How to Plan a Bachelor's or Master's Thesis

Start with your degree requirements, not a chapter count

A bachelor's or master's thesis needs a defensible research question and a plan for answering it. A long outline is not enough: each chapter must have a purpose, suitable evidence and a realistic path to completion. This guide offers a planning worksheet, not a universal university template.

Collect your department's current handbook, assessment criteria, submission deadline and approved template. Record the required language, length, citation style and any rules about AI assistance. Ask your supervisor to resolve ambiguous requirements before drafting. Do not assume that a structure accepted in another course will satisfy yours.

Terminology and structure vary across institutions and disciplines. Monash University's thesis-structure guidance illustrates this variation. Use your own institution's requirements to decide which sections belong in the submitted document.

Write a one-page thesis brief

Before generating an outline, fill in six fields: the question; the object or population being studied; the method; the sources already available; the boundaries; and the next decision needing supervisor feedback. A missing field identifies a planning task, not a reason to invent an answer.

For example, “AI in education” names a field rather than a manageable question. An illustrative narrower question is: “How do two selected university policies describe student disclosure of AI assistance?” That question suggests a defined document set and a comparative analysis. It does not establish what those policies say or whether the comparison is sufficient for a particular degree.

State exclusions alongside the question. A document comparison does not measure student behavior. A model evaluation on one dataset does not establish performance in every deployment setting. These boundaries make later conclusions easier to check.

Match the structure to the kind of research

Empirical or experimental work

Plan how methods, data provenance, analysis and findings will connect. Keep a protocol separate from completed results. If a comparison has not been run, describe it as planned and leave the result-dependent argument unfinished. A paper about an algorithm cannot supply evidence of your implementation's performance.

Literary, historical or other interpretive work

Organize chapters around an argument, theme, case or comparison. Record primary-text editions and passage locations, distinguish your interpretation from another scholar's, and address alternative readings. A chapter can require close reading or documentary evidence without requiring an experiment report.

Literature-based research

Define what literature you include and how you compare it. A set of summaries is not yet a synthesis. Keep your search and selection decisions visible, and do not call a review systematic unless its methods justify that description. Confirm the suitability of a literature-only project with your supervisor.

A chapter-planning worksheet

The rows below are an illustrative planning aid, not a required five-chapter thesis. Split, combine or replace them according to the question and approved format. Give each actual chapter its own row and connect it to a claim-to-source evidence map.

Separate missing evidence from unfinished writing

For every dependency, record its source, owner and next action. Public literature may need further retrieval; a private dataset may need permission from its owner; an experiment may still need execution. These are not interchangeable problems, and repeated requests to continue writing cannot resolve all of them.

Draft sections whose evidence is ready while tracking the sections that remain blocked. Keep placeholders in working notes rather than treating them as final results. If a source cannot be obtained, discuss narrowing the question or removing the unsupported claim. Do not replace missing research with plausible-looking numbers.

Plan revisions before the submission deadline

Schedule separate milestones for source collection, a reviewable chapter draft, feedback, revision and final formatting. Reserve time to check cross-chapter consistency after individual chapters have been revised. The University of Sheffield's dissertation-planning guidance provides further advice on planning and managing the research project.

Review evidence and expression separately. First check whether citations, quotations and numerical claims can be traced. Then examine structure, transitions and academic wording. Correct prose does not repair unsupported evidence, and a valid source does not automatically make an argument coherent.

At assembly, check abstract requirements, bibliography consistency, figure placement, captions and the actual exported document. Compare word counts with your institution's counting rules, not just a generic minimum. Prepare any required declarations, acknowledgements and appendices yourself where the workflow does not provide them.

Use writing tools without outsourcing research judgment

Paper Terminator organizes scope, evidence, outlines, chapter checkpoints and review. It does not replace research, supervisor approval or institutional assessment. Selected abstract and unified-reference options are part of manuscript assembly; they are not evidence that the manuscript satisfies every university rule.

Read the research writing workflow and inspect the public spam-classification example before deciding whether it fits your project. The example shows one project, not a template every thesis should copy. Use only AI assistance permitted by your institution and verify any generated draft yourself.

Check your draft length

Use the free academic word counter to count the full paper or exclude references.

Organize your bibliography

Use the reference converter and deduplicator for GB/T, APA and IEEE formatting.