Thesis Studio

How to do a master thesis

Make a real contribution, but control the scope tightly enough to finish.

A master thesis should show independent research judgment. It needs a sharper question, stronger method, and more defensible contribution than a class project.

gumu supports long-running thesis work by keeping drafts, source PDFs, experiments, supervisor notes, and PDF edits in one workspace.

On this pageWorkflow Structure Checklist Mistakes FAQ

Workflow

Do the work in the right order.

Define the contribution level

Decide whether the thesis contributes a method, system, dataset, experiment, theory, replication, or evaluation. This controls the entire plan.

Ask gumu to classify your idea by contribution type and identify what evidence each type requires.

Write a thesis proposal

Before heavy execution, write the question, motivation, related work, method, expected outputs, risks, and timeline.

Use the proposal helper to turn rough notes into a supervisor-ready plan.

Run a rigorous literature review

A master thesis needs synthesis, not a source list. Use the review to justify both the question and the method.

Build a literature matrix and ask for gap statements, competing methods, and missing baselines.

Design evaluation before implementation ends

Define datasets, cases, baselines, metrics, or analysis criteria early. This avoids a thesis that only describes what was built.

Ask gumu to turn your contribution into research questions, evaluation metrics, and baseline requirements.

Write while executing

Draft background, method, and experiment design while work is ongoing. Save the final result narrative for when evidence stabilizes.

Keep experiment logs and chapter drafts synchronized so changed results update the thesis text.

Prepare the defense from the manuscript

Your defense should explain the problem, contribution, evidence, limits, and next steps. Do not create a different story from the written thesis.

Ask for likely defense questions and a slide outline tied to thesis chapters.

Structure

Use a format that makes the answer obvious.

Abstract and contribution summary.
Introduction with question, stakes, and thesis claims.
Related work or background with a clear gap.
Method, system, dataset, or theoretical framework.
Evaluation, analysis, or case studies.
Discussion, threats to validity, and limitations.
Conclusion and future work.
Appendices for details that support but distract from the main argument.

Checklist

Before you call it done.

The contribution type is explicit.

The literature review explains why the contribution is needed.

The method is reproducible enough for the field.

Evaluation choices match the research question.

Limitations are specific and defensible.

The defense story matches the written thesis.

Mistakes

Common failure modes.

Building too much and evaluating too little.
Using the literature review as a background dump.
Changing the contribution after results are known without revising the thesis frame.
Leaving writing until the implementation is complete.
Preparing defense slides that overclaim beyond the manuscript.

How gumu helps

One workspace for the argument and the artifact.

Start from a rough prompt, paper draft, PDF, review, call text, or supervisor note. gumu keeps the chat, source, references, and PDF regions together so every answer can become a concrete edit.

Try it in gumu

FAQ

Practical answers.

What is a good master thesis topic?

A good topic has a clear contribution type, accessible data or materials, a feasible method, and a way to evaluate the result.

How original does a master thesis need to be?

It usually needs a meaningful independent contribution, but it does not need to redefine a field. A careful replication, extension, or evaluation can be strong.

When should I start writing?

Start once the proposal is approved. Early chapter drafts reduce risk and expose missing definitions, sources, and evaluation details.