Dissertation Proposal Planning: From Topic Feasibility to Approval-Ready Structure

A strong dissertation proposal is not a decorative introduction to a project you have already decided to do. It is a test of whether the proposed research is sufficiently clear, justified and feasible to deserve the next stage of work. The most useful proposal therefore makes its logic visible: what you want to investigate, why the question matters, what existing work makes the question defensible, how you could answer it, and whether the plan can actually be completed under the rules and constraints that apply to you.

Reader outcome

After using this guide, you should be able to turn a broad dissertation idea into a proposal plan that can be checked against your programme requirements, discussed productively with a supervisor, and revised before you invest heavily in data collection or writing.

This guide deliberately does not give one universal proposal template. Universities, schools, modules and research degrees can specify different word limits, required sections, assessment criteria and approval processes. Your current programme documentation remains controlling.

Start with the rules that actually control your proposal

The first planning mistake is often starting with a template found online. A better first step is to identify the documents and people that define what your proposal must do.

Build a short requirement sheet before you draft prose.

This step is not administrative housekeeping. It changes the research plan. Oxford’s current graduate guidance, for example, tells applicants to check the relevant course page first because proposal requirements and assessment criteria differ by course. Edinburgh similarly advises students to consult programme information and subject-specific guidance because dissertation formats vary by discipline, question and project type.

Do not copy a word limit or heading sequence from this article, Aston, Oxford, Sheffield, Edinburgh or any other institution unless your own programme tells you to use it.

Test feasibility before polishing the title

A proposal can sound sophisticated and still be impossible to complete. Before you spend time improving the title, test the project against the constraints that would determine whether it can move forward.

A practical feasibility screen

Rate each area as green, amber or red.

A red rating does not automatically mean the topic is bad. It means the proposal needs redesign. You might narrow the population, change the evidence source, reduce the number of research questions, choose a more feasible design, or move from primary to defensible secondary data.

Edinburgh’s dissertation guidance explicitly asks students to consider feasibility, time and resources when choosing a project. Aston also frames the proposed methodology as something that must be logical and feasible, not merely impressive.

Build a research logic chain

A proposal becomes easier to evaluate when each major decision follows from the previous one. One useful planning chain is:

context or problem → current evidence → unresolved issue → aim → research question(s) → design → data or material → analysis → feasible contribution

The arrows do not represent a universal section order. They represent logical dependence.

Context or problem

Define the setting in which the research makes sense. Depending on the discipline, that might be an organisation, policy environment, population, archive, text, technology, clinical setting, market, theory or technical problem.

Avoid beginning with claims such as “this is a major global problem” unless you have evidence that establishes the scale and relevance of the problem.

Current evidence

Identify what credible literature or authoritative evidence already establishes. This is where you begin separating what is known from what you merely suspect.

Unresolved issue

A research gap is not the same as “I could not find many papers.” A defensible gap or unresolved issue has to emerge from a topic-specific review of current scholarship. Generic proposal guidance cannot prove that a particular field lacks research.

If the evidence is not yet strong enough to claim a gap, write the uncertainty more cautiously: for example, the relationship remains contested, evidence is concentrated in a different context, existing studies use designs that do not answer this question, or the available literature suggests a question that requires further examination. The exact wording must match the literature you actually find.

Aim and research questions

The aim should state what the project is trying to accomplish at a high level. Research questions break that aim into answerable inquiries.

A useful question is not just grammatically neat. It is narrow enough that you can identify what evidence would answer it, yet important enough to justify the work.

Design and analysis

The design should follow from the question. If you want to understand experience or meaning, your evidence and analytic approach will differ from a project estimating an association, testing an intervention, modelling a system, interpreting documents, comparing cases or developing an artefact.

Do not select a method because a keyword appears in the title. “Impact,” “effect,” “relationship,” “experience,” “influence” and similar words do not mechanically determine a valid design.

Make the literature review perform a job

In a proposal, the literature section should do more than prove that you have read papers. It should help the reader understand why the proposed question follows from the current state of knowledge.

A useful planning structure is:

1. What is already reasonably established? Identify the main concepts, findings or debates relevant to the question.
2. Where is the evidence limited, inconsistent or context-bound? Distinguish genuine uncertainty from a simple lack of reading.
3. What perspective or theoretical framing is relevant? Use theory only when it helps explain the question or design.
4. Why does the proposed study follow logically? Connect the literature directly to the aim and questions.

Aston’s current proposal guidance explicitly links a brief critical literature review to the rationale and theoretical underpinning of the proposed study. The important word is critical: a paragraph-by-paragraph inventory of authors is weaker than a synthesis showing patterns, disagreements and limitations.

For a dissertation proposal, you usually do not need to settle every debate before approval. You do need enough evidence to show that the question is informed by scholarship rather than invented in isolation.

Justify the methodology instead of naming it

A methodology section becomes convincing when the decisions are connected.

At proposal stage, try to answer these questions:

• What research design or overall approach is being proposed?
• What kind of evidence, data or material will answer the research question?
• Who or what will be included, and why?
• How will the evidence be collected, accessed or generated?
• How will it be analysed?
• What limitations are already foreseeable?
• What ethical, legal, safety, privacy or access issues may arise?
• What software, facilities or specialist resources are genuinely required?

Sheffield’s guidance is useful here because it distinguishes a detailed methodological justification from simply listing a method: research design, data collection and analysis all need to be considered. Aston similarly asks applicants to explain the proposed methodological approach, planned data, practical and ethical considerations, and analysis.

Method fit matters more than method complexity

A modest design that can answer the research question is stronger than an advanced design that cannot be executed or justified.

For example, a student with three months, no organisational access and no recruitment route should be cautious about proposing a large multi-site primary-data study. A narrower analysis using accessible data may produce a more defensible proposal even if it looks less ambitious.

Methodological sophistication should come from alignment and justification, not from the number of techniques named.

Treat access and ethics as design variables

Students often write “ethical approval will be obtained” as a final sentence and assume the issue is solved. At proposal stage, ethics and governance can change the design itself.

Ask early:

• Will you recruit human participants?
• Will you work with vulnerable groups or sensitive topics?
• Will personal, confidential or proprietary information be used?
• Is gatekeeper permission required?
• Does the project use organisational records, clinical data, school data or restricted archives?
• Are recordings, images, geolocation, biometric information or identifiable online data involved?
• Are there safety, export-control, data-protection or other governance requirements relevant to the field?

The correct approval route depends on the institution and project. This guide cannot tell you which formal review you need. The proposal should instead show that the issue has been identified and that the research plan will follow the current institutional process.

Access belongs in the same conversation. If your project requires interviews with senior executives, hospital records or a private dataset, “I will request access later” is not a feasibility plan. Identify the route, timing, dependencies and fallback options before the proposal is treated as workable.

Plan time around dependencies, not just months

A timeline should show that you understand the order in which research activities depend on one another.

A proposal timeline might include:

• requirement confirmation and proposal revision;
• targeted literature review;
• design refinement;
• ethics, governance or access steps where applicable;
• instrument, protocol or data-source preparation;
• pilot or feasibility work where appropriate;
• data collection, extraction or corpus construction;
• analysis;
• interpretation and discussion;
• drafting and redrafting;
• supervisor feedback cycles;
• final formatting and submission checks.

The exact stages depend on the project. What matters is that the schedule includes the work that can block later work.

For example, if data collection cannot begin before ethics approval and gatekeeper permission, the timeline should not place recruitment in the same week that approval is submitted. If a dataset must be purchased or requested, access lead time belongs in the plan.

A useful question is: What is the first date on which this stage could realistically begin, given everything that must happen before it?

Use an adaptable proposal skeleton

Once the research logic is sound, you can map it into the structure required by your programme. Where your programme does not prescribe headings, the following is a practical planning skeleton, not a universal template.

Working title

Make the focus identifiable without forcing every variable, method and location into one sentence.

Research context and rationale

Explain the problem or intellectual context, what current evidence establishes, and why the proposed inquiry is worth pursuing.

Aim, objectives or research questions

Use the terminology required by your programme. Keep the relationships among aim, objectives and questions explicit.

Literature and conceptual positioning

Synthesize the most relevant current work, show the unresolved issue carefully, and identify theory or concepts only where they genuinely help frame the study.

Proposed methodology

Explain the design, evidence, sampling or case logic where relevant, collection/access, analysis, foreseeable limitations and why these choices fit the questions.

Feasibility, ethics and resources

Show that access, time, skills, software, facilities, budget and governance constraints have been considered. Not every proposal needs a separate heading for each of these; the programme requirements control.

Research plan or timeline

Map the major stages and dependencies to the available period.

References

Use the required referencing system and make sure every substantive literature claim can be traced to the source actually supporting it.

A worked planning example

Suppose a student begins with this topic:

The impact of artificial intelligence on university students.

The title sounds current, but the proposal is not yet researchable. “Artificial intelligence” could mean generative AI, automated feedback, recommendation systems or institutional analytics. “Impact” could refer to grades, confidence, study behaviour, writing practices, employability or something else. “University students” could mean millions of people across different degrees and countries.

A feasibility-driven refinement might proceed like this:

1. Context: first-year international master’s students in one programme.
2. Phenomenon: use of generative AI during independent academic writing.
3. Research interest: how students describe changes in planning, drafting and checking their work.
4. Evidence route: interviews or another qualitative design, if recruitment and ethics are feasible.
5. Question: “How do first-year international master’s students in [programme/context] describe the ways generative-AI use influences their academic-writing process?”

This is still only a hypothetical example. The wording does not prove that interviews are the best design, that the population is accessible, or that the study is original. Those claims would require programme-specific and topic-specific evidence. The example simply shows how feasibility, scope and answerability can narrow a broad idea.

Prepare for a useful supervisor conversation

Instead of asking only “Is this topic okay?”, send a compact planning note that lets the supervisor see the decisions you have already made.

A useful discussion document can include:

• one-sentence research focus;
• provisional aim and one to three research questions;
• two or three evidence-backed reasons the study may be worthwhile;
• proposed data or material;
• proposed methodological approach and why it may fit;
• known access, ethics or resource constraints;
• the two or three decisions on which you need feedback.

Questions that invite actionable feedback include:

• Is the proposed scope realistic for the available dissertation period?
• Does the research question match the level of the programme?
• Is the proposed evidence sufficient to answer the question?
• What methodological risk should I resolve before submitting the proposal?
• Which programme requirement or disciplinary convention am I currently underestimating?

Oxford and Sheffield both encourage applicants to seek feedback where appropriate, while still making clear that the proposal represents the applicant’s own research work. Feedback should therefore sharpen your decisions, not outsource them.

Run an approval-readiness audit

Before you call the proposal “finished,” audit the reasoning rather than just the grammar.

Common proposal failures to catch early

Copying a universal template

A template can help you remember possible planning areas, but it cannot override the current requirements of your programme.

Making the topic broad to sound important

Breadth usually increases the number of concepts, evidence sources and methodological decisions you must justify. Narrowing often makes a project more rigorous, not less ambitious.

Manufacturing a research gap

Do not write “no studies have examined…” unless a defensible topic-specific search supports that statement. Absence claims are difficult to establish and easy to overstate.

Naming methods without showing fit

“Mixed methods,” “SEM,” “thematic analysis,” “machine learning,” “case study” or “interviews” are not justifications. Explain why the evidence and analysis can answer the research question.

Assuming access

If the research depends on participants, organisational permission, restricted data, specialist equipment or travel, access needs a route and a fallback.

Treating ethics as a formality

Ethics, privacy, safety and governance can affect recruitment, data collection, storage, analysis and dissemination. They belong in design planning.

Making the timeline start after approval

Proposal work should account for the time required to secure access, refine materials, obtain approvals where required, analyse the evidence, write, receive feedback and revise.

Writing to the wrong word limit

Aston publishes a word range for its own proposal guidance, while Oxford directs applicants to the relevant course page for the controlling limit. That difference is exactly why copying another institution’s number is unsafe.

Final checklist

Before submission, you should be able to answer yes to the following:

• I checked the current programme, school or department requirements.
• I can state the research focus without relying on a vague title.
• My rationale is based on evidence rather than enthusiasm alone.
• I have not invented an originality or “no previous research” claim.
• My aim and research questions follow logically from the problem and literature.
• My proposed methodology is justified for the question.
• I know what data, material, participants, texts, cases or technical resources the project requires.
• I have identified access and ethics/governance issues that could block the work.
• I have considered skills, software, equipment, budget and time where relevant.
• My timeline includes analysis, writing, feedback and revision, not only data collection.
• The proposal follows the required word limit, format and referencing rules.
• The final submitted work represents my own research decisions and complies with my institution’s academic-integrity and AI-use requirements.

A proposal is ready for serious review when the reader can see not only that the topic is interesting, but that the proposed research has a coherent route from question to evidence to analysis within the constraints of the actual programme. Approval itself remains a decision for the relevant supervisor, programme, admissions process, ethics body or other institutional authority; no generic guide can guarantee that outcome.