Published guides put freelance grant writers at about $30 to $150 an hour, with mid-level writers near $50 to $95, and $1,500 to $5,000 for a standard foundation proposal. AI should change how you price, not necessarily how much: quote a fixed fee for the deliverable, because hourly billing pays you less each time drafting gets faster.
Treat those figures as planning ranges, not a survey. The two guides behind them come from an AI grant-writing software vendor and a grant-listing service, and both sell something close to what you do. Whatever you charge, never base it on a percentage of the award: the Grant Professionals Association and AFP codes both rule that out, and the worked quote below uses a fixed fee.
What two published guides say, and who wrote them
I compared the two most recent guides I could verify. Grantable, which sells AI grant-writing software, states that its figures are industry ranges consistent across published rate guides as of July 2026, not quotes. GrantWatch, a grant-listing service, published its guide on 28 October 2026 and calls its numbers approximate planning ranges, not fixed industry prices.
| Item | Grantable (July 2026) | GrantWatch (28 October 2026) |
|---|---|---|
| Hourly rate | $30 to $150 overall: junior $30 to $50, mid-level $50 to $95 with about $75 as the midpoint, senior $95 to $150 or more | $50 to $125 or more |
| Letter of inquiry or simple application | $400 to $1,200 | $500 to $1,500 |
| Standard foundation proposal | $1,500 to $5,000 | $1,500 to $5,000 |
| Large public-funder application | $5,000 to $15,000 or more | $5,000 to $10,000 or more for a complex one |
| Retainer | $2,000 to $6,000 a month, usually covering one to three proposals plus strategy | Custom, no figure given |
The overlap is the useful part. Both guides put a standard proposal between $1,500 and $5,000 and most hourly work between about $50 and $125. GrantWatch also gives a flat-fee example worth copying: $2,500 for a foundation application including the narrative, a budget review and two rounds of revisions. What neither guide does is say whether its prices assume AI-assisted drafting. That is the gap this tutorial fills, and your own time log fills it better than any guide.
Set your own floor before you lean on anyone's range
A published range cannot tell you what you need to earn. Start from your own end of the sum: target income plus overheads, divided by the hours you can honestly bill. Say you want $72,000 net, your software, insurance and professional fees come to $6,000, and only 1,000 hours a year are billable once admin, sales and unpaid proposals are taken out. That is ($72,000 plus $6,000) divided by 1,000, or $78 an hour. It is your floor. Any quote that works out below $78 an hour on an honest hours estimate loses you money, however it compares with a guide.
The figure lands inside the guides' mid-level band of about $50 to $95, which is a reasonable sign. If your floor comes out above $125, a better price will not fix it. You need a narrower specialism, such as large public-funder bids, where the fees are higher because the work is.
The Complete Excel Course with AI Integration
From formulas and pivot tables to Power Query, DAX, VBA and AI with Copilot, ChatGPT and Claude — hands-on from the first lesson.
By Sagnik Bhattacharya and Virginia Thorn, Primetime Emmy Award winner
Buy nowHourly, fixed fee or retainer: who carries the risk
The choice of model decides who pays when a job takes longer than expected, and AI changes the maths for each.
| Model | Best for | Who carries an overrun | What AI does to it |
|---|---|---|---|
| Hourly | Editing a client's own draft, advice calls, post-award reporting | The client | Your billable hours fall, and some clients start asking for an AI discount |
| Fixed fee per application | Standard proposals with a clear brief | You | Faster drafting widens your margin, but you need a revision limit to cap overruns |
| Phased fixed fee | Larger bids where fit is uncertain | Shared | The draft phase shrinks while the checking phase grows |
| Retainer | A steady pipeline of one to three bids a month | Shared | You can take on more bids within the same hours, so state a capacity |
| Percentage of the award | Never | Not applicable | Not applicable: both professional codes rule it out |
A phased fixed fee on proposals, plus an hourly rate for everything with open scope, is a sensible default. The phases give a client a cheap first step, and they give you a go or no-go decision before you invest 30 hours in a bid for which the organisation turns out to be ineligible.
Why hourly billing shrinks when AI drafts faster
The trap is easy to miss because AI genuinely does speed up drafting. Below is the time log, with illustrative hours, for a standard foundation proposal written two ways. Drafting drops sharply. Checking goes up, because a draft that arrives in minutes still needs every claim and figure matched to the client's own documents.
| Task | Hours, manual drafting | Hours, AI-assisted drafting |
|---|---|---|
| Discovery call and eligibility check | 2.5 | 2.5 |
| Funder guidance and past awards research | 4 | 3 |
| Outline and answer plan | 3 | 2 |
| First draft of the narrative sections | 14 | 6 |
| Budget and budget justification | 5 | 4 |
| Two revision rounds with the client | 6 | 5 |
| Fact and figure check against client documents | 2 | 4 |
| Formatting, portal entry and submission | 3.5 | 3.5 |
| Total | 40 | 30 |
At the $75 midpoint both guides circle, the manual job bills $3,000 and the AI-assisted job bills $2,250. The client receives an identical document, arguably a better-checked one, and you earn $750 less. The more you automate, the less you are paid. A fixed fee of $3,000 turns the same 30 hours into an effective $100 an hour, which is where a writer who has invested in a method should be. The budget step stays slow on purpose: the model can word a justification, but the arithmetic belongs in a spreadsheet, as the tutorial on drafting a budget narrative with AI and reconciling every figure explains.
A worked quote for a mid-size foundation proposal
This is the quote a freelancer might send a small charity for a mid-size foundation application. The $3,000 total sits in the middle of the published $1,500 to $5,000 range, and the phases keep the client's first commitment small.
| Phase | What is included | Fixed fee |
|---|---|---|
| 1. Fit check | Discovery call, funder guidance read, written go or no-go on eligibility | $350 |
| 2. Draft | Narrative, budget justification, two revision rounds, outcomes section | $2,400 |
| 3. Finish | Final fact check, portal entry, submission support | $250 |
| Total | $3,000 |
Beneath the table, a good quote lists four things. First, assumptions: the client supplies a fact sheet, past reports and the budget within five working days, and the quote lapses if the funder's deadline is under three weeks away. Second, limits: a third revision round is $250. Third, payment: half on acceptance and half on submission, whatever the funder decides. Fourth, an AI-use statement, covered below. Extras sit outside the fixed fee: rush delivery inside ten working days adds 25 per cent ($750 here), and each post-award report is $400 when the client supplies the data in your template.
A realistic mistake shows why a fixed fee has to attach to the result, not the hours. Picture a freelancer who quoted "about 25 hours at $75" for a small application, then used AI to finish the draft in far less time. The invoice came to 16 hours, or $1,200, against the $1,875 the client had budgeted. The client was delighted and asked for the same price on the next two applications. The freelancer had cut their own income by a third and set a precedent for it. The mistake was not using AI; it was pricing the hours instead of the result, which is the one thing the client was actually buying.
Building the quote from the brief, not from a feeling
Clients accept a fixed fee more readily when they can see what drives it. Grant fees move with the complexity of the bid more than with the size of the award, so build the quote from a base and add-ons. These figures are illustrative and should be replaced with your own after a few jobs.
| Component | Fee | Why it is there |
|---|---|---|
| Base fee, private foundation proposal | $2,000 | Narrative, standard budget and submission |
| Partner and match-funding letters coordinated | +$300 | Chasing third parties is slow and cannot be automated |
| Logic model and outcomes framework | +$250 | An extra deliverable that needs client input |
| Three-year budget with narrative | +$450 | More lines to reconcile and justify |
| Quote | $3,000 | The same figure as the phased quote above |
Showing the build also makes negotiation honest. If the client drops the logic model, the fee falls by $250 and nobody feels cheated. If they ask for a rush, the 25 per cent surcharge applies to the full $3,000.
Before you can build that quote, you need six answers from the client, and asking for them is part of the service. Which funder and which call? What is the amount being requested? Is there a previous application or a funded example to build from? What attachments are required, and who holds them? Who signs off, and by when? What is the real deadline once the client's own internal review time is counted? Each answer moves the fee or the schedule, and a client who cannot answer the first three is telling you the fit-check phase is where to start. Take the answers in writing, before the quote, and quote only from them.
No percentage-of-award fees: what the two codes say
The Grant Professionals Association's code of ethics says members shall not accept or pay a finder's fee, commission or percentage compensation based on grants, and shall take care to discourage organisations from making such payments. Its footnotes define a commission as a flat-rate fee or a percentage paid for services when a grant is awarded. AFP's current Code of Ethical Standards takes the same line for fundraisers: pay may include bonuses or merit pay in line with an organisation's practices but may never be based on a percentage of funds raised, and members decline to receive or pay finder's fees, commissions or percentage-based compensation. The GPA's ethics page sets out the full code, including a separate rule that any performance bonus must follow the usual practice of the member's own organisation and never be based on a percentage of grant money.
These are professional codes, so they bind members rather than everyone, but clients and funders read them as the standard. Both guides above warn against percentage pricing, and one says most funders prohibit paying writers from grant funds, which is worth checking in each funder's rules before you assume the grant covers your fee. The arithmetic is also unattractive to the client: ten per cent of a $60,000 award is $6,000 for the same 30 hours that a fixed fee prices at $3,000.
When a client asks for no win, no fee, a short reply works. Explain that your professional standards do not allow pay that depends on the award, offer the phased fee so the first step costs $350, and point out that a fixed fee means you are paid for the quality of the application rather than for the funder's decision. A flat bonus paid only if the bid wins is no way round the rule, because the GPA's definition of a commission includes a flat-rate fee paid when a grant is awarded.
Six situations and what to quote
A letter of inquiry only. Published guides range from $400 to $1,500. A fixed $650 fits a job that took about eight hours by hand and about six with AI, and it makes a good entry point for a cautious client.
A repeat client on their third application this year. Their fact sheet, past answers and budget template already exist, so discovery and research shrink. A 20 per cent reduction takes the $3,000 job to $2,400 and rewards loyalty without touching your rate.
A rush job. Add 25 per cent, which is $750 on a $3,000 base, and refuse the job if the deadline leaves no time for the fact check. That check is the step AI makes more important, not less.
Reviewing a client's own AI-written draft. Price it as review: $75 an hour with a six-hour minimum, or $450. Require the funder's guidance and a fact sheet, because verifying claims against sources is most of the job.
Post-award reporting. A fixed $400 per report when the client supplies project data in your template, hourly when they do not. Drafting the narrative from data is where AI saves real time, so the grant report template tutorial pairs well with this add-on.
A retainer for two proposals a month. Two proposals at $3,000 plus four hours of strategy come to roughly 64 hours of your time. Priced at ten per cent under the $6,000 the two proposals would cost on their own, with the strategy hours included, that is $5,400 a month, or about $84 an hour, near the top of the $2,000 to $6,000 retainer range the guides quote. State the capacity: two proposals a month, with extras billed separately.
What AI costs you, and why it is not a line item
The tools are cheap next to the fee. ChatGPT Plus is $20 a month, Claude Pro is $20 or $17 a month on annual billing, and Google AI Pro is $19.99. One of the guides above notes that $0 to $150 a month of AI software is roughly one to two hours of a mid-level writer, so itemising it on a quote invites haggling over a rounding error. Put it in your overhead.
Where the choice matters is confidentiality. Client financials and strategy documents belong in a business plan, such as ChatGPT Business or Claude Team, which do not train on business content by default, whereas the consumer plans rely on a privacy setting you must switch off yourself. Both business plans need at least two seats, so a one-person practice pays for two (Claude Team lists $25 a seat monthly or $20 annually), about $40 to $50 a month, which is still under two per cent of one $3,000 job. The comparison of AI grant writing software costs for freelancers goes through the dedicated tools.
When a client asks for an AI discount
It will happen, usually politely. Do not defend your hours, because you have just seen that your hours fell. Defend the deliverable: a checked, funder-specific application submitted on time. Here is a reply you can adapt. It uses placeholders, not real names.
Hi [first name],
Thanks for asking. I do use AI tools to draft faster, and I check every
figure and claim against your documents, which takes longer than the
drafting now does. The fee covers the finished, checked application and
submission support, not the hours.
If the budget is tight, I can offer a smaller scope: a fit check and
outline for $350, or a letter of inquiry only. I would rather scope the
work down than discount the part that protects your application.
Kind regards,
[your name]If clients keep pressing, the wider conversation is about pricing models across consulting work, which the tutorial on fixed fee, retainer or value pricing after AI covers.
Check your effective hourly rate after five projects
Every fixed fee is a bet on your hours, so keep a log. After five jobs the average tells you whether the bet is paying. This is a filled-in log with illustrative figures.
| Project | Fixed fee | Hours logged | Effective rate |
|---|---|---|---|
| Foundation proposal | $3,000 | 31 | $96.77 |
| Letter of inquiry | $650 | 6 | $108.33 |
| Foundation proposal, repeat client | $2,400 | 28 | $85.71 |
| Post-award report | $400 | 4 | $100.00 |
| Large public-funder application | $5,000 | 52 | $96.15 |
| Total | $11,450 | 121 | $94.63 |
The weakest line is the repeat client at $85.71: the discount was too generous for the work saved, so next time cut it to 15 per cent. The strongest is the letter of inquiry. If your target is $90 an hour and the average comes out near $95, the model is working. If it falls below, change the scope limits before you change the rate, because overruns usually come from unlimited revisions rather than from a low fee.
What to do with the hours AI frees
Fixed fees let you keep the time AI saves, and the saving shows up as capacity. With 120 billable hours a month, 40-hour proposals mean three a month, or $9,000 at $3,000 each. At 30 hours each you can fit four, or $12,000, a third more revenue, provided you have clients to fill the slot. If you do not, spend the spare ten hours on things clients value and competitors skip: a mock review that scores the draft against the funder's criteria before submission, a post-award calendar of reporting dates, and a client fact library that makes next year's bid cheaper to write.
An AI-use clause for your proposal
Clients and funders both care how AI was used, and a short clause protects you. Funder rules on AI differ and are still changing, so read each call for any statement on AI use, and the series has a guide to using AI for grant writing without funders rejecting it. For your side of the relationship, adapt this wording.
AI use: I may use AI tools to prepare drafts. All AI-assisted text is
reviewed and edited by me. Every figure, date and factual claim is checked
against documents you supply, and nothing is submitted without your
approval. Your information is processed only in business accounts that do
not use content for model training by default. If the funder requires a
statement about AI use, I will help you draft it truthfully.Whether to mention AI at all is a question for any freelancer, and the tutorial on whether freelancers should tell clients they use AI sets out the case for saying so up front. The short version for a grant writer: clients who find out later feel misled, and funders who ask need your honesty to be on record.
Grant writer pricing questions
Should a grant writer charge hourly or per project?
Per project suits most applications, because the client wants a known cost and AI makes your hours less predictable. Keep hourly for editing someone else's draft, open-ended advice and post-award reporting where scope is hard to fix. Either way, set a minimum fee and log your real hours on every job so you can see your effective hourly rate.
Can I charge a success fee on top of a fixed fee?
Not under the GPA code. The Grant Professionals Association says members shall not accept or pay a finder's fee, commission or percentage compensation based on grants, and it defines a commission as a flat-rate fee or percentage paid when a grant is awarded, so a flat win bonus counts. AFP's code also rules out pay based on a percentage of funds raised. Keep every fee payable whatever the funder decides.
Do I have to tell clients and funders I use AI?
Tell clients in your proposal, because it affects data handling and quality checks. For funders, read each call for any statement on AI use, because policies differ from funder to funder and are still changing. If a funder asks, answer truthfully, and remember that the person who signs the application carries responsibility for every claim in it.
How many hours does a grant application take?
Published guides do not agree. One describes short applications as taking several hours and detailed proposals dozens, without giving a range. In the worked example here a standard foundation proposal takes 40 hours without AI-assisted drafting and about 30 with it, but your own time log after five projects is the only figure that should drive your quotes.
What if a client's budget is below my fee?
Offer a smaller scope rather than a lower rate: an eligibility check and outline, a review of their own draft, or a letter of inquiry only. Published guides put letters of inquiry at roughly $400 to $1,500, so a fixed-price first step can fit a tight budget without you discounting the full application.
Further reads
- Grant Writing AI Agent: What It Can Draft in a Business Grant Bid — What an AI assistant can draft in a business grant bid, and what it cannot.
- ChatGPT Prompts for Grant Writing: Needs to Outcomes, Fact-Checked — Prompts for the needs and outcomes sections, fact-checked.
- How Can AI Help Freelancers? Proposals, Invoicing and Late Payers — Proposals, invoicing and late payers handled with AI.
- Virtual Assistant Pricing When AI Does the Drafting: Rates and Packages — The same repricing problem in another freelance trade.
- Subtitling Rates per Minute: What to Charge When AI Drafts First — Per-unit pricing when AI produces the first draft.
- Best AI for Freelancers: Check a Client Contract for Red Flags First — Check a client contract for risks before you sign.
- Logic Model Template: Build One With AI From a Client's Programme — Build a logic model from a client's programme.
- Freelance Day Rate Calculator: Set Your Rate When AI Speeds Up Delivery — Work out the day rate behind your fixed fees.
- Free AI for Grant Writing: ChatGPT, Claude and Gemini for Business Bids — Where free ChatGPT, Claude and Gemini plans cope with a small-business grant bid, and where their limits get in the way.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Grantable's guide to what a grant writer costs (July 2026), GrantWatch's guide to grant writer fees (28 October 2026), the Grant Professionals Association ethics page, the AFP Code of Ethical Standards (archived copy, October 2026), and OpenAI, Anthropic and Google pricing pages for the tool costs.