Years ago, I interviewed for a contract grant writing position with a VA hospital. The conversation went well. We talked through their funding priorities, their team, and their previous proposals. Toward the end, they offered to pay me on commission.
I told them that was unethical and that I could not work that way.
They told me they would just find someone else willing to do it.
I told them good luck with that, and I walked out the door.
What I took from that meeting was not anger. It was a practice change. From that day forward, I began sharing my rates with prospective clients in writing and asking them to acknowledge receipt before we ever scheduled an interview. If a client could not get past my fee structure, I did not want to spend an hour explaining grant writing to them, only to land in the same conversation I just had.
That experience is one of dozens that have shaped how I think about ethics in this field. And ethics in grant writing is rarely the clean, comfortable territory that workshops and listservs make it out to be. The real questions are the ones working grant writers ask each other privately, in DMs, over coffee, when they are not sure what the right answer is, and they do not want to look unprofessional for asking.
Here are eight of those questions, and how I work through them.
Table of Contents
Is it ever OK to work on commission or contingency?
No. And this is the question I get asked more than any other, so let me be direct.
Working on commission, contingency, or a percentage of the awarded grant violates the ethical standards of every major grant writing professional association, including GPA, AFP, and AGWA. It is also opposed by virtually every major funder, including most federal agencies, which can disallow grant funds used for this purpose. The Grant Professionals Association Code of Ethics is unambiguous on this point, and so are the codes of every adjacent profession that touches charitable fundraising.
The reason it is unethical is not arbitrary. Tying a writer's pay to the outcome of a single proposal creates a direct financial incentive to inflate, distort, and shade the truth in ways that serve the writer's paycheck and not the funder, the organization, or the people the organization is supposed to serve. It also forces the organization to compete with the grant writer for the funder's money, which is backwards.
I have written about this in more depth in a separate article. [LINK to existing contingency article]
The short version: if a prospective client offers to pay you on commission, that is your signal to share your rates in writing, get acknowledgment of receipt, and only schedule an interview after the client confirms they understand and accept your fee structure. If they say they will find someone else, let them. The someone else they find is rarely the writer you wanted to compete with anyway.
Can you write for two clients applying to the same funder for the same deadline?
Yes, and you do not have to tell either of them.
This is the question where I diverge from a lot of grant writing commentary, so I want to walk through my reasoning.
A few years ago, I had two clients, both planning to apply to the same United Way grant cycle, for the same funding pool, for the same deadline. I wrote both proposals. Each one was the best version of that organization's case I could produce. I did not tell either client about the other.
Here is why.
The argument against doing this usually goes one of two ways. Either you should refuse one of the clients, or you should disclose the situation and let them decide whether to keep you. I think both options put the writer's discomfort ahead of the client's outcome.
If I refer one of the clients to another writer, I have no idea whether that writer is going to do work as strong as mine, or even finish the proposal on time. The client trusted me with this opportunity, and I am sending them to someone unknown because I find the situation awkward. That is not better service. It is offloading my problem onto the client.
If I disclose, the most likely outcome is that one or both clients fire me to find a writer with no conflicts, and then they end up with the same uncertainty about quality. I have also created drama where none needed to exist. Funders do not pick proposals based on which writer was used. They pick proposals based on the strength of the case, the alignment with funder priorities, and the organization's track record. Two strong proposals from two strong organizations both deserve to be in the running.
Now, a word about conflict of interest. The GPA Code of Ethics requires grant professionals to disclose actual conflicts of interest, and even the appearance of conflicts of interest, to clients and employers. That is the rule I am working under when I make these decisions, and it deserves a direct response here.
A conflict of interest exists when a grant writer has a competing personal, financial, or professional interest that could compromise, or reasonably appear to compromise, their judgment or loyalty to a client. A board seat at a competing applicant is a conflict, because board service creates a fiduciary duty that runs against a paid contract to write for a different applicant. A close personal or familial relationship with someone on the funder's review committee is a conflict, because the appearance alone is enough to require disclosure. Being paid by the funder while also writing for the applicant is a conflict, even when rare. The common thread is a structural tie that pulls the writer's professional judgment in a direction that does not serve the client fully.
Writing for two clients applying to the same funder is not that. Here is the key point that often gets missed: grant proposals are scored against a rubric, not against each other. Reviewers evaluate each proposal on its own merits, against the published criteria, and assign a score. Two proposals to the same funder are not in a head-to-head competition where my work on one diminishes the other. They are two independent evaluations against a fixed standard. If both proposals score well, both can be funded. If only one scores well, only one is funded, and that outcome reflects the strength of the case and the fit with funder priorities, not which proposal I wrote second.
My loyalty to each client is whole. My effort on each proposal is full. Nothing about the structure of the engagement pulls my judgment away from either one. Wanting both clients to receive funding is a single aligned interest, not a competing one. That is why this is not a conflict of interest, and why disclosure is not required.
What I do instead:
Write each proposal to the highest standard the organization can support. No coasting on either, no recycling language between the two.
Do not bill either client for shared time. If I read the funder guidelines once, that hour belongs to me, not split across two invoices. If I attend a single funder webinar, same. The shared learning is on me.
If a client ever finds out and asks, tell the truth. Both organizations were my clients. I wrote the best proposal I could for each. I am not in the business of choosing winners for the funder.
I have never had a client ask. I have never had a client complain. The warnings I have heard from other grant writers about this practice have not matched my experience.
What do you do when you come in to replace a previous grant writer?
Listen carefully, do not assume, and nip problems in the bud.
Many of my best client relationships started with a conversation about what went wrong with the previous grant writer.
The reasons varied, but they were rarely what people assume. Most of the time, the breakdown was about fit. The client wanted the proposal out the door and the writer wanted more time to refine it. The writer wanted frequent check-ins and the client wanted the writer to work independently. The client expected a quick turnaround and the writer worked on a longer, more deliberate timeline. None of those are quality failures. They are mismatches in working style, communication preference, and expectation, and they are the most common reason these arrangements end.
Sometimes there were performance issues too. Missed deadlines, work that did not match the client's voice, an inability to take feedback. But in my experience those cases were the exception, not the rule. The previous writer was usually a competent professional whose working style did not click with this particular client.
Here is what I do not do. I do not assume the previous writer was bad. I do not assume the client was difficult. I do not gossip about the previous writer with mutual contacts. None of that is useful, and most of it is unkind.
What I do is listen carefully for the specific mismatch in the way the client talks about the previous arrangement. If the client says, "They wanted to keep tweaking it forever," I make a clear schedule with submission dates that respect their pace. If the client says, "They were always emailing me," I set communication expectations up front, including how often they want to hear from me. If the client says, "We needed someone faster," I deliver a tight first draft early to show I can match their tempo.
The warning signs are not always about the previous writer. Sometimes they are about the client. If three previous writers all left, that is information. If the client cannot articulate what went wrong, that is information. If the client expects me to fix problems that are organizational and not writing-related, I name that early and offer capacity building as a separate engagement.
Either way, the only ethical move is to do my best work and let the previous arrangement stay in the past where it belongs.
How do you bill ethically?
Someone in the field once asked whether he should bill a client for time spent thinking about their project in the bathtub.
My answer: if you are seriously considering billing for bathtub thinking time, you are probably not charging enough.
Your rate should be set so that you are only billing for actual sitting-at-the-computer writing and researching time. The thinking, the marinating, the shower insights, the drive-home connections, all of that is built into the rate. If the rate is too low to absorb the cognitive overhead of doing the work well, raise the rate. Do not parse minutes.
Better yet, move away from hourly billing entirely. Hourly billing creates an ethical tension that never fully resolves. The hourly writer has a financial incentive to spend more hours. The client has a financial incentive to want fewer hours. Both parties end up watching the clock instead of focusing on the proposal. Value-based pricing, where you set a fixed fee for the deliverable based on the value to the client, removes that tension. You get paid for the result. The client knows the cost up front. Neither of you is incentivized to game the clock.
If you do bill hourly, here are the standards I hold myself to:
Bill only for time you can clearly attribute to the client's work. Reading the funder guidelines, writing, editing, research specific to their proposal, meetings, drafting attachments.
Do not bill for general professional development. Reading a Chronicle of Philanthropy article that might come up in someone's proposal someday is on you.
Do not bill for rabbit-hole research beyond what the proposal needs. If you were researching sphagnum bogs, got interested in the topic, and spent three more hours reading after you already had what you needed for the proposal in the first 45 minutes, those three hours are on you. The proposal needed 45 minutes of research. Your curiosity needed the rest. The client should not pay for the difference.
Honor the original quote even when scope creeps modestly. If the scope changes meaningfully, have the conversation up front before you do the extra work.
What about referral fees?
I have never paid one. I have never taken one. I do not have a problem with people who do, as long as the arrangement is transparent and the client is not paying more because of it.
If a colleague spends real time vetting a client, learning their needs, matching them to the right writer, and making the introduction, that is professional work and it deserves to be paid. A flat referral fee is a reasonable acknowledgment of that effort. Many fields, including law and consulting, have referral fee structures that work fine ethically as long as they are disclosed.
I make referrals because I want my students and colleagues to succeed, and because building a network of writers I trust makes my own work better. If I receive a request for grant writing services that I cannot take, I interview the organization briefly to understand whether they need a writer or whether they need capacity building. If they are ready and just need a writer, I refer them to one of my top graduates and make myself available for questions or review if the new writer wants feedback.
I have not charged a referral fee for these introductions. Two reasons. First, I want the graduate to build their practice. Second, the goodwill of an unpaid referral tends to come back around. Writers I have referred to have referred clients back to me. The network is the compensation.
One thing I might ask instead: if the engagement goes well, would the graduate be willing to participate in a brief case study I could use to show prospective Certificate students what the work looks like in practice? That trade serves both of us and the field, and no one feels like the referral is transactional.
If you do charge or pay referral fees, disclose the arrangement to the client in writing. In practice, that means two places. The referring writer should mention the arrangement when making the introduction, in the same email or call. Something like, "Full disclosure, Jamie pays me a referral fee if you decide to work together. That does not affect my recommendation, and your cost does not change because of it." The receiving writer should then include a line in the contract confirming the arrangement: that the referral fee exists, who pays whom, and that the client's cost is not increased because of it. The client should always know whether the recommendation they received was incentivized, and they should know it before they sign anything.
What do you do when you do not trust the organization to deliver?
This question has two very different answers depending on what kind of distrust you are dealing with.
If you have information suggesting fraud, you need to leave the engagement immediately so you are not exposed. Fraud is not a wait-and-see situation. It is also a tricky place to land, because the threshold for whistleblowing is high, and you need to be absolutely sure before you take that step. If you are not sure but you are concerned, the order of operations I recommend is:
Be direct with your supervisor or main contact. Name the specific concern, ask for the explanation, and give them the chance to clarify. Sometimes what looks like fraud is poor record-keeping or a misunderstanding.
If the explanation does not hold up, escalate to the board. Specifically, the board chair or the audit committee chair, if there is one. Embezzlement, misuse of restricted funds, or knowingly false reporting to funders are board-level concerns.
If the board response also does not hold up, leave the engagement. Document what you observed and what you reported, in case it comes up later.
If what you have is not evidence of fraud but a personal opinion that the organization will not deliver well, the path is different. You are not the funder, and your job is not to gatekeep funding.
In this case, I work with the organization to create a written work plan with concrete milestones and persons responsible. If they can commit to the work plan, I write the proposal and trust that the plan reflects real intent. If they cannot commit to the work plan, I point out that they may not be ready to apply this year, and offer that as a finding, not a judgment. If they insist on applying anyway, I have two choices. I can write a truthful proposal that does not stretch anything, including the outcomes section. Or I can politely recuse myself from that particular proposal while staying on for other work.
What I do not do is write a proposal that overstates the organization's capacity. That is where my personal opinion crosses into a truth-telling question, and we are about to get to that one.
How honest do your numbers have to be?
All the way honest. There is no version of grant writing ethics where it is acceptable to put numbers in a proposal that you know are not true.
This sounds obvious, but the working reality of the field involves a lot of pressure to nudge numbers in ways that feel small in the moment and add up to a real problem over time. Here are the most common temptations, and the line I hold on each:
Inflating the population served. If you reach 800 people, you write 800. Not 1,000. Not "approximately 1,000." If 800 sounds too small for the funder's expectations, the right move is to reconsider whether this funder is the right fit, not to round up.
Inflating outcome projections because the funder "expects" big numbers. If the realistic projection is a 15 percent improvement on the target measure, you write 15 percent. If 15 percent is not competitive, that is information for the organization to use in deciding whether to apply, not information to write past.
Writing letters of support that the partner did not actually write. Even if the partner agrees to sign whatever you draft, the letter should reflect their actual voice and commitments. A boilerplate letter the partner signed without reading is a fragile partnership. It’s okay to provide a template, just make sure they edit it to reflect their voice and commitments. For more on how to draft letters of support that the partner actually wants to sign, see my article on letters of support.
Using outdated data because the current data is less flattering. If the 2019 number tells a better story than the 2024 number, you still use the 2024 number, or you contextualize the change honestly.
Cherry-picking statistics that are technically true but misrepresent the picture. A single number out of context is often more misleading than no number at all.
Claiming evidence-based practices the organization is not actually implementing with fidelity. Saying you use a model is not the same as using the model. Reviewers can usually tell the difference, and program officers definitely can. Be specific about the ways the model is replicated and how it has been adapted.
Grant proposals are formal communications to funders, and the relationship of trust between the field and its funders is the single most important asset any of us have. Every inflated number, every recycled letter of support, every cherry-picked statistic erodes that trust a little. Multiply that across the thousands of proposals being submitted every week, and you can see why this matters.
Do you have to disclose AI use?
Yes. Disclose it to your client in your contract, with specifics about how you use AI and ensure confidentiality.
This is the newest live question in grant writing ethics, and the field has not settled it yet. Most of the commentary I see is hand-wavy. "Be transparent." "Use AI responsibly." That guidance is fine as far as it goes, but it does not tell anyone what to actually put in a contract.
Here is what I recommend, drawn from what I do in my own client agreements:
State that you use AI, and how. For example: "I use AI as a thought partner for brainstorming, structuring outlines, and editing drafts. I do not use AI to write proposals on the client's behalf, and I do not paste finished AI-generated text into proposals without substantial editing."
State what you do not use AI for. If you never use AI for parts of the work, say so. This is often more reassuring to clients than the list of permitted uses.
Address confidentiality explicitly. For example: "I do not include personally identifying information of clients, beneficiaries, board members, or staff in any prompt to an AI tool. I do not upload client documents containing confidential financial, personnel, or programmatic information to AI tools."
Name the tools you use. Clients increasingly want to know whether you are using ChatGPT, Claude, Gemini, or a specialized grant writing tool. Naming the tool builds trust and lets the client raise concerns if they have a corporate policy against a specific platform.
Provide an opt-out and honor it. If the client prefers you not use AI on their work, state in the contract how that election works. Some clients will prefer AI-free work; some will not care; some will explicitly want you to use AI for efficiency. Let them choose.
Address funder disclosure separately. Some funders now require disclosure of AI use in proposals. That is a different conversation from client disclosure. If a funder requires it, you and the client need to agree on what gets disclosed and how.
I do not think the field is going to land on a single answer about AI in the next year. Tools are evolving too fast, funder policies are evolving too fast, and individual writers are making different choices about what feels right. But the writers who put this in their contracts now, in specific language, are the ones who will not be caught flat-footed when a client or funder asks.
Working through these questions before you face them
Every question in this post is one I have had to work through in real time, with a real client, with real money and real reputational stakes on the line. The grant writers I see thriving in this field are the ones who have thought through their answers before the moment of decision arrives. The ones who struggle are the ones who improvise under pressure and then second-guess the choice for months afterward.
My Business of Freelance Grant Writing course is built around exactly these dilemmas, the kinds of questions that do not have universal right answers but do have better and worse ways of working through them. If you are a freelance grant writer or thinking about becoming one, I would love to help you think through your own answers before you need them.
FAQ
Is contingency-based grant writing illegal?
It is not illegal in most jurisdictions, but it violates the ethical standards of every major professional association in the field, and most major funders will disallow grant funds used to pay grant writers on contingency. Federal funds in particular cannot be used for contingency-based fundraising compensation.
What if my client cannot afford my standard rate?
You have several options that do not require crossing into contingency territory. You can offer a sliding scale based on organization budget size, offer a smaller scope of work that fits their budget, recommend they fundraise for grant writing support before they hire, or refer them to a newer writer whose rates are lower.
How do I tell a client I cannot work on commission without losing the contract?
Share your rates and fee structure in writing before the interview, and ask them to confirm receipt. If the client cannot get past your fee structure in writing, you have not lost a contract, you have saved an hour. If they confirm receipt and then propose commission anyway, you have a clear record that the terms were set up front.
Do I have to tell a client I used AI on their proposal?
Yes, and the cleanest way to handle it is in your contract, not in a one-off conversation. State in writing how you use AI, what you do not use it for, how you handle confidentiality, and what opt-outs the client has.
What do I do if I find out mid-engagement that my client misrepresented something to me?
First, find out whether the misrepresentation is in something already submitted or in something still being drafted. If it is in a draft, you fix it before submission. If it is in something already submitted, you have a harder conversation about whether the proposal needs to be amended or withdrawn. If the client refuses to correct it, leave the engagement.
Is it OK to use language from one client's proposal in another client's proposal?
No. Specific language about a specific organization's programs, outcomes, or partners belongs to that client and should not be reused.
Can I list specific funded grants in my portfolio?
Never list the client by name. List the funder and the range of the award amount. The client's identity, their program details, and the specifics of their funded work belong to them, not to your marketing.
Can I use past grants written for clients in my writing portfolio?
Yes, but only if they are properly anonymized. The client's name, identifying details about their programs, and specifics that could trace back to them all need to come out before the writing can be used as a sample. I have written more about how to anonymize past work for portfolio.
About the Author
Allison Welch, M.Ed., GPC is the founder of Spark the Fire Grant Writing and one of approximately 30 GPCI-approved trainers nationally. With 25+ years of experience in the field, she has trained more grant writers than she can count and serves as a reviewer for private foundations and government funders. She is the creator of the Certificate in Grant Writing and the author of the forthcoming book, The 'Of Course' Factor: A Guide to Meaningful Grant Writing (October 2026).