For nonprofits, foundations and funded programmes
Those decisions are the organisation’s judgement and its legitimacy rests on them being made by people accountable for them. What can be carried is the machinery around them: which restricted fund may pay for what, which funder report is due, which application is missing a document, and the acknowledgement that has been on somebody’s list for three weeks.
The money arrives restricted. This grant funds programme staff but not administration. That one funds a specific county. This foundation requires a match. That government pass-through has procurement rules attached that the organisation did not know applied until an audit mentioned them.
The mission does not respect those boundaries. A family that walks in needs three things, one of which is fundable by the current grant, one of which is fundable by a grant that ended in June, and one of which nobody funds because it is not a programme, it is a bus fare.
The staff are few and the roles are fictional. The person titled programme director does the reporting, the outcome measurement, some of the fundraising, and covers the front desk on Thursday. Anything requiring a specialist happens badly or not at all.
Reporting consumes a share of capacity that funders systematically underestimate. Each funder wants its own outcomes in its own format on its own calendar, and none of those match each other or the way the organisation actually thinks about its work.
And the board oversees, cares, and meets six times a year. It sees a financial summary and a programme narrative, and the gap between those and the operating reality is where most nonprofit governance failures live.
Restrictions, periods, match requirements and pass-through rules live across grant agreements, a bookkeeping system and institutional memory — so an organisation cannot easily say what it may spend on what, and decisions get made against a guess.
The consequence is not usually fraud, it is drift: a cost charged to the fund that had room rather than the fund that should carry it, discovered at an audit and disallowed years later against an organisation with no reserve.
So the first act is the same one the government budget page describes, at a scale where it matters more: write down every fund with its restriction, its period, its match requirement, its reporting calendar and the rules attached to it. One document. Most organisations under a certain size have never had it and every one of them has the information scattered.
What can then be carried is the administrative band: assembling a funder report from data already held, chasing the sub-grantee or partner document that has not arrived, flagging a fund approaching the end of its period with money unspent, checking a grant application is complete against the funder’s own requirement list, and getting acknowledgements out inside the window.
What never moves: any decision about who receives a grant or a service; any prioritisation of a person, family or applicant by any predicted characteristic; any determination that a cost is allowable against a restriction; and any programme judgement. The first two are what the organisation’s legitimacy rests on and the second two belong to the executive and the board.
Whether the restrictions are written down at all — measured by whether one document holds every fund with its restriction, period, match and rules — most organisations start at zero.
Funds approaching period end with money unspent — measured by count surfaced with lead time, against those previously discovered at closeout.
Time to assemble a funder report — measured by elapsed hours from a report becoming due to submission, per funder.
Applications complete on first submission — measured by count returned by a funder for a missing item, before and after.
Acknowledgements sent inside the window — measured by elapsed days from gift received to acknowledgement sent.
Staff hours on reporting rather than programme — measured by time-on-task sampling across both, taken the same way before and after.
What the board can see between meetings — measured by whether fund position and reporting status are available without staff assembling them.
any decision about who receives a grant or a service; any prioritisation, scoring or triage of a person, family or applicant by any characteristic or prediction; any allowability determination against a restriction; and any programme judgement. Nothing here reads an application to form a view about an applicant, and nothing decides who goes first.
It reads what the organisation actually has, which is frequently a small bookkeeping package, a spreadsheet, a shared drive of grant agreements and a donor database. That is the real starting condition and a design that assumes an enterprise finance system will not survive contact with it.
Restrictions are transcribed from the grant agreements themselves rather than interpreted. Where two documents disagree, that is reported as a question for the executive, because resolving it is a determination.
Nothing becomes a second record of a donor, a grantee or a participant. A second donor record in particular is how acknowledgement failures multiply.
And participant-facing surfaces, where any exist, are built for the population served rather than for the staff — which in most organisations means a phone, a screen reader, and a language other than English.
An organisation this size has no reserve to absorb a disallowed cost, a lost donor record or a missed reporting deadline. The consequences of a vendor failure here land on the people the organisation serves, and that should govern how much is delegated and how fast.
Which is why the first phase is a document rather than an operation, and why the document is useful even if nothing else follows. A written fund map is an asset the organisation keeps.
Participant and donor data stays inside your tenancy, on your retention schedule, exportable by you, and is not used to train anything serving another organisation. Participant data in particular frequently discloses circumstance — housing status, immigration-adjacent facts, health — and is never training material at any tier.
And on assurance: an independent SOC 2 Type II attestation is in progress and no report exists yet. Where a government pass-through attaches a specific safeguarding obligation, terms are agreed in writing during scoping and may close the scope.
The executive director is usually the whole approval process, and the document that matters to them is the fund map — because it is the thing they have needed and not had time to build.
The board treasurer will ask whether anything can touch the books. Nothing can; reconciliation means comparing two of your own figures and reporting the difference to a person who decides.
The board should see the same fund position the staff see, between meetings rather than six times a year. That is a governance improvement independent of anything else here, and several organisations have found it the most valuable part.
And the auditor should review the record format before anything runs, particularly where a government pass-through is involved.
Every current fund mapped to one document — restriction, period, match requirement, reporting calendar and attached rules — transcribed from your own agreements, with no system access required.
This phase can be done from the grant agreements alone. No integration, no credentials, no access to your bookkeeping — which matters when the organisation has neither the capacity nor the appetite to set any of that up.
The map is very often the whole value, and stopping there is the expected outcome rather than a disappointing one. An organisation that gains a written fund map has gained the thing its executive has been carrying in their head, and it survives them leaving.
If it continues, the first operated act is usually the reporting assembly for one funder, because it is the one that returns capacity immediately and risks nothing.
Both are usually true, and the fund map phase is deliberately shaped for it: no integration, no credentials, no recurring cost, and an output your organisation keeps. If the map shows nothing you did not already know, you have lost very little. If it shows a fund ending in eleven weeks with money unspent — which is the commonest finding — it has already paid for itself in a way your board will understand.
Nor should you, and nothing here is. No decision about who receives a grant or a service, no prioritisation of anybody in need, no scoring, at any tier. Your legitimacy rests on those decisions being made by people accountable for them, and a vendor that offered to triage a waiting list would be offering to take the one thing you cannot delegate. If a proposal includes it, refuse it.
No system will, and nothing here tries to harmonise them. What it does is assemble each report in each funder’s own format from data you already hold, which is different from reconciling their definitions. The measurement to watch is hours per report: if it does not fall, the assembly is not helping and the honest answer is that the constraint is the funders rather than the process.
It stays in your systems, on your retention schedule, exportable by you, and is never training material at any tier. And the fund map phase needs none of it — it works from grant agreements alone. Where a government pass-through attaches a safeguarding obligation we cannot meet, the honest answer is that the scope closes rather than that we interpret the obligation generously.