How to Write a Research Proposal
That Actually Gets Funded

Most research proposals fail not because the research idea is bad, but because the proposal cannot communicate it. A committee reading 200 proposals in two days has limited time per application. If your proposal cannot explain what you are doing, why it matters, and how you will do it, in that order, it will not survive shortlisting.

I have written proposals for the NST Fellowship, for BCSIR, and for internal university funding. I have also helped colleagues revise proposals that were rejected and then funded on second attempt. The difference between a funded proposal and a rejected one is almost never the quality of the underlying research idea. It is almost always how the idea is written.

This post walks through every section of a science research proposal, with examples drawn from environmental science, agricultural research, and related fields at Bangladeshi universities.

The Purpose of a Research Proposal

A proposal has one job: to convince a review committee that you know what problem you are solving, why it has not been solved before, and that your plan for solving it is realistic within the time and resources requested.

It is not a thesis chapter. It is not a literature review. It is not a description of your field of interest. It is an argument: structured, specific, and directly responsive to what the committee needs to see before approving funding.

—Before You Write: Three Questions First

Everything else in the proposal expands from these three. If you cannot answer them concisely before you start writing, you are not ready to write yet.

1Title

The title should be specific enough that a reviewer in your field knows exactly what the study is about. It should not be a slogan, and it should not run to three lines.

The strong version names the parameters, the spatial and temporal scope, the location, and what is being assessed. That is the entire study in twenty-two words.

2Background and Rationale

This section has a funnel structure: broad context narrowing to the specific gap, ending at the research objective. Two to four paragraphs is typical for most fellowship proposal formats.

Start with the broad problem and make it concrete quickly. For a water quality study, this means moving from "water contamination is a global issue" to "the Buriganga receives over 1.5 million cubic metres of industrial effluent daily" within the first paragraph. Use numbers. Cite them.

Narrow to what is already known about your specific area or topic, and then to what is not known. This is the gap paragraph, and it requires you to have actually read the literature — not skimmed abstracts. A gap that says "despite extensive research, this topic requires further study" is not a gap. It is a placeholder. A gap that says "existing studies in this river system have characterised total dissolved solids and faecal coliforms; heavy metal speciation and source contribution at the spatial scale of individual industrial discharge points has not been done" is a gap.

End the section with a clear objective statement, not a continuation of the literature review.

3Research Objectives

Structure them in two levels. First, a single broader goal statement in one sentence. Then, two to four specific numbered objectives that collectively achieve it.

Broader goal: "To assess the distribution, sources, and ecological risks of heavy metal contamination in the Buriganga river system adjacent to Dhaka's industrial zone and provide a science-based contamination baseline for policy use."

Specific objectives:

1. To determine seasonal concentrations of Cr, Pb, Cd, and As in surface water and sediment across twelve sampling points.
2. To identify dominant contamination sources through enrichment factor and geo-accumulation index analysis.
3. To evaluate ecological risk using established risk indices.
4. To produce a spatial contamination map using GIS interpolation.

Each objective is verifiable. At the end of the study, you will either have produced a contamination map or you will not. You will either have seasonal concentration data or you will not. Objectives written this way hold the researcher accountable and give the committee something concrete to fund.

What to avoid: writing objectives as questions ("What are the heavy metal concentrations in the Buriganga?"). Objectives should be active, declarative statements of what will be determined or produced.

4Methodology

This is where committees make their shortlisting decisions. A vague methodology signals that the researcher has not fully thought through how the study will actually run. Specificity here is the difference between shortlisting and rejection.

Organise the methodology as numbered Research Tasks, each corresponding to one or more objectives. This format is clear, auditable, and directly mirrors how the work will actually run. For each task, address: what materials or subjects are involved, what procedures are followed, what measurements are taken, and how the data will be analysed.

The specific version names the equipment, volume, preservation method, storage conditions, measurement instrument, and laboratory. A reviewer can evaluate whether this is realistic and whether the protocol is appropriate for the analytes being measured.

Do not write "appropriate statistical methods will be used." Name them: "Data will be analysed by one-way ANOVA with Tukey HSD post-hoc comparison using R version 4.3.1; significance set at p < 0.05. Spatial distribution will be mapped by kriging interpolation in QGIS 3.34."

For computational work: name the software and version. For laboratory work: name the instrument and model. For field work: name the design and replication structure. Reviewers who work in your field can immediately tell whether a methodology is thought through or borrowed from a previous proposal. Specificity is the signal that separates the two.

5Timeline

A Gantt chart — a table with research tasks on the left and months across the top — is the standard format. Mark which tasks run during which months. Do not separate phases that naturally overlap: data collection and preliminary analysis often run concurrently, and a timeline that shows them as strictly sequential is not believable.

Dark cells = primary activity. Light cells = overlap period. A timeline with zero overlap between phases is rarely believable.

For a twelve-month MSc proposal, this structure is typical and defensible. The key is that overlapping phases (months 7-9 in the example above) are shown honestly. A timeline where every task runs strictly back-to-back usually signals that the researcher has not thought about how the work will actually flow.

6Expected Outputs

This section answers: what will exist at the end of your research that did not exist before? It is the section most often written badly, because it is the easiest to write vaguely.

Real outputs are specific and countable: a peer-reviewed manuscript targeting a named journal (Environmental Pollution, Chemosphere, Science of the Total Environment); a validated seasonal dataset of heavy metal concentrations and ecological risk indices for the Buriganga system; a contamination map usable by city planners and the Department of Environment; a methods report documenting the sampling protocol for replication in other river systems.

Named, specific, concrete. If you cannot name the outputs, the committee cannot evaluate whether the research is worth funding.

7Budget

Not every proposal format requires an itemised budget. NST Fellowship proposals do not. DAAD, BCSIR, and USAID-funded applications typically do.

Even when a budget is not required in the submission, think through one anyway. It forces you to check whether your methodology is actually executable. A researcher who cannot estimate the cost of their own field sampling has likely not thought carefully enough about the logistics.

When a budget is required, every line needs a justification alongside the figure.

Do not inflate. Committees that fund environmental and agricultural research have seen chemical and equipment budgets before. Inflated budgets are a faster route to rejection than tight ones. They raise questions about the researcher's familiarity with actual costs in the field.

Group budget items logically: consumables and reagents; field equipment and transport; laboratory access and instrument time; dissemination (conference registration, open-access publication fee if applicable); contingency (typically 5-10%).

8References

Cite what you have read and can discuss in an interview. Use a reference manager (Mendeley or Zotero) and do not format references by hand. Follow whatever citation style the funding body specifies. A proposal that lists 60 papers but cannot defend three of them at interview is more damaging than a proposal with 20 well-chosen, well-understood sources.

9The Interview

Most national fellowship committees, including NST, shortlist candidates for a 20 to 30-minute interview after reviewing proposals. Know your proposal thoroughly. Know why you chose your methodology over alternatives. Know the three or four most important papers directly relevant to your specific topic and what they found. Know your preliminary data if you have it.

The committee is not trying to fail you. They are confirming that the researcher behind the document is real, engaged, and prepared to execute what they have written. A confident, specific, honest interview is more persuasive than a polished proposal from a researcher who cannot explain what they wrote.