How do I build a budget for a grant application?

England and Wales: This page sets out the position in full. Scotland: the same position, with no difference in principle — grant terms are funder contracts, and nothing in the terms read for this guide varies with where the charity is registered. Northern Ireland: the same position, for the same reason.

Overspend and underspend: the funder’s own terms

One funder clause shapes a grant budget more than any other, and the treasurer should read it first. The budget presents the figure you reached when you calculated the project’s full cost, including core costs to one funder. Before writing any line, read what the funder does when that figure turns out wrong. The National Lottery Heritage Fund, verbatim:

If you spend more on the project than you were expecting, we are unable to give you more than the grant we awarded.

If you spend less on the project than you were expecting, then we will not pay you the full grant.

Two sentences, two opposite errors, and both of them land on the charity.

Overspend is yours. If the project costs more than the budget said, the funder does not top it up. The difference comes from unrestricted funds the charity may not have. A lean budget submitted to look competitive is a liability written in advance.

Padding is not banked. If the project costs less, you are not holding the surplus. The grant is reduced to what you spent. Building slack into the lines buys nothing and, if the underspend is large, it invites a question about whether the bid was accurate.

What both point at is the same discipline. Budget for the work’s true cost, show your working, and tell the funder early if it changes.

Garfield Weston puts the variation side of that plainly. It asks to be told immediately about proposed changes so it can approve them. It reserves the right to reclaim some or all of the grant if changes happen without its agreement.

Why a grant budget starts from the full cost

Work out what the project costs in full first, then decide what to ask this funder for. Most weak budgets get the order the wrong way round.

The full cost is the direct costs plus a fair share of the overheads — full cost recovery. The method belongs to the money hub’s step to calculate your full cost recovery rate. The number this budget uses comes from there, not from an estimate made fresh for the bid.

Once you have the full cost, the budget is a presentation problem, not an arithmetic one. The lines the funder will pay for come from the funder’s own “what you can spend the money on” page. Anything the full cost includes that this funder will not pay stays in the project. It is covered from somewhere else, and the bid should say where.

Which funders pay core costs in a grant budget?

Each funder decides for itself, and there is no sector rule: it is the single biggest variable in the budget. The table shows each major funder’s published position.

The National Lottery Community Fund row is the one worth quoting. It is the biggest funder of small UK charities, and the answer is commonly assumed to be no. Its own programme page says, verbatim: "We can fund day-to-day running costs. These are sometimes called 'core costs' or 'fixed costs'." Examples given include core salaries, rent, utility bills and equipment.

And the Wolfson row is the one that catches people out. Put a project manager’s salary into a Wolfson bid and you have budgeted a line the funder says, in terms, it does not pay.

The cost nobody budgets for: a grant audit certificate

A grant audit certificate is an independent certificate on how the grant itself was spent. No published budget guidance found for this guide mentions it. It is a real line.

Several major funders require an independent certificate on the grant’s own expenditure. That is over and above any audit or independent examination duty the charity already carries under charity or company law. It is usually needed before the final payment. It is a separate piece of professional work, and it carries a separate fee.

What the evidence supports, funder by funder, is uneven, and this page says so rather than generalising:

• Horizon Europe requires a Certificate on the Financial Statements from an independent qualified auditor, submitted with the final report. The European Commission’s own Model Grant Agreement sets the trigger: Standard threshold (beneficiary-level): requested EU contribution to costs ≥ EUR 430,000 over the whole project. Special (higher) threshold for beneficiaries holding a Systems and Process Audit (SPA) rating: ≥ EUR 725,000. For this funder — unusually — the certificate’s own cost is an eligible cost that can be charged to the grant. Check that row before assuming the certificate is always an extra.
• UKRI, including Innovate UK requires an independently examined statement of account or an independent accountant’s report. The requirement is established in the funders’ own published terms. The banded thresholds widely repeated by accountancy firms are not in those terms, so this guide does not publish them as fact.
• Central government grants — DCMS’s own standard terms require the recipient’s external auditors to certify that the grant was applied in accordance with the grant agreement. No monetary trigger is published in that document, and it is one department’s terms rather than a rule common to all.

For the National Lottery Heritage Fund, Arts Council England, the National Lottery Community Fund and Sport England, this was not established. Their own terms pages did not return readable content when they were checked. This guide publishes no figure and no requirement for those four.

If you are applying to one of them, ask the funder directly whether a certificate is required and what it will cost. Put the answer in the budget.

Whichever funder you are applying to, treat this as a question to ask rather than an assumption to make. “Does this grant require an independent certificate at the end, and is its cost eligible?” is one email, and it is much cheaper before the budget is fixed than after.

Other lines a grant budget needs

A few lines are routinely missed, and each of them is real money:

• Staff time on the grant’s own administration. Reporting, monitoring, evidence collection and procurement paperwork. Some funders pass procurement duties down: competitive quotes above a threshold, documented tender evidence above a higher one. That is somebody’s job, and it is not free.
• The evaluation. If the final payment depends on an evaluation report, the evaluation is a project cost, not an afterthought.
• Irrecoverable VAT. For most charities VAT on project costs is a real cost rather than a recoverable one, so the budget should carry the VAT-inclusive figure. Whether that is true for you is answered when you determine whether you must register for VAT. It turns on whether what you have is a grant or a contract.
• Match funding and in-kind contributions. Match funding is money another source has promised to the same project. Show them, and show them as what they are. Unsecured match funding is a published reason for refusal in its own right, so do not present a hoped-for contribution as a secured one.

On contingency, be careful. A contingency line is not a way of banking slack, because underspend is not paid. Where the funder permits one, it belongs against a named risk with a stated basis, not as a percentage sprinkled across the budget.

When does a grant budget need an accountant?

Routine project budgeting does not need a professional. Building the budget is arithmetic and reading, and a competent treasurer can do all of it.

Three situations are different, and each is specific:

• The grant requires an independent certificate on its expenditure. The funder will expect it to be issued by someone independent and qualified. It cannot be done in-house, and the fee belongs in the budget from the start.
• The project’s VAT position is unclear — particularly where the money may be a contract rather than a grant, or where the charity is partly exempt. Getting this wrong misstates every line in the budget.
• The overhead allocation basis is contested, or a funder has challenged it. That is a conversation about method, and it is worth having with someone who can defend the method.

Before paying anyone, use the free routes. The funder’s own grants team will answer eligibility and cost questions about its own programme, and usually quickly. Your local council for voluntary service, or its equivalent, often has a finance adviser.

The Charity Commission’s and OSCR’s own published guidance on restricted funds and full cost recovery is free and authoritative. Charity-specialist accountants also publish a great deal at no charge. Exhaust those before you commission anything.

Reconciling the grant budget

Finish with three checks, in this order.

1. The budget reconciles to the full cost calculation from the money hub.
2. Every line pays for an output or outcome from the table you made when you set your outputs and outcomes. Every promise has a line paying for it.
3. The total matches the figure in the application form, including the parts of the full cost this funder is not paying.

Then keep the working, not just the total. The grant report at the end of the project is written against these same lines. A budget whose basis nobody can reconstruct in a year’s time is a reporting problem waiting to happen.

The next step is to write the monitoring and evaluation section, which the evaluation line above pays for. After it you assemble your supporting documents.

Common mistake: budgeting optimistically and planning to ask for more later

A lower ask feels more competitive. The Heritage Fund’s own terms close the door on it, verbatim: “If you spend more on the project than you were expecting, we are unable to give you more than the grant we awarded.” The overspend is the charity’s, met from unrestricted funds it may not have.

Common mistake: budgeting generously to build in slack

Contingency feels prudent, and here it is not. The same funder: “If you spend less on the project than you were expecting, then we will not pay you the full grant.” Padding is not banked. It is not paid at all.

Worked example

Saltmarsh Trust budgets a project at what it believes the work will cost. Then it trims the figure before submitting, on the view that a smaller ask stands a better chance.

The project runs over by roughly a tenth. Saltmarsh asks the funder to cover the difference. The funder answers in the words already in Saltmarsh’s own grant agreement: it cannot give more than the grant it awarded. The overrun comes out of unrestricted reserves.

The mirror case is worth sitting with, because it looks like the safe error and is not.

Had Saltmarsh padded the budget instead and come in under, the unspent portion would not have been retained. The grant would have been reduced to what was spent. An inaccurate budget never leaves the charity better off. That is the argument for building it from the full cost and defending the figure.