Most Pakistani applicants write a research proposal the night before the deadline, copy the shape of a thesis introduction, and wonder why the result is a rejection. A proposal is not a summary of what you already know. It is a short argument that you found a real gap, that you can close part of it in the time the award allows, and that a specific supervisor is the right person to help you do it.
This guide covers the proposal for research-led applications: MEXT research student and doctoral routes, CSC postgraduate applications, DAAD doctoral research grants, HEC overseas PhD awards and the Commonwealth PhD scholarship. For the shorter study plan used in taught master's applications, read the study plan guide instead.
What universities say a proposal is for
Read what admissions offices write and one pattern repeats. The proposal is how a department decides whether your project is original, doable in the time allowed, and supervisable by someone on staff.
- The University of Manchester asks for evidence of intellectual purpose and originality, and says it needs the proposal to identify a suitable supervisory team. It also says you can refine details with that team after you are accepted, so you do not need a finished methodology on day one.
- Cambridge says a proposal should persuade the department that the project is valuable, achievable and a good fit for a supervisor or research group. It warns that you may not be admitted if no supervisor is available in your area, if the topic cannot be covered in the time allowed, or if the resources do not exist.
- The LSE Law School says applications without a proposal that answers its six questions (topic, questions, key literature and its limits, hypotheses, methodology, case studies) will not be considered. It also asks you to name three potential supervisors from permanent faculty.
How long it should be: the honest answer is "check the call"
Length varies so much that any single number would mislead you. These are figures published by the bodies themselves, as of 2026. Always check your own call.
| Body | What it says about length |
|---|---|
| LSE Law School PhD | No more than 1,500 words. Selectors are not required to read past that. |
| University of Manchester | Typically 1,000 to 1,500 words, but each School can set its own. Check your School. |
| Cambridge | No central figure. Check the course page or Applicant Portal, including whether footnotes count. |
| TU Munich, School of Medicine and Health exposé | 600 to 800 words of running text, bibliography excluded, signed by supervisor and applicant, in English or German. Other TUM schools differ. |
| DAAD doctoral research grants | The proposal and your description of previous research together can run to 10 pages, plus a schedule agreed with your supervisor. |
| CSC postgraduate study or research plan | Published copies of the CSC notice set a minimum of 800 words for postgraduates, and some Chinese provincial and university notices ask for 1,000. Follow the notice of the university you apply to. |
| MEXT, HEC, Commonwealth | We found no fixed figure on official pages. Use the embassy, HEC or Commission instructions for your round. |
Where CSC figures conflict, follow your university notice. If it gives a minimum, a 1,000-word plan covers both readings.
What our scholarship pages say about each program
- MEXT: a detailed, typed research proposal or study plan covering objectives, methodology and expected outcomes. Our scholarship page advises starting 3 to 4 months ahead and tailoring it to specific Japanese faculty. Use the Embassy of Japan in Pakistan's MEXT page for current forms.
- CSC: a study plan or research proposal, typed and signed. Type B applicants (applying through a university) also need to contact potential supervisors first.
- DAAD doctoral research grants: a research proposal, a time schedule and a letter confirming supervision by a doctoral supervisor in Germany. DAAD's database page also lists a declaration from the German supervisor on safety and ethical aspects of the project. Funding can run up to four years, usually awarded for three first.
- HEC overseas PhD: a research proposal is on the document list, and our scholarship page advises aligning it with HEC priority areas.
- Commonwealth PhD: your plan must fit one of the six development themes, you need to agree the plan with a UK supervisor, and it should show the benefit to your home country. Our scholarship page describes a 36-month award.
The standard sections and where the weight sits
Use the headings below unless the call gives its own template. If it does, follow the template exactly, even when the order feels wrong.
| Section | What it must do | Rough share of a 1,500-word text |
|---|---|---|
| Title | Name the topic and setting in one line. Specific beats clever. | One line |
| Background and problem | Set the field and state what is unknown, using a few sources, not thirty. | 15 to 20% |
| Research questions | Two to four questions your method can actually answer. | 5 to 10% |
| Literature gap | What existing work did and what it missed. | 15 to 20% |
| Methodology | Data, tools, steps, analysis, how you will check results, one risk and a fallback. | 30 to 35% |
| Timeline | Phases mapped to the funded duration. | A small table |
| Expected outcomes | Papers, a dataset, a model, a thesis, and who benefits. | 5 to 10% |
| References | Only what you cited, in one style. | Usually outside the count, but check |
The percentages are my reading of how reviewers weigh a proposal, not an official rule. They show where weak drafts go wrong: half the length on background, one paragraph on method.
Choose the supervisor first, then the topic
This is where Pakistani applicants lose the most ground. They take their BS or MPhil thesis topic and look for any professor "in the same area". Reverse the order.
- List 8 to 10 professors at your target universities. Use department pages and Google Scholar, not rankings.
- Read at least their last three papers each. Note their methods, their data, and the questions they say are still open.
- Pick a topic where what you can already do meets what those papers leave open.
- Email the professor a one-paragraph version of the idea. The guide to emailing professors has a structure that works.
- Rewrite using their reply. If a supervisor says "narrow it to X", that is what the committee will want.
Several universities say the same thing. Manchester and Cambridge both encourage you to talk to a possible supervisor before you submit. DAAD requires the proposal to be discussed with your academic adviser and a supervision letter. A proposal that names the lab's real method and data reads like the start of a collaboration. A generic one reads like a mass mailing. And do not say a professor's work inspired you unless you read it.
A problem and gap a reviewer can find in ten seconds
Put the problem and gap in plain sentences, early, not buried in a literature review.
A weak gap: "Not much research has been done on this topic in Pakistan." That is usually false, and never enough. A stronger gap has three parts: what is known, what is missing, and why the missing piece matters. The sample below uses a made-up applicant.
Sample only. Studies of canal irrigation scheduling in the Indus basin rely mostly on seasonal averages and rarely test whether farm-level sensor data can improve water allocation at the distributary level. Without that evidence, water authorities cannot judge whether low-cost sensing is worth deploying. This project tests that question on three distributaries using one season of sensor readings and compares the results with current scheduling practice.
It does not promise to solve water scarcity. It names one testable thing, one setting and one comparison.
A methodology you can actually finish
A reviewer asks one question here: could this person do this, in this time, with these resources? DAAD names feasibility and the consistency of the work schedule among its criteria for the project, and Cambridge says outright that a topic which cannot be covered in the time allowed is a reason for refusal.
- Say what data or material you will use and where it comes from. If it is a public dataset, name it. If it is fieldwork, name the site, sample and how you will get access.
- Describe the analysis in steps and name the technique.
- Say how you will check the result: a baseline, control or validation set.
- Admit one real risk and give a fallback. "If interview access is limited, I will use existing survey data" shows more maturity than a plan where nothing can go wrong.
Scale down. Pick one setting, one dataset, one main method. A committee prefers a small project with a clear path over a large one in fog.
Before and after: a methodology paragraph
Before (sample only):
The study will use modern machine learning techniques to analyse large amounts of data and produce accurate results that will benefit society.
After (sample only):
I will train a gradient-boosted model and a baseline logistic regression on a public dataset of hospital discharge records, splitting training and test sets by hospital to avoid leakage. I will compare both on recall for readmission within 30 days and report where they fail. The last months of year one are reserved for testing on a smaller local dataset, if access is approved.
Replace the invented details with your real dataset name and size.
A sample timeline table
A timeline shows you understand how research runs. Match it to the funded duration in your call. This sample is for a 36-month doctorate; stretch or shrink the columns for your award.
| Phase | Months 1 to 6 | Months 7 to 12 | Months 13 to 24 | Months 25 to 30 | Months 31 to 36 |
|---|---|---|---|---|---|
| Literature review and refined questions | X | X | |||
| Ethics approval and data access | X | X | |||
| Pilot study or method testing | X | ||||
| Main data collection or experiments | X | ||||
| Analysis and first paper draft | X | X | |||
| Second study and conference paper | X | ||||
| Thesis writing and defence | X | X |
Put the hard part, usually data collection, in the longest block, and leave a buffer at the end. For DAAD, the schedule must be agreed with your supervisor.
If you have no publications
Many applicants hold a BS or MPhil from a department where publishing is not routine and assume the proposal is dead before it is written. It is not. Some programs list publications or research experience for some routes, but a proposal can carry the weight instead.
- Cite recent papers accurately and name their limits. This is the nearest substitute for a publication record.
- Use your thesis or final-year project as evidence, and include a pilot result, even a small one.
Pair the proposal with an honest academic CV that states your research experience plainly.
Plagiarism, similarity scores and the HEC context
Manchester tells applicants the proposal must be their own original work in their own words, even after discussing it with a supervisor. In Pakistan, HEC's anti-plagiarism policy sets the similarity index it allows at 19% or less, and HEC thesis checklists ask for a Turnitin report under that figure. Treat that as a rule for theses and papers at Pakistani universities. A scholarship panel abroad may use its own check, and a proposal is short, so a few copied sentences move the number a lot. HEC's own guidance says the score is an alert, not proof of plagiarism, but a panel will not read it kindly.
Check the current HEC policy version, since HEC lists an updated Anti-Plagiarism Policy 2023.
- Write from your own notes, not from another proposal open on the next screen. A senior's proposal is for structure only.
- Paraphrase by understanding the paper, then writing without looking. Swapping synonyms is still plagiarism.
- Cite every claim that is not common knowledge, in one style such as APA or IEEE.
- Do not paste your own MPhil introduction unchanged.
- Cambridge asks applicants to read its policy on AI in postgraduate applications, and our GKS page says its screening looks for AI-generated and plagiarized content. If you use a tool for grammar, keep the ideas and argument yours, and check that every reference it mentions exists.
Mistakes that sink proposals
- A topic far too broad: "Impact of climate change on Pakistan's economy" is a career, not a proposal.
- No named supervisor or lab, and three pages of background over a four-line methodology.
- Claims of "novel" and "first" results with no sign you checked the literature.
- A timeline that ignores ethics approval, data access, or visa and arrival delays.
- Ignoring the call's theme. A Commonwealth plan outside the six development themes is weak from the first page, and HEC pages advise alignment with priority areas.
- Breaking the limit. LSE tells selectors they need not read past 1,500 words. Missing signatures count too (CSC asks for typed and signed, and the TUM exposé needs both signatures).
Checklist before you upload
- I read the official call and followed its limit and template.
- The problem and gap are stated in plain sentences on the first page.
- The method names data, tools, analysis and a way to check results.
- The timeline fits the funded duration and includes ethics, writing and a buffer.
- Every source is real and cited in one style.
- I ran a similarity check and can explain every match.
Keep the proposal consistent with the rest of the file. Your statement of purpose, CV and recommendation letters should all describe the same research direction. For the papers and attestations that travel with it, see the documents checklist for international scholarships.
Official pages worth opening before you write: the University of Manchester proposal guidance, the Cambridge research proposal page, the LSE Law PhD FAQs, the DAAD doctoral programmes page, and HEC's plagiarism policy page.
Frequently asked questions
How long should my research proposal be?
Follow your call. Published figures range from 600 to 800 words for a TUM medical exposé, 1,000 to 1,500 for a typical Manchester proposal, up to 1,500 for LSE Law, and up to 10 pages for DAAD doctoral grants. CSC notices vary between 800 and 1,000 words as a minimum. If nothing is stated, ask the department.
Do I need a supervisor's agreement before I submit?
For DAAD doctoral research grants, yes: you need a letter confirming supervision and a schedule agreed with your supervisor. The Commonwealth PhD asks you to agree the plan with a UK supervisor. Universities such as Manchester and Cambridge encourage contact before applying. Contact supervisors early everywhere.
Can I use my MPhil thesis topic?
You can build on it, and often should. But the proposal must move forward with a new question, dataset or method. Do not paste your thesis abstract and call it a proposal.
What if my field does not match HEC priority areas?
Our HEC pages advise aligning with priority areas. Check the current HEC call for the list, frame your question around the real problem it addresses, and do not bend the science to fit.
What similarity score is acceptable?
HEC's anti-plagiarism policy sets 19% or less for theses and papers at Pakistani universities, and PhD criteria add a lower cap for any single source. Scholarship panels abroad may set no number at all, so aim well below it, and check your own university's rules.
Start with the supervisor's papers, not a blank page. A proposal built on someone's real open questions, with a method you can finish, is ahead of most of the pile.