Last reviewed: 4 November 2026. Grant rules and deadlines change with each call, so confirm them on the funder's official page before applying.
A research proposal is defendable when its logic and evidence are clear enough that a reviewer does not have to fill gaps with assumptions. Clear reasoning reduces room for personal interpretation and makes the proposal easier to evaluate fairly. Every major claim should answer a question the reviewer would otherwise have to ask.
Most guides tell you which sections a research proposal should contain. This guide starts with a different question: what must the proposal prove to a reviewer?
Why this framework is useful It works across PhD proposals, fellowship applications and research grants because it focuses on what reviewers need to see: a clear problem, a believable plan and a realistic path to results. |
A real funding example from India is included later to show how the same principles appear in practice, without treating one funder's format as a universal rule.
The five steps at a glance
Step | Reviewer's question | What to show |
1. Prove the problem and the gap | Why does this problem need research, and what is still missing? | A concise evidence-based case that establishes the problem and ends in one clear gap |
2. Show why you can deliver it | Why should the reviewer believe that this project can be executed? | Your expertise, plus the people, facilities, collaborators or institutional support needed to carry out the work |
3. Build a clear research plan | Does your research plan clearly show how you will answer your research question? | Objectives, methods and work packages that clearly connect to one another |
4. Plan work that can be finished on time | Can you finish the project within the allowed time? | A realistic timeline with clear phases and milestones |
5. Define outcomes and justify resources | What will the project produce, and why are these resources needed? | Specific outputs, outcomes and resources that clearly connect to the research plan |
Step 1: Prove the problem and the gap
A funding panel rarely consists only of experts in your exact niche. Write for technically informed readers who may work outside your specialisation. First establish why the problem matters; then show what existing research still does not resolve. In ANRF's 2025 PM ECRG proposal-writing session, awardees emphasised making the core problem and project logic understandable to readers who may not be specialists in the exact niche.
How to do it
• Summarise the most relevant national and international work briefly and with citations.
• Move from the existing evidence to one clear gap that a non-specialist could repeat.
• State what your study will add, test, explain or improve compared with the current approach.
Common mistake a long literature summary that never establishes a research gap. The reviewer should be able to say in one sentence what is still unresolved and why it matters. |
Funded-project example: ARUN-SSW A publicly described ANRF PM ECRG project at Physical Research Laboratory states a concrete limitation: there is no solar radio imaging facility in the Asian time zone. The proposed response is equally specific: a solar radio heliograph covering 1 to 18 GHz. A clearly stated limitation leads directly to a clearly stated research response. Source: Physical Research Laboratory, ARUN-SSW (Advanced Radio Telescope Udaipur Network for Solar and Space Weather Research), project description in a 2025 job advertisement. |
Step 2: Show why you can deliver it
A good idea still needs a credible route to execution. Show what you can do, what other people or facilities are needed, and how those roles will be covered. Relevant evidence may include earlier research, publications, preliminary results, methods you have already used, specialist facilities, institutional support or collaborators with clearly defined roles.
How to do it
• Cite your own relevant work, including co-authored papers, where it directly supports the proposed direction.
• Show how earlier results, methods or technical experience support specific parts of the proposed plan. If relevant work has already been published, presented, patented, deposited as data/software, or reported in a thesis or technical report, cite it where appropriate.
• If the project needs expertise, facilities or data that you do not personally have, explain who or what will provide that capability and how access will be secured. Do not make the plan depend on unsupported assumptions.
Common mistake: a long publication list that never explains how the investigator's experience, team or research environment supports the proposed work.
Funded-project example: Fair Federated Learning IIT Gandhinagar's computer science department lists an ANRF PM ECRG project, "Fair Federated Learning Framework in the Presence of Heterogeneous, Strategic, and Malicious Clients," led by Manisha Padala, whose department profile lists fair and federated learning among her research areas. In the ANRF session, the awardee described extending earlier work on fairness in classification models into federated learning. The lesson is not that every grant must continue a PhD topic. It is that the capability link should be visible: what you already know, what new problem you are taking on, and why the next step is credible Source/credit: IIT Gandhinagar CSE research pages; ANRF PMECRG Proposal Writing Session, 22 November 2025. |
Step 3: Build a clear research plan
A reviewer should be able to see how your research question will be answered. Each objective should lead to a method or activity, and each activity should produce evidence or a result. Three terms are often mixed up:
• Objective: what the research aims to achieve.
• Research question: what the study is trying to find out.
• Hypothesis: what you expect the study may show, when a hypothesis is appropriate for the design.
How to do it
• Use a simple diagram if it helps the reader see the flow from problem → objective → method or work package → expected result. The diagram is a communication aid, not a requirement.
• Build the plan around what can realistically be achieved within the time allowed by the funder or institution. If you have a longer-term vision, mention it briefly, but keep this proposal focused on what this project will actually deliver.
• Keep objectives short and make sure each objective clearly connects to a method, activity or work package. Repeating the main aim for emphasis is fine; repeating the full methodology in several sections is not.
Common mistake: the objectives, methods and work packages are all present, but the reader cannot see how they connect.
Simple test For every objective, ask: What will I do? How will I do it? What evidence will it produce? If those three answers do not connect, the research plan needs work. |
Funded-project example: Operational Approaches to Information Leakage IIIT Hyderabad's profile of ANRF awardee Gowtham Raghunath Kurri describes a project with a clearly stated objective: build a theoretical framework for information-leakage measures and apply it to private information retrieval and privacy-preserving machine learning. In the ANRF session, the same proposal was explained as state of the art → gap → objectives → research plan. Source/credit: IIIT Hyderabad, 'ANRF Early Career Research Grant Winners', 2025; ANRF PMECRG Proposal Writing Session, 22 Nov 2025. |
Step 4: Plan work that can be finished on time
Your plan should fit the time you are actually given. A timeline helps the reviewer see what happens first, what depends on what, and whether the workload is realistic. If important activities would only happen after the approved period ends, the project is too large or the sequence needs redesign.
How to do it
• Break the work into phases or work packages with clear milestones. A simple table or Gantt chart is usually enough.
• Allow realistic time for recruitment, approvals, data collection, fabrication, experiments, computation, analysis and writing — whichever apply to your project.
• Remove, reduce or redesign activities that only become possible if the project runs beyond the allowed period.
Common mistake: an ambitious plan that looks possible only if major activities take longer than the stated project period.
Illustrative activity | Year 1 | Year 2 | Year 3 |
WP1: Refine design / approvals | ● | ||
WP2: Data collection / experiments | ● | ● | |
WP3: Analysis / validation | ● | ● | |
WP4: Outputs / dissemination | ● |
Illustrative Gantt chart only — adapt the activities and timing to your own project and funder.
Funded-project example: a three-year instrument project The ARUN-SSW PM ECRG project is publicly listed as a three-year project. In the awardee session, the investigator described dividing the work into three work packages across the project period. The lesson is simple: the work plan and the funded time must visibly fit each other. Source/credit: Physical Research Laboratory / Udaipur Solar Observatory, ARUN-SSW; ANRF PMECRG Proposal Writing Session, 22 Nov 2025. |
Step 5: Define the outcomes and justify the resources
A reviewer should understand both what the project will produce and why the requested resources are necessary to produce it. Separate direct outputs from the outcomes those outputs are expected to enable. This makes the research value and the resource logic much easier to evaluate.
How to do it
• Define the expected outputs: direct products of the project, such as a dataset, algorithm, instrument, prototype, protocol, publication or open-source tool.
• Define the expected outcomes: what those outputs will enable, improve, explain or make possible if the project succeeds.
• Tie each major resource request - equipment, personnel, consumables, travel or contingency - to a specific activity or work package.
• Support large or unusual requests with a clear justification or simple calculation, such as required compute hours, sample numbers, instrument time or fabrication needs.
Funded-project example: TVARAK-AI An ANRF PM ECRG project at IIIT Hyderabad proposes a scalable heterogeneous accelerator chip for next-generation AI models. The direct output is a hardware accelerator; the intended outcome is more energy-efficient and scalable AI inference. This is the distinction to make in your own proposal: what will you produce, and what will that product make possible? Source/credit: IIIT Hyderabad, 'Meet IIIT Hyderabad's Young Researcher With The Breakthrough Proposal to Build India's Own AI Chip', 2026. |
Common mistake: ambitious outputs with no credible route to achieving them, or budget items that the research plan never explains.
Worked Indian example: how ANRF PM ECRG reflects these five principles
ANRF PM ECRG is used here as one current Indian funding example, not as the template for every research proposal. The practices below were described by awardees in a public ANRF proposal-writing session in November 2025. They are individual experiences, not ANRF rules, and are paraphrased here.
Principle | What PM ECRG awardees described in the session |
1. Prove the problem and the gap | Used the state of the art to map relevant work, then used the origin/rationale section to explain why the unresolved problem mattered and how the proposed work differed. |
2. Show why you can deliver it | Cited relevant own and co-authored work and often positioned the proposal as a credible next step from prior research. This was awardee practice, not an ANRF requirement. |
3. Build a clear research plan | Used objectives, work packages and, in some cases, an input-to-output diagram to make the project easy to follow. Several awardees advised writing for technically informed reviewers who may not be specialists in the exact niche. |
4. Plan work that can be finished on time | Kept deliverables within the three-year PM ECRG project period and used milestones or timeline charts to show sequencing. |
5. Define outcomes and justify resources | Separated tangible outputs from what the project would enable, and justified manpower, equipment and other costs against the work. Individual budgeting heuristics mentioned in the session should not be treated as ANRF rules. |
A short case: an awarded instrument project built around one gap
Dr. Anshu Kumari's awarded ANRF PM ECRG project, 'Advanced Radio Telescope Udaipur Network for Solar and Space Weather Research (ARUN-SSW)', provides a concrete example. Public project material states that it addresses the lack of a solar radio imaging facility in the Asian time zone by developing a 1–18 GHz solar radio heliograph. Her public profile also records earlier work in solar/radio instrumentation, which supports the capability argument. In the ANRF proposal-writing session, the project was described as three work packages across the three-year grant period, with preliminary simulations and calculations used to show what the proposed instrument could deliver. This is an example of one funded project's logic, not a universal ANRF template.
Before you submit: test it on an outsider
Give the draft to a capable reader from another field and ask for a two-minute summary. If the problem, gap, plan and reason the work is credible do not come back clearly, the proposal is not yet defendable. Outside review is useful precisely because it reveals the connections that may feel obvious only to the person who wrote the proposal.
The defendability checklist
☐ The problem matters and the gap fits in one clear sentence
☐ Relevant prior work or expertise is linked to the proposed plan
☐ Objectives, methods and work packages clearly connect to one another
☐ The scope and timeline fit the approved or funded period
☐ Expected outputs and outcomes are specific and realistic
☐ Every major budget item maps to an activity or work package
☐ A non-specialist can summarise the proposal
☐ Word limits and call guidelines are met
☐ The main selling point is easy to spot
ANRF PM ECRG example: current call snapshot
• Dates: the current call runs from 2 November 2026 to 2 December 2026, 5:00 PM.
• Funding: up to Rs 60 lakh plus overheads for three years. Project tenure cannot be extended.
• Eligibility: early-career researchers who apply within two years of joining the institution, with an age limit of 42 years (three years' relaxation for SC, ST, OBC, women and physically challenged candidates) and a PhD in science or engineering, or an MD, MS, MDS or MVSc degree. Read the official call for the full conditions.
• Structure: the application format illustrates the framework above. State of the Art and Origin of the Proposal establish the problem and gap, the Research Plan shows execution, expected outputs and outcomes clarify what the project should produce, and the budget must support the proposed work.
Use this only as a worked example. Other funders and universities may ask for different sections, timelines, eligibility conditions and budget categories, and the official ANRF portal is the final source before applying.
Frequently asked questions
Does this framework apply only to ANRF proposals?
No. The five questions are intended as a general reviewer-facing framework for research proposals. ANRF PM ECRG is included only as a worked Indian funding example.
How long should a research proposal be?
It depends on the funder, university or programme. Follow the stated word or page limits rather than a general figure. A shorter proposal still has to make the problem, gap, plan, feasibility and expected contribution clear.
Do I need preliminary data?
Not always. Some calls require preliminary evidence and others do not. Even when it is not mandatory, relevant prior work or early evidence can strengthen feasibility if it genuinely supports the proposed research. Check the specific call or institutional guidance.
Can I use AI tools to write my proposal?
Check the funder's and institution's rules on AI use first. The ideas, claims and citations must be your own and verified.
Does ResearchEdit4u write proposals?
No. The team reviews drafts and helps strengthen the logic, without writing the proposal for the applicant.
Want a second reader for your proposal?
A proposal improves when someone outside the project tests its logic. ResearchEdit4u's research planning support reviews the problem-gap logic, capability case, research plan, timeline, expected outcomes and resource justification without writing the proposal for the applicant.