grants_public

Skill: Academic Grant Writing from Notes

Purpose

Transform one or more Markdown inputs containing ideas, outlines, meeting notes, rough drafts, and reference fragments into a polished academic grant proposal written in LaTeX, with a matching BibTeX file.

Input Format

The input directory may contain any of the following Markdown files:

These files may be incomplete, repetitive, messy, or contradictory. Your job is to synthesize them into a coherent grant application.

Output Format

Produce:

The main output should be a clean, compile-ready LaTeX manuscript.

Core Task

Use the Markdown notes to draft a grant proposal with the following qualities:

Preferred Grant Structure

Unless the user specifies otherwise, use this structure:

  1. Title
  2. Lay Summary
  3. Vision (Background, Aims, and Impact)
  4. Research Questions / Aims and SMART Objectives
  5. Methodology / Work Packages
  6. Outcomes, Impact, and Translation Plan
  7. Workplan and Timeline (Gantt chart described in prose or tabular form)
  8. Applicant and Team Capability
  9. Ethics and Responsible Research Innovation (RRI)
  10. Risks and Mitigation
  11. Resources and Cost Justification
  12. References

If the notes suggest a different funding format (e.g. UKRI, Horizon Europe, ERC, Wellcome, Gates), adapt the structure accordingly.

Understanding the Review Process

When drafting, always write with the actual reviewer audience in mind. Proposals pass through multiple stages:

Write each section with its primary reader in mind.

Vision Section

The Vision section is the narrative heart of the proposal. It must build a coherent story arc:

Problem → Gap → Aim → Why solving it matters

Structure it across seven elements in this order:

  1. The big picture — why this research matters; present key facts and evidence (statistics, citations) that establish the problem at scale. Tailor to the funder’s priorities and the expected scale of impact (e.g. a £500K proposal reads differently from a £5M one).
  2. What others have done — a selective review establishing that serious prior work exists but that a gap remains.
  3. The gap in knowledge — the single most important missing piece. Be explicit: this is what you will address.
  4. Your aim and research questions — signposted clearly: “This project aims to…” followed by specific research questions.
  5. Novelty and timeliness — why now, why this approach, what makes it different from prior work.
  6. Expected outcomes and impacts — what changes in the short and long term; address academic, societal, economic, and/or environmental impact as appropriate.
  7. Who will benefit and how — close the loop by connecting beneficiaries back to the problem described at the start. Do not say “researchers” — be specific (e.g. “clinicians working in early cancer diagnostics”, “rural teachers without access to digital infrastructure”).

Ensure the Vision section does not merely describe — it must persuade. Back every claim with a citation or a statistic.

SMART Objectives

Each objective must be Specific, Measurable, Achievable, Realistic/Relevant, and Time-Bound.

Format each objective as:

O[N]. [Short title]: To [action verb] [specific target] in [context/location] [timeframe; lead institution or person]

Example:

O1. Soil microorganism survey: To characterise microbial abundance and population diversity in working farm fields in the UK treated with paper pulp waste [Y1Q1–2; UoX lead]

Include an overarching Aim (one high-level goal) followed by 3–5 SMART objectives. Each objective should map directly to a Work Package.

Work Package Methods and Outputs

Each Work Package (WP) corresponds to one objective. Each WP entry must include:

WP[N]: [Title] [Month range, e.g. M1–M12]
Lead: [Name (role, institution)]
Background: [One to two sentences on the specific problem this WP addresses]
Activities:
  - [Activity name]: Practical details and justification
  - [Activity name]: Practical details and justification
Risk and mitigation: [Specific risk and how it will be reduced]
Key deliverables:
  - D[N.1] [Description — e.g. open-access dataset submitted to repository X]
  - D[N.2] [Description — e.g. paper submitted to target journal Y]

Keep a clear 1-to-1 mapping: Objective N → WP N.

When writing methods:

Reviewers rarely complain about too much methodological detail; they frequently criticise proposals that are too vague or that show a disconnect between aims and approach.

Outputs, Outcomes, and Impact

Distinguish clearly between:

Outputs are more than just papers. They include methodologies, prototypes, trainings, exhibitions, business plans, policy frameworks, new collaborations, monographs, and websites.

Translation of Outputs into Outcomes and Impact

After the work packages, include a dedicated section titled “Translation of outputs into outcomes and impact” with three parts:

  1. Output clusters: group related outputs by theme or stakeholder and write 1–2 sentences per cluster describing what users will do differently as a result. Example: “The open-access benchmarking dataset and accompanying workshop will enable clinical NLP researchers to evaluate models against real-world minority language patient queries for the first time.”

  2. Outcome-to-impact mapping: trace each cluster to its broader impact. Be specific about who benefits. Example: “Clinicians in regions with large Sylheti-speaking populations will have access to validated triage tools, reducing diagnostic delays and improving patient safety.”

  3. Funder alignment: 2–3 sentences explicitly connecting the project’s impacts to the funder’s stated strategic priorities. Use phrases like “will inform”, “is expected to”, or “could lead to” to balance ambition with realism.

Theory of Change: Consider presenting a Theory of Change table:

Level Content
Inputs Funding, facilities, time, team expertise
Activities Experiments, modelling, community co-design
Outputs Open datasets, trained models, workshops, papers
Outcomes Adoption of tools by clinicians; policy uptake
Impact Reduced health inequity; improved patient outcomes

Lay Summary

The lay summary is read by committee reviewers who are non-specialists, not just by members of the public. It is often the first thing read and sets the tone for everything that follows. If the reviewer is not excited after reading it, the opportunity is lost.

Structure a 150–500 word lay summary as follows (budget roughly 10 sentences for a 200-word version):

1–3. The problem/need (facts, scale, urgency)

  1. The research aim 5–6. Simplified methods 7–9. Expected benefits and impact (close the loop to the problem)
  2. How this addresses the funder’s objectives

Rules for lay writing:

Most of the lay summary text should draw from the Vision section, not the detailed methodology.

Narrative CV Guidance

When drafting researcher statements, use the R4RI (Resume for Research and Innovation) four-module format:

Module 1: Contributions to the generation of new ideas, tools, methodologies, or knowledge (outputs) For each contribution state: (a) the personal contribution, and (b) the significance in the field context.

Module 2: Contributions to the development of others and effective working relationships Includes: supervision, line management, mentoring, teaching, collaborations, strategic leadership.

Module 3: Contributions to the wider research and innovation community Includes: organising workshops/conferences, peer reviewing, invited talks, EDI contributions, positions of responsibility.

Module 4: Contributions to broader audiences and societal benefit (outcomes/impact) For each example state: (a) how target beneficiaries were engaged, and (b) short- and long-term changes resulting.

The researcher statement section must be persuasive — not modest. Every claim must be supported by concrete evidence with specific outcomes (e.g. “17% improvement in early diagnosis”, “£5,000 secured”, “100 survey responses”). Favour fewer, richer examples over many superficial items.

For a group CV, organise by output themes aligned to the proposal rather than listing by person.

Workplan / Gantt Chart

Describe or present a Gantt chart that maps:

The WP names and task titles in the Gantt must match exactly those used in the written methodology.

Writing Rules

Citation Rules

LaTeX Rules

BibTeX Rules

Workflow

  1. Read all Markdown inputs.
  2. Identify: project goal, research problem, contributions, methods, required references, missing information.
  3. Reconstruct the review pipeline in mind: eligibility check → expert peer review → committee review. Draft each section for its primary audience.
  4. Draft the Vision section using the seven-element structure.
  5. Write SMART objectives and map each to a Work Package.
  6. Draft work packages with activities, risks, and deliverables.
  7. Write the outcomes/impact section including output clusters, outcome-to-impact mapping, and funder alignment.
  8. Write the lay summary last, drawing from the Vision.
  9. Convert all bibliographic mentions into BibTeX entries.
  10. Write the final LaTeX and BibTeX outputs.
  11. Check consistency: names, acronyms, citations, section ordering, terminology.
  12. Ensure the output compiles cleanly.

Style Guidance

Write in a tone that is professional, precise, credible, concise, and academically persuasive.

Avoid: hype, vague language, unsupported promises, repetitive phrasing, overlong sentences, endashes, passive constructions where active is possible.

Handling Incomplete Notes

When the notes are incomplete:

Acceptable placeholders: [FUNDING SCHEME], [PROJECT DURATION], [NAME OF HOST INSTITUTION], [REFERENCE NEEDED]

Author Details

Quality Checklist

Before finishing, verify:

This site is open source. Improve this page.