At European universities, the student runs the thesis project and the supervisor guides it. Supervision guidelines from Norway, Sweden, Denmark and Finland describe much the same division: you choose and own the topic, book the meetings, send text on time and make the basic decisions; the supervisor advises on topic, structure and method, reads and comments, and does not fix every detail. This page sets out that division with quotes, and lists what to agree on in the first meeting.
No one sees their own blind spots, with or without AI help. Esmond points them out in your draft from within your field, along with factual errors and inconsistencies.
The University of Helsinki gives its teachers a table of rights and obligations for both sides. Among the student's obligations: the student “makes the basic decisions concerning their thesis”. Among the supervisor's: the supervisor “provides expertise-based feedback on the student’s work and reads and comments on the thesis manuscript”. The other universities quoted here divide the work in a similar way.
| Task | Usually the student | Usually the supervisor |
|---|---|---|
| Topic and research question | Proposes and is responsible for the choice (NHH) | Advises and may suggest changes (NHH, Lund) |
| Meetings | Requests and books them, with the points to discuss (NHH, LTH, Aarhus) | Gives feedback within a reasonable time (NHH) |
| Drafts | Sends text on time and prepares for meetings (Helsinki) | Reads and comments on the manuscript (Helsinki) |
| Method and structure | Decides and justifies | Advises on structure, method and how to work (NHH, Lund) |
| Language and data | Checks their own text and data | General comments only; no detailed language check or error check of data sets (NHH, LTH) |
| Schedule | Keeps to the agreed schedule (Helsinki) | Helps structure the process into steps (Helsinki) |
A hypothetical example of a commentYour introduction says the study covers 2015 to 2024, but the data chapter stops in 2021. Is the period you name the one your analysis actually covers?
A first comment on one part of your own draft, right away and without an account.
Get a free previewSeveral guidelines put the agreement at the start. NHH's guidelines, which it says should be reviewed together in the first meeting, ask both sides to agree on the format of supervision, the expected progress and the intended submission date, and add: “There should be at least three meetings between supervisor and students”, at the beginning, in the middle and towards the end.
The reason is practical. Helsinki's guidance for supervisors notes: “Ambiguities related to supervision are often due to the parties’ different expectations regarding the content and responsibilities of the supervision and the fact that the parties are often unaware of the others’ expectations.” Helsinki suggests a written supervision plan or agreement, which can also be used if problems arise or the work falls behind schedule.
Aarhus University's student guide lists questions to settle at that first meeting: how often you will meet, how you communicate between meetings, when either of you is unavailable, and “What, and how much, of your text will the supervisor read?” That last question decides a great deal of your planning, and it is easy to leave unasked.
In most of these guidelines, you do. The Department of Design Sciences at Lund University (LTH) tells students: “As a rule, you are responsible for initiating supervision meetings.” It asks you to send the date, time and points to discuss in advance, so the supervisor can prepare. NHH says students are expected to request meetings. Aarhus puts it this way: “Often, it will be your responsibility to contact the supervisor, schedule the meetings, and communicate clearly what you need from the supervision.”
For a student used to a teacher who sets the pace, this can be the biggest change. A supervisor who hears nothing for two months may assume the work is going well.
Supervisors read for the thesis as a whole. NHH's guidelines say the supervisor should give overall, constructive comments on all parts of the thesis, along with general comments on the language. They also set the limits: the supervisor “Is not responsible for pointing out every detail that can be improved or provide detailed feedback on language.” and “Is not expected to check excel sheets/data sets/analysis for errors.”
LTH says the same about language: the supervisor is not responsible for correcting it, in Swedish or English, and it points students to the university's academic support service for that. So the detail stays with you. If a table in chapter four contradicts a sentence in chapter five, your supervisor may notice, but the guidelines don't promise it.
Aarhus suggests sending drafts well in advance with a short note: how rough or finished the draft is, what you are unsure about, and what you want feedback on. That note tells the supervisor where to spend limited reading time. For how to handle the comments that come back, see how to read advisor feedback.
European guidelines stress that the thesis is the student's work. An NTNU memo for company supervisors states: “A master thesis is independent work.” It says a supervisor can give direction and feedback on ideas, but should not hand the student a solution; the student comes up with the ideas and proposes solutions.
Helsinki describes the supervisor's position from the other side: “The supervisor plays a dual role: while you provide support to a goal-oriented project in accordance with the objectives set for the thesis, you also have to support the student as an independent actor.”
Independence does not mean working alone. Aarhus warns against "performing" for your supervisor by hiding doubts and unfinished text, and adds: “Remember: if you were expected to know everything in advance, the whole idea of supervision would be meaningless.” Showing an early, rough chapter is part of the arrangement. Some programs also assess independence under supervision as part of the thesis, so check the assessment criteria your program uses.
The amount varies by university and department, and it is often written down. At Lund's Faculty of Social Sciences: “If the right to supervision is limited in scope and time, information on this is to be included in the course syllabus.” The Department of Biology at NTNU puts a number on it for its own students: “The main supervisor is committed for 25 or 50 hours of guidance of the master student (25 hours per 30 credits master thesis and 50 hours per 60 credits thesis).”
That figure is one department's rule. Look up the figure for your own program, if there is one, and plan your meetings around it.
In many European systems your supervisor is not the only person, or not the person, who decides on the thesis. At Lund's Faculty of Social Sciences the supervisor and the person who examines the degree project cannot be the same, and the project is presented at a seminar with one or more appointed reviewers. NHH's guidelines add that the supervisor should not commit to, or suggest, the result the student can expect.
A supervisor's approval of a chapter is therefore advice; the decision on the thesis comes later. The final readers come to the text without the history of your meetings, so the thesis has to make its case on paper.
Aarhus advises raising problems directly with your supervisor first, and talking to your student advisor if a change of supervisor seems necessary, which it says is rare. Lund's Faculty of Social Sciences sets out the right to a change: “The student has the right to be allocated a new supervisor within a reasonable period of time, if there are special reasons for this.” At NHH, problems the two parties cannot resolve together must be reported to the section for exams.
Before that step, the written agreement from the first meeting is useful: it shows what both sides agreed to.
This example is made up. A master's student in public health agrees on a topic in September and then works alone until February, planning to show a complete draft. The supervisor, who expected monthly meetings, reads the draft and finds the research question narrower than the one agreed, and a survey that no longer fits it. The arrangements above are meant to catch this earlier: a short note with the first rough chapter, sent in October with the question "is this still the question we agreed on?", costs an hour and saves a rebuild.
Drawn from the guidelines quoted above. Your program's rules come first.
Esmond evaluates every assignment before you submit. It never writes it.
Esmond is an AI tool that doesn't write for you. If your course doesn't allow any AI tools, don't use Esmond for that work. If you're unsure, ask your instructor or advisor.
Esmond is not a supervisor and doesn't replace one. It reads the draft you made before you send it and flags factual errors, inconsistencies and blind spots, the kind of detail the guidelines above leave to you. It adds no text to the thesis. Your supervisor and your committee keep the final say. Tell your supervisor if you use it.
For who else can read a draft first, see feedback before your supervisor reads it. For the European framework behind the review, see why a European review.
In the guidelines quoted here, the supervisor advises on the topic, structure and method, reads and comments on the manuscript, and helps structure the process. The supervisor is not expected to point out every detail, give detailed language feedback or check your data for errors.
Usually the student. LTH at Lund says that as a rule you are responsible for initiating meetings, and NHH and Aarhus say much the same.
It depends on your program. NHH sets at least three, at the beginning, middle and end; NTNU's Department of Biology commits a number of hours per credit size. Check the rule in your own program and agree on a schedule early.
Often yes, for good reasons. Lund's Faculty of Social Sciences gives students the right to a new supervisor within a reasonable time if there are special reasons. Talk to your supervisor first, then to the student advisor or the office your program names.
Read on November 7, 2026.
Get your paper evaluated before you submit