with figures that later change | The most damaging of all |
The first two rows differ only in whether you sent a message. That message costs nothing and it is the difference between a scheduling note and an incident.
And there is a direct financial consequence people forget: where reporting releases the next tranche, late reporting is late money. An organisation with tight cash flow can end up borrowing to cover a gap created entirely by its own paperwork.
The trust cycle, in both directions
This is the part that rarely gets said explicitly, and it matters more over a multi-year relationship than any single report.
A funder who receives complete reports on time reads them lightly. Queries are rare, requests for evidence are occasional, and the relationship is about the work. That is a cheap relationship to maintain, on both sides.
A funder who receives late or amended reports reads everything closely. More queries, more evidence requested, more scrutiny of things that would previously have passed — not out of hostility, but because their own assurance obligations require it once a pattern exists.
Both states are self-reinforcing. Light scrutiny leaves you time to report well; heavy scrutiny consumes the time you needed to improve. Which cycle you are in is largely determined in the first two reports of a relationship.
That is a strong argument for over-investing at the start. The first two claims of a new grant deserve more care than they seem to warrant, because they set the terms for everything after.
It also argues for raising problems yourself. A recipient who says “we found an error in last quarter’s figures” is treated very differently from one whose error is found by the funder — even when the error is identical.
Four things that actually fix it
In order of effect per unit of effort.
One · Capture documents at the moment of spending. A photograph to one place, taken by the person who spent. This addresses the largest cause and it needs no system, only a habit and a destination that is genuinely two seconds away.
Two · Put one name against the deadline. Free, and it converts the work from something that happens under pressure into something that happens on a schedule.
Three · Process in batches through the period. A week or a fortnight at a time rather than a quarter at the end. The extraction takes the same total effort; what disappears is the reconstruction of what each cost was for, which is where the real time goes. The mechanics are on batch processing and on drawdown reporting.
Four · Agree the late-cost convention with the funder at the start. Ten minutes in month one that removes a recurring dilemma for the life of the grant.
Notice that only the third involves software at all, and that the first is both the largest and the cheapest. That ordering is not accidental — the biggest problem in grant reporting is a logistics problem about paper, not a data problem.
The one-person organisation
Everything above assumes several people. A great many funded programmes are run by one person who delivers the work, spends the money and writes the report.
The good news is that reason one mostly disappears: the documents are with the person who needs them. The bad news is that reason three gets worse, because the same person’s delivery work always feels more urgent than their reporting work, and it usually is.
What works at this scale is smaller and more frequent. Fifteen minutes a fortnight — photograph anything outstanding, extract the batch, put the rows in the table — rather than a day at quarter end that will be postponed twice.
And be more willing to ask the funder for a workable deadline. A sole operator with a two-week reporting window after a period whose invoices arrive over five weeks is being asked to do something arithmetically impossible, and funders generally understand that when it is put to them plainly.
The compounding argument matters most here too. A quarter where the fortnightly habit held costs almost nothing; a quarter where it lapsed costs a day. The gap between those two is the entire difference between a manageable programme and a stressful one.
Measuring your own lag
Two numbers, both computable from what you already have, and both more useful than any general claim about how long reporting should take.
Days from period end to submission. Track it for four quarters. A stable number means you have a process, whatever its length; a number that swings wildly means the reports are being produced by whoever happens to have capacity.
Share of costs whose document was captured within a week. This is the leading indicator. It moves months before the submission date does, and it is the only one of the two you can act on directly.
Watch the second and the first will follow. Chasing the first on its own produces a rushed report rather than an earlier one, which is how organisations end up submitting on time and then amending twice.
Both are cheap to keep. The first is a date subtracted from a date; the second falls out of whatever you use to capture documents, since it already knows when each one arrived. Neither requires a new system, and having them turns “we should be better at this” into something with a direction.
One caution about the first number: do not compare it with other organisations. Supplier terms, funder deadlines and team size vary so widely that a benchmark tells you almost nothing. What matters is the direction of your own figure over four quarters.
Three objections
“Our funder does not mind.”
Possibly true of one report and rarely true of a pattern. The consequence is not a complaint; it is a shift in how closely everything else gets read, and that shift is not announced.
“We do not have time to do it fortnightly.”
The fortnightly version is less total time, not more — the quarter-end version includes reconstructing what each cost was for. The objection is real, but it is about attention rather than hours.
“The reporting requirements are disproportionate.”
Often true, particularly for small grants, and worth saying to the funder rather than absorbing silently. Funders do reconsider requirements when recipients explain the cost, and never when they do not.
The third deserves more airtime than it gets. Reporting burden is a legitimate topic in a funding relationship, and the organisations that raise it constructively — with an account of what the requirements actually cost them to meet — are taken seriously more often than the ones that simply struggle quietly.
From the funder’s side of the desk
It is worth understanding what a late report looks like to the person receiving it, because it explains behaviour that otherwise seems disproportionate.
A grants officer is usually managing dozens of relationships against their own internal deadlines. They report upwards on portfolio spend, they have their own year end, and they are accountable for money released against reports they have not yet received.
So a late report is not an inconvenience to them personally. It is a gap in something they have to submit, on a date they do not control, and the number of chasing emails they send is a measure of their own pressure rather than of their opinion of you.
That reframing makes two things obvious. A short message saying when the report will arrive is genuinely useful to them — it converts an unknown into a planned item. And a report that arrives complete is worth more than one that arrives fast and then gets amended, because an amendment means redoing whatever they built on top of it.
It also explains why patterns matter more than incidents. An officer who has been surprised twice starts building slack into their own planning, and that slack takes the form of asking you earlier, asking for more, and looking more closely at what arrives.
None of that is adversarial. It is what anyone would do managing a portfolio of commitments where some of the inputs are unreliable — and it is entirely reversible by being predictable for two consecutive quarters.
What none of this fixes
Suppliers who invoice late
Nothing on your side changes when a supplier bills. What changes is whether that is a planned-for condition or a quarterly surprise.
A receipt that was never kept
No extraction recovers a document that does not exist. That is why the capture habit is first on the list and everything else is second.
A genuinely unreasonable deadline
It has to be renegotiated. Process improvements can absorb a difficult window and not an impossible one.
Staff cost apportionment
The largest heading in most grants comes from payroll and timesheets, and follows your funder's rules rather than any general method.
Deciding what a cost was for
Only the person who spent it knows. Software can carry that answer once it exists; it cannot supply it.
The last one is the honest limit of every tool in this area. The document can be captured, read and filed automatically. What it was for is a fact that exists only in someone’s head, and it has a short shelf life.