How to Write a Grant Proposal

Reviewers are looking for reasons to say no, because there are more good applications than there is money.

A grant proposal is a persuasive document with an unusually well-defined audience: a panel reading many applications, working from stated criteria, under pressure to reject most of them. Understanding that reading situation matters more than any writing technique, because it explains why clarity and fit beat ambition.

Read the call before having the idea

The most common cause of rejection is misfit — an excellent project sent to a funder that does not support that kind of work. Read the call document, the funder’s stated priorities, and ideally a list of recently funded projects. If your project needs reframing to fit, decide whether the reframing is honest or whether you should find a different scheme.

Where the call names assessment criteria, treat them as your section headings. Panels frequently score against them line by line.

Lead with significance

The opening must establish why the problem matters before it explains what you will do. Reviewers form an impression in the first paragraph, and a proposal that opens with method has answered a question nobody has yet decided they care about.

Significance is specific: who benefits, how much, and what changes if the work is done. Avoid claims of general importance that could preface any proposal in the field.

State the aims plainly and keep them few

Two or three aims, each expressed as something that will be known or achieved, is usually right. Long lists suggest a project that has not been scoped. Each aim should map visibly onto a work package and a deliverable later in the document.

Make the method credible, not exhaustive

Reviewers need enough method to believe the project is feasible and rigorous — not a protocol. Cover design, sample or data, procedure, and analysis, and justify each choice against the aims. Where a technique is unusual, say why it is appropriate and demonstrate that you or a collaborator can execute it.

• Name the design and why it answers the question.
• Justify sample size or data volume.
• Identify what could go wrong and how you would respond.
• Show that access, permissions and ethics are achievable.

Build a budget the panel will believe

Budgets are scrutinised, and errors damage credibility disproportionately. Every line should be justified in the narrative and traceable to the workplan. Under-costing is as damaging as over-costing — a proposal that cannot be delivered for the money requested suggests the applicant has not thought it through.

Check what the scheme actually funds. Many exclude certain staff costs, equipment above a threshold, or overheads, and a budget containing ineligible items is sometimes rejected without further review.

Show a workplan with milestones

A Gantt chart or timeline demonstrates that the work fits the funding period. Include dependencies, and be realistic about approval and recruitment, which routinely take longer than planned. A workplan where everything begins in month one is not credible.

Address risk honestly

Naming the main risks and your mitigation strengthens rather than weakens a proposal. Panels know research is uncertain; what they are assessing is whether you have anticipated the uncertainty. A proposal with no risk section reads as either naive or evasive.

Write for a mixed panel

Most panels include specialists in your area and generalists from adjacent fields. Write so that a generalist can follow the significance and the shape of the work, while a specialist can verify the method. Technical detail belongs in a clearly signposted section rather than spread through the narrative.

Leave time for the unglamorous parts

Letters of support, costings signed off by a research office, data management plans and ethics statements all take longer than expected and are routine causes of missed deadlines. Start them early.

Finally, have someone outside your field read the case for support. If they cannot explain back what you propose to do and why it matters, the proposal is not yet ready — and the generalist on the panel will have the same difficulty.

Show why this team, and this moment

Panels fund people as much as projects. Make the case that you or your team can deliver the work: relevant prior output, access already secured, techniques already demonstrated, collaborators with the missing expertise. For early-career applicants, a named mentor or an institutional commitment carries weight.

Timeliness matters too. Say why the work should happen now — a policy window, a newly available dataset, a method that has just become viable.

Write the lay summary carefully

Most schemes require a plain-language summary, and many applicants treat it as an afterthought. It is often the first thing read, sometimes by a panel member outside the field or by a public contributor with a formal role in scoring. Write it without jargon, without acronyms, and with a clear statement of who benefits.

Plan for what happens after

Funders increasingly assess dissemination and impact rather than treating them as optional. Say concretely what will be produced, who will receive it, and how anyone would know the work made a difference. Vague promises to publish and present are visibly weaker than a named plan with identified audiences.