How to Manage Grants for a Nonprofit: The Full Lifecycle, Step by Step

What are the steps in managing a grant?

Grant management is not one task but a cycle that repeats every year. Small organisations often treat it as writing applications, which is only one step. The steps below are the whole job, in the order they usually happen. Several run at once: you will be reporting on last year's grants while scoring next year's funders.

• 1. Strategy: decide the role of grants in your income, a target, the funder types you prioritise and a limit on dependence on any one funder.
• 2. Prospect research: find funders and score their fit against an eligibility gate and weighted factors.
• 3. Pipeline and calendar: track every opportunity by stage with a next action, and every deadline with an internal date.
• 4. Case for support: keep a reviewed library of need statements, theory of change, indicators, profiles and policies.
• 5. Writing: letters of inquiry, concept notes and proposals in each funder's format and limits.
• 6. Budget: direct costs, indirect costs at your true rate, match funding and a budget narrative.
• 7. Decision: due diligence documents, thanks after a yes, feedback after a no, and a careful read of the grant agreement.
• 8. Compliance: restricted fund tracking, procurement, visibility, budget change approvals and close-out.
• 9. Reporting and stewardship: honest interim and final reports, stories with consent, and contact between reports.
• 10. Rhythm: a weekly meeting, monthly finance checks, quarterly reviews and an annual plan.

How do you decide which funders to apply to?

Start with eligibility, because it is a gate rather than a score. Are you the type of organisation the funder supports? Do you work where it funds? Does your work match a stated priority, in its words? Can you legally receive this kind of money, for example foreign contributions or government funds? Does the timing work? Is anything on its exclusions list true of you? One honest no ends the assessment.

Then score fit on a few factors from 1 to 5, with a line of evidence for each: mission, geography, grant size, how similar past grantees are to you, timing and relationship. Weight mission most heavily. The factor most often skipped is past grantees. Guidelines describe what a funder hopes to fund; its list of recent grants shows what it actually funds. If none of its grantees look like you in size, place or approach, a perfect mission match is not enough.

Set two thresholds on the weighted score: one for Apply and one for Maybe. Apply on a Maybe only when you can talk to the funder first or the past-grantee evidence is reasonable. Record every no-go and its reason. Over a year, those notes stop you returning to the same poor fits.

How do you keep track of grant deadlines and the pipeline?

Keep one pipeline with a row per opportunity. Give each a stage (research, letter of inquiry sent, invited, submitted, awarded, declined), the amount requested, a next action and a date. No open row without a next step. Record staff days spent so you can see, over a year, what each type of funder costs you to pursue.

Multiply each open amount by a probability for its stage to get a weighted pipeline. Treat the probabilities as placeholders until you have a year of your own success rates, and never budget on the weighted figure. Its value is comparison: is the weighted pipeline growing, and does it come near the gap between what has been awarded and the year's target?

Put every date in one calendar: letter of inquiry and proposal deadlines, report due dates, renewal windows, document renewals and board meetings. Set an internal deadline at least five working days before each funder deadline, so the finance check, the director's read, the attachments and the portal upload do not happen in the last hour.

What should a grant proposal include?

Follow the funder's format first: its questions, order, headings and limits. Where it leaves the format to you, a strong proposal covers a short summary, who you are, the need, your approach and theory of change, objectives with indicators, activities and timeline, monitoring and learning, safeguarding and risk, sustainability, organisational capacity and the budget.

Write the need from the side of the people affected, with two or three facts from named sources. Describe the change as a results chain: activities lead to outputs, outputs to outcomes, outcomes to long-term change, each link resting on assumptions you state. Many bilateral and multilateral donors ask for a logframe, which puts that chain in a table with indicators and means of verification. Make indicators SMART: specific, measurable, achievable, relevant and time-bound, and choose ones your monitoring system can actually collect.

If your report will later be judged against evaluation criteria, it helps to know them now. The OECD's Development Assistance Committee publishes six widely used criteria: relevance, coherence, effectiveness, efficiency, impact and sustainability. A proposal that shows how each will be evidenced is easier to score and easier to report on.

How do you budget a grant at full cost? A worked example

Full cost is the direct cost of a project plus a fair share of the indirect costs that keep the organisation running: finance, management, office, IT, governance and audit. In a 2009 Stanford Social Innovation Review article, Ann Goggins Gregory and Don Howard described the nonprofit starvation cycle, in which unrealistic expectations about overhead lead organisations to under-report and under-fund it, weakening the systems that good work depends on.

Find your true indirect rate once a year: divide your shared support and governance costs (and fundraising costs, if you choose to include them) by your direct programme costs. Then build each budget at full cost and compare it with what the funder will pay. Here is a worked example in US dollars for a small project. Direct costs are $80,000. The organisation's true indirect rate is 22%. The funder will pay 90% of the direct costs and allows indirect costs of 10% on what it pays. The remaining 10% of direct costs is match funding the organisation brings.

In this example the funder's money covers about 41% of the project's true indirect cost, and $10,400 must come from somewhere else. That figure should reach the director and board before submission, not after the award. Some government funders publish a default indirect rate for organisations without a negotiated one; in the United States, the Uniform Guidance allows a de minimis rate of up to 15% of modified total direct costs. Know your own rate regardless, because the difference is real money.

What happens after a grant is awarded?

Thank the funder within 48 hours. Read the whole grant agreement with your finance lead and turn every condition into an owner and a date: payment schedule, reporting dates, budget change rules, procurement, visibility and branding, equipment, data and consent, audit, records and unspent funds. If a clause is unclear or unusual, ask the funder and take advice.

Set up a restricted fund or project code in your accounts and load the budget by line. Each month, after the accounts close, compare the share of the grant spent with the share of the grant period gone. A large gap either way is a conversation with the funder, early. Before moving money between budget lines beyond what the agreement allows, ask in writing and wait for approval.

Start close-out about 60 days before the end date: check the remaining budget, request a no-cost extension in writing if you need one, reconcile the grant to the accounts, follow the funder's instruction on unspent funds, and archive records for the period your agreement and law require.

How do you write grant reports and keep funders engaged?

Report against what you promised, using the same indicators as the proposal. Name problems along with what you did about them; reports that admit difficulty are usually more credible than reports that describe only success. Ask your monitoring team for data two weeks before your internal deadline.

Stories show what numbers cannot, but consent is a hard rule. Explain in plain words how a story or photo will be used, ask for consent in writing or record verbal consent, let people refuse without any effect on the help they receive, and let them withdraw. For children, a parent or guardian consents, the child is asked too, and your safeguarding policy decides what can be shared.

Between reports, keep in touch: an informal update every two or three months, an invitation to visit, a year-end thank-you and a renewal conversation six to nine months before the grant ends. If you want an outside check on how you handle the judgment calls between donors, team and board, AssessAll's Three Rooms: Donor, Team and Board assessment is one option.

Can you use AI to write grant applications?

AI tools can help with parts of the work: summarising long guidelines, reading a funder's grants list for patterns, outlining a theory of change from programme notes, checking budget arithmetic, cutting text to a limit and compiling a first draft of a report from data you supply. They also invent plausible numbers and sources, and in grants an invented figure is a serious problem.

Use a simple human review rule: AI drafts, summarises and checks; a named person who knows the work checks every fact, figure and claim and signs it off. Never paste personal data about participants, staff or donors into a tool your organisation has not approved for that use. Read each funder's policy. UK Research and Innovation, for example, asks applicants to declare substantive use of generative AI and not to use it to generate an entire application.

Common mistakes in grant management

Most of these are system failures rather than writing failures. A scored funder list, one pipeline, a reviewed content library, a full cost budget and a monthly check prevent nearly all of them. To build those habits into a personal development plan for a grants officer, see how a Jobulary development plan works at jobulary.com/how-it-works. The Grant Lifecycle Kit from Bodhih Training packages the workbook, templates and checklists for each step.

• Applying outside the funder's priorities or ignoring its exclusions.
• Writing the need about the organisation instead of the people it serves.
• Statistics with no source, or out of date.
• Outcomes that are really activities, and indicators nobody can measure.
• A narrative and budget that do not match, or arithmetic errors.
• Leaving out indirect costs, then running down reserves to pay for them.
• Submitting in the last hour with no internal review.
• Moving money between budget lines without approval.
• Late reports, and silence between reports.
• Letting AI write the application and not checking the facts.