Two prior pieces on this site benchmarked named institutions’ own published doctoral requirements — MIT and Cambridge’s thesis format regulations and TU Delft’s propositions and plagiarism-scan requirements. Both are useful, and both share a structural trait: they are format-and-submission rules layered onto an admissions process that looks broadly similar to most research universities’. The Technical University of Munich (TUM) is a genuinely different case. Its own graduate-school pages, read directly for this piece, describe an admission model built around securing a named supervisor before you can apply at all, five distinct doctoral pathways rather than one generic route, no tuition fee, and an academic threshold expressed as a specific grade cut-off with a public conversion formula for international applicants. For a policy team building a comparative reference set, that is a materially different institutional shape worth having on file.
Why a third benchmark, and why TUM specifically
MIT’s thesis specifications are a format-and-submission document: title-page components, an abstract length guideline, PDF/A-1 requirements. Cambridge’s published rules are a word-limit regime set by Degree Committee, with a formal exceptions process. TU Delft’s doctoral regulations add a propositions requirement and a mandatory pre-approval plagiarism scan run by the library before submission. All three answer the question “what does the finished document have to look like, and how is it checked?” TUM’s own pages answer a prior question that the other three largely assume is already settled: “how does a candidate get admitted to a doctorate here at all, and under what kind of appointment?” That is the gap this piece fills.
| Institution | Country | What its own published rules are mainly about |
|---|---|---|
| MIT | US | Thesis format and submission specification (title page, abstract length, file format) |
| Cambridge | UK | Word-limit regime by Degree Committee, with a formal exceptions process |
| TU Delft | Netherlands | Propositions requirement and a mandatory pre-approval plagiarism scan |
| TUM | Germany | Admission-gate design (supervisor-first), doctoral pathway typology, fee and funding model, academic threshold |
The admission gate: a supervisor before an application
TUM’s own guidance is unusually direct about the order of operations. Its published requirements page states the point without qualification: “If you have not found a supervisor for your research project, you will unfortunately not be able to do your doctorate at TUM.” Finding a professor willing to supervise the specific proposed project is described as the most critical step in the whole application, ahead of assembling academic transcripts or a formal research proposal, and the expected route is direct email contact with individual professors rather than a central application portal that assigns a supervisor after acceptance.
This matters for a benchmarking exercise because it inverts the sequence many institutions run. Where an institution centrally allocates supervisors to admitted candidates, replicating TUM’s model wholesale would require restructuring the admissions office’s relationship with academic departments, not just adopting a form. Where an institution already expects applicants to secure informal supervisor interest before a formal application — common in many STEM fields even without TUM’s explicit framing — TUM’s language is a useful, citable articulation of a practice a committee may already be running informally.
What a supervisor-first gate would require of your own admissions process
Adapting TUM’s model, even partially, is an operational change before it is a policy change. Three questions a committee tends to surface once it looks seriously at this: first, does the institution publish, department by department, which faculty currently have capacity to take on a new doctoral candidate — without that list, “find a supervisor first” becomes an unstructured cold-email exercise rather than a navigable process; second, who owns the decision when a candidate secures informal supervisor interest but the formal application is later rejected on academic grounds — TUM’s own pages frame the supervisor conversation and the formal academic threshold as two separate gates, both of which must be cleared; third, how does the institution avoid the gate becoming a de facto informal-network filter that disadvantages applicants without existing contacts in the department, a fairness question TUM’s own published pages do not address and that any adopting institution should expect its own equity or admissions-oversight office to raise.
Five doctoral pathways, not one
TUM’s graduate-school pages name five distinct routes into a doctorate, rather than treating “PhD candidate” as a single undifferentiated status:
| Pathway | Basic shape |
|---|---|
| Doctorate as a Research Associate | Employment-based — the candidate holds a university staff contract |
| Doctorate with Scholarship | Funded by an external or internal scholarship rather than an employment contract |
| Cooperative Doctorate | Run jointly with an industry or external-organisation partner |
| External Doctorate | Independent research, typically for a candidate not resident at or employed by TUM |
| Guest Doctorate | For a visiting researcher already pursuing a doctorate registered elsewhere |
Naming these as distinct, formal categories — each presumably carrying its own funding, supervision-intensity and reporting expectations, though TUM’s introductory pages do not detail each pathway’s specific requirements individually — is itself the useful comparative data point. An institution auditing its own policy for gaps (a common exercise behind pieces like this site’s postgraduate enrolment by field of study analysis) can ask a concrete question prompted by this typology: does our own doctoral population already sort into functionally distinct groups — staff, funded, industry-partnered, external, visiting — that our formal policy currently treats as one undifferentiated category? The Cooperative Doctorate route in particular is worth flagging to any institution running or considering industry-sponsored doctoral programmes: naming it as a distinct formal pathway, with presumably its own disclosure and intellectual-property conventions, is a more legible structure than folding an industry-sponsored project into the same generic policy that governs a fully independent academic doctorate.
Academic threshold: “above-average,” with a named conversion formula
TUM’s stated bar is a degree described as “above-average” — a master’s, Diplom or Magister — which its own pages operationalise as a final grade of 2.5 or better on the German grading scale (where lower numbers indicate a stronger result), or an equivalent “good pass.” For international qualifications, TUM points applicants to the Bavarian Formula as a self-estimation tool for converting a foreign grade onto the comparable German scale, while stating plainly that “the responsible TUM School always decides on the recognition of your degree” — the formula is a candidate-facing estimation aid, not a binding determination. TUM does not offer a bachelor’s-only fast-track route into a doctorate on the pages reviewed.
No tuition, no central deadline
Two further, easily overlooked data points from TUM’s own pages: doctoral study at TUM carries no tuition fee, with candidates expected to cover living costs through an employment contract, a scholarship, or another funding source depending on which of the five pathways applies; and TUM states there is no general, university-wide application deadline for doctorates — individual TUM Schools or chairs may set their own, but absent that, candidates can be admitted at any point in the year. Both are structural, budget-relevant facts for a policy team modelling what a comparable no-tuition or rolling-admission policy would actually require to administer: a no-tuition model shifts the institution’s cost recovery entirely onto the employment, scholarship and external-funding routes that fund each of the five pathways, and a rolling-admission model shifts workload from a small number of concentrated intake cycles to a continuous, year-round administrative process that a batch-oriented admissions office may not currently be staffed for.
What TUM’s pages did not specify, on this review
In the interest of the same honesty this site’s TU Delft piece applied to its own gaps: the two TUM graduate-school pages reviewed for this article reference a “Good Scientific Practice” framework by name, without this review locating the specific provisions of that framework on those pages, and neither page detailed the examination or defence procedure, or stated an institution-wide AI-use or plagiarism-detection policy for doctoral submissions. That is a materially different disclosure posture from TU Delft, whose own regulations page states a mandatory pre-approval library plagiarism scan explicitly and by name. This piece does not assert TUM has no such policy — only that it was not found on the pages opened for this review, and a policy team citing TUM’s practice on integrity screening specifically should locate and open TUM’s own Good Scientific Practice document directly rather than relying on a secondary summary.
Using this as a fourth data point, not a template
The value of a named-institution benchmark series like this one, this piece and its two predecessors, is not that any single institution’s rules should be copied. It is that a committee drafting or revising its own doctoral policy gets to see, in the institutions’ own published language, how much genuine variation exists across admission-gate design, pathway typology, fee model and disclosure depth — and can use that variation to test whether its own draft policy is actually deliberate, or just an unexamined default. TUM’s supervisor-first gate and five-pathway typology are the two elements from this review most worth raising explicitly in a policy committee discussion, whether or not the eventual decision is to adopt anything resembling them. A useful next step for a policy office running this kind of exercise is to hold its own current doctoral handbook up against this four-institution table and mark, honestly, which cells are a deliberate institutional choice and which are simply silent.
Where Tesify fits
None of the admissions, pathway or funding decisions above are ones a writing platform touches — they stay with the graduate school and the individual chair or department. Where Tesify for Institutions is relevant is downstream of admission, once a candidate is registered under whichever pathway applies: giving supervisors visibility into a doctoral candidate’s drafting progress against the institution’s own formatting and citation conventions, regardless of which of the five TUM-style pathways funds the candidate’s place. A free departmental pilot lets one doctoral programme test that visibility before any procurement decision.
Frequently asked questions
What makes TUM a useful third institutional benchmark alongside MIT, Cambridge and TU Delft?
Its admission model is structured around securing a supervisor before applying, it offers five distinct named doctoral pathways rather than one generic PhD route, and it charges no tuition — a genuinely different profile from the format-and-word-limit rules MIT and Cambridge publish or TU Delft’s propositions-and-plagiarism-scan model.
Does TUM require a minimum grade for doctoral admission?
TUM’s own pages state applicants need an “above-average” master’s, Diplom or Magister degree, operationalised as a final grade of 2.5 or better on the German scale, or an equivalent “good pass.” International applicants can self-estimate using the Bavarian Formula, though the responsible TUM School makes the final recognition decision.
Can you start a TUM doctorate without a supervisor already agreed?
No. TUM’s own admission guidance states this directly: without a secured supervisor for the specific research project, admission is not possible, regardless of how strong the academic record is.
Does TUM publish a fixed doctoral application deadline?
No. TUM states there is no general, university-wide deadline for doctoral applications unless a specific TUM School or chair sets one, and candidates can be admitted throughout the year.
Did TUM’s own pages specify an AI-use or plagiarism-scan policy for doctoral submissions?
Not on the pages reviewed for this piece. They reference a “Good Scientific Practice” framework by name without detailing its provisions, which is a materially different disclosure level from TU Delft’s own pages, which state a mandatory pre-approval plagiarism scan explicitly.
Does TUM offer a route into a doctorate straight from a bachelor’s degree?
Not on the pages reviewed. TUM’s stated prerequisite is an above-average master’s, Diplom or Magister — a fast-track bachelor’s-only route was not found on the pages opened for this review.
How should a policy team use a benchmark like this one?
As one documented data point in a small, named-institution set — useful for showing a committee that admission-gate design, pathway structure and fee models genuinely vary by institution, not as a template to copy without adapting to the reader’s own governance structure.