How does a scientist write a grant proposal that gets funded?

A fundable grant proposal is not primarily a description of interesting science. It is an argument that:

There is an important problem, we know enough to tackle it now, your specific project can make a meaningful advance, and you have a credible plan for producing interpretable results. The exact criteria depend on the agency. For example, current NIH research-project review emphasizes importance of the research, rigor and feasibility, and the expertise/resources needed to execute it, while NSF evaluates intellectual merit and broader impacts, along with the quality of the plan, team, and resources.Grants.nih+1

1. Start with the funding opportunity—not your experiment

Before writing, answer:

• What scientific question does the funder explicitly want addressed?
• Why does this question matter to that agency?
• What does the solicitation say reviewers will evaluate?
• What kinds of projects has the program funded recently?
• What is the expected budget and project duration?
• Who is likely to review this?

A brilliant proposal submitted to the wrong program can be a poor fit.

Think of the solicitation as the exam rubric. Your proposal should make it easy for the reviewer to check every box.

2. Reduce the whole proposal to one central argument

You should be able to explain the proposal in roughly four sentences:

Problem: What important problem exists?

Gap: What don't we currently know?

Hypothesis/idea: What do you think is happening, and why?

Test: What will you do that will distinguish your explanation from alternatives?

For example:

Chronic exposure to X is associated with Y, but the mechanism connecting X to Y is unknown. We hypothesize that X alters pathway Z, which subsequently produces Y. We will test this hypothesis using three complementary approaches that measure Z at molecular, cellular, and organismal levels. These studies will establish whether Z is a causal mediator and identify circumstances in which the mechanism operates. That's much stronger than:

"We will investigate the effects of X on Y using several experimental approaches." The latter tells me what you'll do. The former tells me why I should care.

3. Make the "why now?" unmistakable

Reviewers should quickly understand the chain:

Important problem → critical knowledge gap → opportunity → proposed solution → consequence of success

NIH specifically asks reviewers to consider whether proposed research addresses an important gap, solves a critical problem, or creates a valuable conceptual/technical advance.Grants.nih

A useful test is:

If this project succeeds, what will scientists know or be able to do that they cannot today? If the answer is vague, the proposal probably needs conceptual work before it needs better prose.

4. Design aims that work together

A common mistake is writing a collection of interesting experiments rather than a coherent scientific story.

A strong set of aims often looks like:

• Aim 1 — Establish: Determine whether the phenomenon exists and characterize it.
• Aim 2 — Explain: Determine the mechanism responsible.
• Aim 3 — Test/translate: Determine whether manipulating that mechanism changes the outcome.

The aims should be related but not completely dependent. If Aim 2 fails, Aim 3 shouldn't automatically become impossible.

NIH specifically identifies overly ambitious projects, unfocused aims, and unclear goals as common application problems.National Institute of Mental Health

5. Write for the reviewer, not for yourself

You know your field's literature. Your reviewer may be an expert in the broad area but not your exact niche.

So don't make the reviewer work to reconstruct your argument.

Use:

• informative headings;
• short paragraphs;
• explicit hypotheses;
• figures and conceptual diagrams;
• defined acronyms;
• clear transitions;
• strategically repeated key ideas.

NIH explicitly recommends outlines, subheaders, and strong organization to help reviewers navigate the Research Strategy, and advises minimizing jargon and using graphics when useful.Grants.nih+1

A good proposal should be easy to summarize after one reading.

6. Make feasibility almost boring

This is one of the biggest differences between an exciting idea and a fundable proposal.

For every major experiment, answer:

What will you measure? Why is that measurement appropriate? What is the sample/design? What controls are required? How will you analyze the data? What result would support the hypothesis? What result would contradict it? What will you do if the experiment fails?

The reviewer should rarely have to ask, "But what if...?"

NIH's current framework explicitly evaluates scientific rigor and whether the proposed studies can realistically be completed within the proposed timeframe.Grants.nih

7. Treat preliminary data as evidence, not decoration

Preliminary data should answer questions such as:

• Does the phenomenon actually occur?
• Can you perform the proposed assay?
• Is the effect large enough to detect?
• Does your model system behave as expected?
• Does your central hypothesis have plausible support?

Don't dump every interesting result into the proposal.

Instead:

Here is the observation → here is what it implies → here is the unanswered question → this proposal tests it. That creates momentum.

8. Anticipate reviewer objections

This is a surprisingly powerful exercise.

After drafting each aim, write down:

"If I were reviewing this, what would make me nervous?" Then address those concerns proactively.

For example:

Reviewer concern: "The proposed biomarker is nonspecific."

Weak response: Ignore it.

Strong response: Explain why it is appropriate, add an orthogonal measurement, and specify what you will conclude if the measurements disagree.

A proposal becomes more credible when it demonstrates that the scientist has already thought through failure modes.

9. Don't confuse innovation with novelty for novelty's sake

"Nobody has ever done this" isn't necessarily compelling.

Innovation can mean:

• a new biological mechanism;
• a new conceptual model;
• a new methodology;
• a new application of an existing technology;
• combining established approaches in a genuinely new way.

NIH's current framework explicitly notes that a project does not have to be highly novel if it addresses an important problem; innovation is considered in relation to the importance of the proposed research.Grants.nih

10. Make the proposal's significance concrete

Avoid:

"This work will have important implications for human health." Instead:

"If successful, these studies will establish X as a regulator of Y. This would provide a mechanistic explanation for Z and identify a previously unrecognized point of intervention." The second statement tells the reviewer what changes if you are right.

11. Match the proposal to the review criteria explicitly

For NIH, the traditional five criteria are Significance, Investigator(s), Innovation, Approach, and Environment, although NIH has reorganized review of most research project grants submitted since January 25, 2025 into a simplified framework centered on importance, rigor/feasibility, and expertise/resources.Grants.nih+1

For NSF, reviewers consider Intellectual Merit and Broader Impacts, including originality, the soundness of the plan, qualifications, resources, and how success will be assessed.NSF - U.S. National Science Foundation

So literally make a checklist:

12. Write the Specific Aims page last

This is counterintuitive, but extremely useful.

Draft the scientific argument, aims, approach, and expected outcomes first. Then write the aims page once you know exactly what the proposal is saying.

The aims page should function almost like a movie trailer:

1. Establish the important problem.
2. Identify the knowledge gap.
3. State the central hypothesis or objective.
4. Give the aims.
5. Explain the expected outcome.
6. Explain why that outcome matters.

A reviewer who reads only this page should understand the entire intellectual architecture of the proposal.

13. Get someone unfamiliar with the project to review it

Ask them:

"After reading this, what do you think my hypothesis is?" Then:

"What do you think my three aims are?" Then:

"Why does this project matter?" If their answers don't match yours, don't blame the reader. The proposal isn't communicating clearly enough.

NIH similarly advises writing so assigned reviewers can readily understand and explain what you propose to the rest of the review panel.Grants.nih

The simplest mental model

Think of a grant as a chain:

Important problem ↓ Specific gap in knowledge ↓ Strong hypothesis/question ↓ Focused aims ↓ Rigorous experiments ↓ Predefined interpretation ↓ Credible alternatives ↓ Meaningful scientific payoff

If one link is weak, the reviewer feels the weakness even if every individual experiment is technically excellent.

And perhaps the most important principle:

Don't try to convince reviewers that your project is guaranteed to succeed. Convince them that the question is important, the logic is compelling, the experiment is rigorous, and the information gained will be valuable regardless of which plausible result occurs. If you tell me your field, funding agency (NIH/NSF/DOE/etc.), and career stage, I can show you what this looks like using a realistic Specific Aims page + proposal outline for your type of science.

Then:

"What do you think my three aims are?" Then:

"Why does this project matter?" If their answers don't match yours, don't blame the reader. The proposal isn't communicating clearly enough.

NIH similarly advises writing so assigned reviewers can readily understand and explain what you propose to the rest of the review panel.

The simplest mental model

Think of a grant as a chain:

Writing a grant proposal that secures funding—whether from the National Science Foundation (NSF) or the National Institutes of Health (NIH)—is less about writing a standard research paper and more about crafting a persuasive, highly structured business case for your idea. Reviewers are exhausted, busy scientists who grade dozens of proposals at a time.pmc.ncbi.nlm.nih.gov+1

A funded proposal typically relies on a strategic, highly disciplined approach:

1. Master the Guidelines First : Before writing a single sentence, read every rule in the agency's solicitation guide. Violating page limits, font requirements, or missing required supplementary documentation (like data management or mentoring plans) can get your proposal rejected without review.nsf.gov+1
2. Hook Them in the Project Summary/Abstract : This is your first and sometimes only impression. Within one page, you must clearly state the gap in knowledge, the central hypothesis or objective, why this matters (Intellectual Merit ), and what societal or broader change it will enable (Broader Impacts).besteditproof.com+1
3. Lead with a Transformative, Feasible Idea : Reviewers look for high impact combined with low risk. Groundbreaking ideas need to be backed by strong preliminary data or a well-reasoned pilot study that proves your team has the capability to pull it off.pmc.ncbi.nlm.nih.gov
4. Design a Rigorous, Bulletproof Methodology : Detail how you will test your hypothesis. Anticipate alternative outcomes and explain what your plan B is. Reviewers need to see that you have thought through statistical power, variables, and potential pitfalls.besteditproof.com
5. Highlight the Team’s Expertise : Show—don't just tell—that your team is uniquely qualified to execute this specific project. Include complementary co-investigators or consultants if your lab lacks a specific technical skill.pmc.ncbi.nlm.nih.gov
6. Write for the Exhausted Reviewer : Use white space, clear headings, bold text for emphasis, and simple conceptual diagrams. If a reviewer has to hunt for your core hypothesis, you have already lost them. Avoid excessive insider jargon where plain scientific clarity works better.besteditproof.com
7. Justify Every Dollar : Construct a realistic, well-reasoned budget. Inflated budgets look careless; underfunded budgets look unfeasible. Align your budget justification meticulously with the actual labor and resources required.pmc.ncbi.nlm.nih.gov+1

Would you like to focus on a specific funding agency like the NIH or NSF , or do you need help structuring a specific section like the Specific Aims/Project Summary?

A funded grant proposal is usually not won by having the “most exciting” idea alone. It is won by making reviewers quickly understand why the problem matters, why your approach is credible, and why you (and your team) can deliver the promised results. Major funders such as NIH explicitly evaluate factors like significance, innovation, approach/feasibility, investigator expertise, and research environment.National Institute of Mental Health+1

A practical approach:

1. Start with the funding agency, not the experiment

Before writing, identify:

• The agency’s mission and priorities
• The specific funding call (not just a broad topic)
• The review criteria
• The audience and likely reviewers

A brilliant project that does not match the program’s goals often fails. Read funded abstracts and recent awards in the same program area to understand what the agency considers important.

2. Build a compelling “big idea”

A strong proposal answers four questions early:

1. What important problem exists?
2. What critical gap in knowledge prevents progress?
3. What will you discover or create that changes the field?
4. Why is now the right time to do this work?

Weak:

“We will study protein X using technique Y.” Stronger:

“A major barrier to treating disease Z is that we do not understand how protein X controls process Y. We will determine the mechanism and identify intervention points using…” The reviewer should understand the importance before getting buried in methods.

3. Make the aims clear and testable

Many successful proposals are built around 2–4 specific aims.

Each aim should have:

• A clear question
• A rationale
• A hypothesis (when appropriate)
• A measurable outcome
• A logical connection to the overall goal

Avoid:

• Too many aims
• A “shopping list” of experiments
• A plan where success depends on everything working perfectly

Reviewers often criticize proposals that are overly ambitious or have unclear goals.National Institute of Mental Health

4. Write for reviewers, not yourself

Reviewers may be experts but usually are not specialists in every technique you use. Make the logic easy to follow.

A good structure:

Problem → Gap → Idea → Evidence → Experiment → Expected outcome → Impact

Use:

• Short paragraphs
• Informative headings
• Figures or conceptual diagrams
• Plain explanations of technical concepts

Funding agencies often emphasize clarity and organization because reviewers must evaluate many applications efficiently.Grants.gov

5. Show feasibility without making the project boring

Reviewers want ambition and evidence you can succeed.

Include:

• Preliminary data (when available)
• Relevant publications
• Proof-of-concept results
• Expertise of collaborators
• Access to necessary facilities

A proposal should make reviewers think:

“This could change the field, and this team has a realistic path to getting there.”

6. Treat the methods section as a risk-management plan

A common mistake is writing only what you will do. Strong proposals explain:

• Why each method is appropriate
• What controls are needed
• What alternative strategies exist
• What you will do if an approach fails

For example:

Weak:

“We will measure gene expression.” Strong:

“We will measure gene expression using method A because it captures changes at the single-cell level. If signal quality is insufficient, we will use method B, which provides complementary validation.”

7. Make innovation meaningful

Innovation does not always mean inventing a new technology.

It can mean:

• A new biological question
• A new conceptual framework
• Applying an existing method in a new way
• Combining fields in a useful way

A technically complicated proposal is not automatically innovative.

8. Build the budget around the science

Every requested dollar should connect to the plan:

• Personnel → who does the work
• Equipment → why it is needed
• Supplies → which experiments require them
• Travel/collaboration → how it advances the project

An inflated or poorly justified budget can weaken confidence.

9. Get aggressive external review before submission

Strong applicants rarely write alone.

Ask colleagues to review:

• The first page
• The aims
• The overall logic
• The feasibility

Have both:

• Experts who know the science
• Scientists outside your niche who can judge clarity

NIH specifically recommends using pre-readers and revising based on critical feedback.Grants.gov

10. Write the first page as if it decides everything

Many reviewers form an early impression. The opening should contain:

• The importance of the problem
• The knowledge gap
• Your central idea
• Your approach
• The expected impact

A reviewer should be able to summarize your proposal after reading one page.

A simple grant “skeleton”

Title: Specific, memorable, not overloaded with jargon

Project summary/abstract:

• Problem
• Gap
• Approach
• Expected impact

Specific aims:

• Overall goal
• Central hypothesis
• Aim 1
• Aim 2
• Aim 3
• Expected outcomes and impact

Research strategy:

1. Significance — why it matters
2. Innovation — what is new
3. Approach — how you will succeed

Investigator/environment:

• Why this team
• Why this institution
• Why now

The mindset shift

Many scientists write grants as if they are explaining their experiment. Successful proposals are closer to a scientific argument:

“Here is a major unanswered question. Here is why solving it matters. Here is evidence that we have a credible way to solve it. Here is how the results will move the field forward.” That is the story reviewers fund.

That is the story reviewers fund.

A fundable grant proposal is less about writing beautifully and more about making a reviewer think:

“This is an important problem, this team has a credible way to solve it, and the results will matter.” That is essentially the logic behind major research funders’ review systems. For example, NIH currently emphasizes importance, rigor/feasibility, and investigator expertise/resources in its simplified framework.Grants.gov+1

The basic architecture

1. Start with the problem, not the experiment

Weak opening:

“We will investigate protein X using CRISPR and single-cell RNA sequencing.” Strong opening:

“Drug resistance remains a major barrier to treating disease X. A critical unanswered question is why a subset of cells survives initial therapy.” The reviewer needs to understand why they should care before you tell them what you're going to do.

A useful chain is:

Important problem → critical knowledge gap → why the gap exists → your central hypothesis → experiments → expected payoff

NIH explicitly asks whether the research addresses an important problem or critical gap and whether the proposed work could advance the field.Grants.gov

2. Make the central hypothesis unmistakable

A reviewer should be able to answer these questions after reading your first page:

• What do you think is happening?
• Why do you think that?
• What will you test?
• What will we learn if you're right?
• What will we learn if you're wrong?

For example:

Central hypothesis: We hypothesize that mitochondrial stress activates pathway X, which allows tumor cells to survive therapy. Then build the proposal around testing that hypothesis—not around performing a collection of interesting experiments.

3. Limit yourself to a few powerful aims

A common mistake is trying to demonstrate everything you know how to do.

A strong proposal might have:

• Aim 1: Determine whether X causes Y.
• Aim 2: Determine the mechanism connecting X to Y.
• Aim 3: Test whether manipulating X can change the outcome.

Each aim should answer a meaningful question and advance the overall story.

NIH specifically identifies overly ambitious projects, unfocused aims, and unclear goals as common application problems.National Institute of Mental Health

4. Make the significance almost impossible to miss

Don't assume reviewers will infer why your work matters.

Explicitly say:

The critical problem is… The gap in knowledge is… This project will address that gap by… Successful completion will enable… A reviewer should never finish your proposal wondering, “So what?”

And significance doesn't necessarily require flashy novelty. NIH notes that a project can be highly impactful even when it isn't particularly innovative, if it addresses an important problem.Grants.gov

5. Make innovation concrete

Avoid:

“This is a highly innovative study.” Instead explain what is actually different.

For example:

“Existing studies examine pathway X in bulk tissue. We will instead resolve pathway X at single-cell resolution, allowing us to distinguish resistant from sensitive populations.” That's innovation the reviewer can evaluate.

6. Approach: convince them you can actually do it

This is where many exciting ideas die.

For every major experiment, explain:

Question → experimental design → controls → analysis → expected result → interpretation → alternative

For example:

If inhibition of X reduces resistance, we will test whether the effect is mediated through pathway Y using genetic rescue experiments. If inhibition does not alter resistance, we will distinguish between target failure and pathway redundancy using... This tells the reviewer that you've thought beyond the first experiment.

NIH's guidance emphasizes rigor, reproducibility, appropriate design and analysis, feasibility, and consideration of potential problems and alternatives.Grants.gov+1

7. Write for a smart reviewer who isn't you

Your reviewer may be an expert in the broad field but not your exact niche.

So:

• Define specialized terminology.
• Minimize unexplained acronyms.
• Use informative headings.
• Keep paragraphs short.
• Make figures do real explanatory work.
• Put the main conclusion near the beginning of paragraphs.

NIH explicitly recommends clear organization, short paragraphs, whitespace, headings, and minimizing jargon.Grants.gov

8. Give reviewers a story they can remember

Imagine the reviewer explaining your proposal to the rest of the panel:

“This group has preliminary evidence that X drives Y. They have three aims: first establish causality, then determine the mechanism, then test whether the mechanism can be exploited. They have the expertise and preliminary data to do it.” That's what you're aiming for.

If the reviewer instead has to say:

“They do some RNA-seq, then some mouse experiments, then maybe look at pathway Z…” the proposal isn't sufficiently coherent.

9. Write the Specific Aims page first

This is one of the best practical pieces of advice from NIH: draft the aims before diving into the full application.Grants.gov

A useful one-page structure is:

Paragraph 1 — The problem What important problem does the field face?

Paragraph 2 — The gap What don't we know, and why does that matter?

Paragraph 3 — Your solution What is your hypothesis and overall objective?

Aims 1–3 What precisely will you do?

Final paragraph — Payoff What will change if you succeed?

If that page isn't compelling, writing 15 more pages usually won't fix the underlying problem.

10. Treat the reviewer as your customer

Before submitting, have someone read the proposal as a reviewer, not as a friend.

Ask them:

• What's the central hypothesis?
• What's the biggest problem this solves?
• What's innovative?
• What are the three aims?
• What is the biggest weakness?
• Where did you get confused?
• What would make you hesitate to fund it?

NIH recommends getting both domain experts and less-specialized colleagues to review drafts and explicitly asking them to evaluate the application against the review criteria.Grants.gov

The “fundable proposal” test

Before submitting, you should be able to fill in this sentence:

“This proposal should be funded because ______.” A strong answer looks something like:

“This proposal addresses a major barrier to treating X, tests a compelling mechanistic hypothesis supported by preliminary data, uses rigorous and complementary approaches, and will establish whether Y is a viable therapeutic strategy.” If your answer instead sounds like:

“Because nobody has studied this exact combination of techniques before…” you probably need to rethink the proposal.

Finally, write to the actual funding opportunity, not to some generic idea of a grant. The funder's current call/NOFO determines the applicable review criteria, requirements, and page limits. NIH specifically advises applicants to follow the funding opportunity's instructions and review criteria.Grants.gov

If you tell me your field, funding agency/program, and research idea, I can show you how I'd turn it into a compelling Specific Aims page + grant outline, including what a skeptical reviewer would likely criticize.

Aims 1–3 What precisely will you do?

Final paragraph — Payoff What will change if you succeed?

If that page isn't compelling, writing 15 more pages usually won't fix the underlying problem.

10. Treat the reviewer as your customer

Before submitting, have someone read the proposal as a reviewer, not as a friend.

Ask them:

• What's the central hypothesis?
• What's the biggest problem this solves?
• What's innovative?
• What are the three aims?
• What is the biggest weakness?
• Where did you get confused?
• What would make you hesitate to fund it?

NIH recommends getting both domain experts and less-specialized colleagues to review drafts and explicitly asking them to evaluate the application against the review criteria.

The “fundable proposal” test

NIH's guidance emphasizes rigor, reproducibility, appropriate design and analysis, feasibility, and consideration of potential problems and alternatives.

A grant proposal that gets funded is usually not the proposal with the most impressive science. It is the proposal that makes reviewers think:

“This is an important problem, this is a compelling solution, these investigators can actually do it, and I understand exactly why this project deserves funding.” The details vary by funder, but NIH and NSF both make the underlying logic explicit: reviewers assess importance, originality, feasibility, the team’s ability to execute, and the likely impact of the work.Grants.gov+1

1. Start with the funder, not your idea

Before writing, find the specific funding opportunity and work backward from its review criteria.

Ask:

• What problem does this program want solved?
• What kinds of projects does it explicitly prioritize?
• Who will review my proposal?
• What criteria determine the score?
• What would make a reviewer say “fund this”?
• What would make them say “interesting, but not appropriate for this program”?

For NIH applications, the funding opportunity notice and application instructions determine what reviewers are looking for; NIH explicitly recommends making the proposal's fit with the review criteria obvious.Grants.gov+1

A common mistake: writing a great proposal for the wrong program.

2. Build the argument around one important problem

A strong proposal has a simple intellectual spine:

Problem → Gap → Hypothesis/Idea → Experiments → Expected result → Impact

For example:

Problem: We don't know why X occurs. Gap: Existing studies explain A and B, but cannot explain C. Idea: We hypothesize that mechanism Y connects A to C. Plan: We will test Y using three complementary approaches. Outcome: These experiments will distinguish between competing models. Impact: The results will establish a new framework for understanding X. If a reviewer cannot summarize your proposal this way after reading it, the proposal probably isn't sufficiently focused.

NIH's current review framework emphasizes essentially this question: How important is the research, how rigorous and feasible is the approach, and does the team have the necessary expertise and resources?Grants.gov

3. Make the significance almost impossible to miss

Don't merely say:

“Disease X is an important public-health problem.” Explain what changes if you succeed.

A useful structure is:

Current state → Critical limitation → Your advance → Consequence

For example:

Current approaches can detect X but cannot predict Y. This limitation prevents us from understanding Z. We will determine whether mechanism A controls Y. If successful, the work will provide a mechanistic framework for predicting Y and identify a previously unrecognized point of intervention. The reviewer should understand the payoff before they get buried in experimental details.

4. Make the aims feel inevitable

For many research grants, the aims are the heart of the proposal.

Good aims are:

• Specific
• Independent enough that one failure doesn't destroy the entire project
• Logically connected
• Testable
• Achievable within the grant period

Bad aims often amount to:

“We will investigate X, Y, and Z.” Better:

Aim 1: Determine whether X regulates Y. Aim 2: Define the mechanism by which X regulates Y. Aim 3: Test whether disrupting this mechanism changes Z. The aims should collectively answer one big question, rather than being three unrelated experiments.

NIH specifically identifies overly ambitious projects and unfocused aims as common application problems.National Institute of Mental Health

5. Don't confuse innovation with novelty

“It's never been done before” is not enough.

The better argument is:

Existing approach A cannot answer question B. Our approach C can answer it because... Innovation should serve significance.

In fact, NIH explicitly notes that a project does not have to be highly innovative to have major scientific impact if it addresses an important problem.Grants.gov

6. Spend enormous effort on feasibility

This is where many exciting proposals lose.

For every major experiment, the reviewer should be able to answer:

“Why should I believe this will work?”

Give them evidence:

• preliminary data
• published evidence
• validated methods
• appropriate controls
• statistical justification
• expertise on the team
• access to necessary equipment/data/populations
• alternative strategies

And address failure modes proactively:

Potential problem: Method A may not distinguish between X and Y. Alternative: We will use method B, which provides an independent measurement. That doesn't make the proposal look weak. It makes you look like a scientist who has actually thought through the experiment.

NIH's guidance emphasizes rigor, reproducibility, rationale, study design, and feasibility; reviewers are explicitly instructed to consider whether the proposed work can be done well within the proposed timeframe.Grants.gov+1

7. Write for the reviewer, not yourself

Your reviewer may be an expert scientist—but probably not an expert in every component of your project.

NIH recommends clear structure, short paragraphs, limited jargon, informative headings, and writing that lets reviewers quickly identify the important points.Grants.gov

Think:

“Could a smart scientist outside my narrow specialty explain my proposal after reading it?” Use figures strategically. A single conceptual diagram showing:

knowledge gap → hypothesis → aims → expected outcomes

can save a reviewer several minutes of cognitive effort.

8. Treat the review criteria as a checklist

After writing each section, literally ask:

For NSF, for example, reviewers explicitly consider intellectual merit and broader impacts, along with originality, organization, assessment, qualifications, and available resources.NSF - U.S. National Science Foundation

9. Make the budget reinforce the scientific argument

The budget should make sense when someone compares it with the work.

Don't ask:

“How much money can I get?” Ask:

“What resources are necessary to accomplish these aims?” Every major expense should have an obvious connection to the science.

A reviewer should never think:

“Why are they spending $200K on this?” without immediately finding the answer in the proposal.

10. Get people to attack it before submission

One of the highest-value things a scientist can do is give the near-final proposal to several people:

• a subject-matter expert
• someone adjacent to the field
• someone who knows grants but not your science
• ideally, someone willing to be brutally critical

Ask them not “Do you like it?”

Ask:

“What do you think my central hypothesis is?” “What's the biggest reason you wouldn't fund this?” “Which aim seems weakest?” “Where did you become confused?” “What would you tell the review panel about this proposal?” Their answers are extraordinarily useful.

The secret: write the proposal as a decision document

A scientist naturally wants to tell the reviewer everything they know.

A successful grant does something different.

It controls the reviewer's reasoning.

By the end, you want the reviewer to have a chain of thoughts like:

This problem matters. ↓ There is a genuine gap. ↓ This hypothesis addresses the gap. ↓ These experiments directly test it. ↓ The investigators have demonstrated they can do them. ↓ The risks are understood and manageable. ↓ If successful, the result will substantially advance the field. ↓ Fund it.

That's the fundamental skill of grant writing.

And importantly, don't hide the science under grant-writing rhetoric. The proposal should make the scientific case clearer, not make ordinary science sound artificially grand.

For practical templates and current requirements, NIH provides official guidance on application sections and sample applications, while NSF publishes its current merit-review criteria and proposal-preparation requirements.Grants.gov+1

If you tell me your field and the grant (e.g., NIH R01, NSF, ERC, Wellcome, foundation, etc.), I can also show you what a fundable 1-page Specific Aims/Project Summary looks like and walk through it sentence by sentence.

“What resources are necessary to accomplish these aims?” Every major expense should have an obvious connection to the science.

A reviewer should never think:

“Why are they spending $200K on this?” without immediately finding the answer in the proposal.

10. Get people to attack it before submission

One of the highest-value things a scientist can do is give the near-final proposal to several people:

• a subject-matter expert
• someone adjacent to the field
• someone who knows grants but not your science
• ideally, someone willing to be brutally critical

Ask them not “Do you like it?”

Ask:

“What do you think my central hypothesis is?” “What's the biggest reason you wouldn't fund this?” “Which aim seems weakest?” “Where did you become confused?” “What would you tell the review panel about this proposal?” Their answers are extraordinarily useful.

The secret: write the proposal as a decision document

A scientist naturally wants to tell the reviewer everything they know.

A successful grant does something different.

It controls the reviewer's reasoning.

By the end, you want the reviewer to have a chain of thoughts like:

This problem matters. ↓ There is a genuine gap. ↓ This hypothesis addresses the gap. ↓ These experiments directly test it. ↓ The investigators have demonstrated they can do them. ↓ The risks are understood and manageable. ↓ If successful, the result will substantially advance the field. ↓ Fund it.

That's the fundamental skill of grant writing.

And importantly, don't hide the science under grant-writing rhetoric. The proposal should make the scientific case clearer, not make ordinary science sound artificially grand.

For practical templates and current requirements, NIH provides official guidance on application sections and sample applications, while NSF publishes its current merit-review criteria and proposal-preparation requirements.

A scientist writes a grant proposal that gets funded by doing more than describing good science: they build a convincing argument that the proposed work is important, feasible, timely, and likely to change the field. Reviewers are usually asking: Why this problem? Why now? Why this team? Why this approach? Why should we invest limited funds here?Grants.gov+1

A strong proposal typically follows this process:

1. Start with a fundable question, not a favorite technique

Weak:

“We will use single-cell RNA sequencing to study immune cells.” Strong:

“We will determine how a specific immune-cell state drives treatment resistance and identify targets for reversing it.” The fundable version has:

• A clear knowledge gap
• A specific hypothesis
• A meaningful outcome
• A reason the answer matters

Funding agencies generally favor projects with focused, testable goals rather than broad explorations.NCCIH+1

2. Match the idea to the right funding opportunity

Before writing:

• Identify agencies, programs, and review panels that fund your area.
• Read previously funded projects.
• Talk with program officers when possible.
• Make sure your proposal fits the funder's priorities.

A brilliant proposal sent to the wrong program often fails because reviewers see it as misaligned.NCCIH

3. Write the “Specific Aims” page first

For many biomedical grants, the Specific Aims page is the most important page because many reviewers see it early. It should answer:

1. What is the problem?
2. What do we know?
3. What do we not know?
4. What is your central hypothesis?
5. What are the 2–4 major aims?
6. What will change if you succeed?

A common structure:

Opening paragraph

• Big problem
• Current limitation
• Why solving it matters

Hypothesis

• “We hypothesize that…”

Aim 1

• Discover or define something

Aim 2

• Test a mechanism or intervention

Aim 3

• Validate or translate findings

Closing

• Expected impact

NIH guidance specifically emphasizes drafting aims early, making them hypothesis-driven, and ensuring they collectively answer the overall scientific question.Grants.gov

4. Tell a scientific story

Reviewers should feel a logical progression:

Problem → Gap → Idea → Evidence → Experiment → Impact

For example:

Alzheimer’s disease remains difficult to treat because current approaches target symptoms rather than early molecular drivers. Our preliminary data identify pathway X as a regulator of neuronal vulnerability. We will test whether targeting pathway X prevents disease progression. Every paragraph should advance that story.

5. Make significance obvious

Scientists often underestimate this. A reviewer should not have to infer why the work matters.

Explain:

• How many people or systems are affected?
• What limitation exists today?
• Why current approaches are insufficient?
• What becomes possible if you succeed?

Avoid:

“This pathway has not been characterized.” Prefer:

“Because this pathway controls a process required for drug resistance, defining its role could reveal new therapeutic strategies.” Clear communication matters because reviewers may not be specialists in every technical detail.Grants.gov+1

6. Make innovation believable

“Innovative” does not mean “high risk with no evidence.”

Good innovation:

• A new concept
• A new method applied to an important question
• A new way to solve a recognized limitation

Bad innovation:

• “Nobody has ever done this exact experiment.”

7. Design the approach so reviewers trust you

A convincing methods section includes:

• Experimental design
• Controls
• Sample sizes/statistics
• Alternative strategies
• Potential problems
• Backup plans
• Timeline

Reviewers worry about proposals that depend on everything working perfectly.

A strong proposal says:

“If experiment A fails, we will use approach B because both test the same hypothesis.” NIH guidance emphasizes rigor, reproducibility, clear organization, and addressing potential problems.Grants.gov+1

8. Show that the team can actually do it

Include:

• Relevant publications
• Preliminary data
• Expertise of collaborators
• Access to equipment, samples, or facilities

A great idea from a team without the necessary capability looks risky.

9. Write for reviewers, not for yourself

Common mistakes:

• Too much background
• Too much jargon
• Too many aims
• Hidden main point
• Overly ambitious timeline

Successful proposals are easy to review. Use:

• Clear headings
• Short paragraphs
• Figures and diagrams
• Repeated key messages

NIH explicitly recommends readable structure, clear language, and organizing proposals so reviewers can efficiently evaluate them.Grants.gov

10. Get many rounds of criticism before submitting

A good workflow:

6–12 months before

• Define idea
• Find collaborators
• Generate preliminary data

3–6 months before

• Draft Specific Aims
• Get mentor feedback
• Refine hypothesis

1–3 months before

• Write full proposal
• Circulate to experts and nonexperts
• Revise repeatedly

Successful applicants often treat feedback as part of the science, not as editing.NCCIH+1

A simple reviewer test

After reading the first page, a reviewer should be able to say:

• “This is an important problem.”
• “The hypothesis is clear.”
• “The experiments will answer the question.”
• “This team can do it.”
• “The results would matter.”

If any of those answers is “no,” the proposal needs revision.

The core skill is not just writing about research—it is making a reviewer believe that funding this project is the smartest use of their limited research dollars.