Nobody warns you about the upgrade viva.

Candidates arrive knowing there is a viva at the end. Far fewer arrive knowing there is one in the middle — and that at a good many UK universities you are not registered for a PhD until you have passed it.

Birkbeck states the position in one line: PhD programme students are initially registered for an MPhil degree, and proceeding to the PhD requires an assessment to change that registration. “This is known as the ‘upgrade’.” Sheffield frames the same idea differently: the first year is “a probationary year, pending successful completion of the Confirmation Review”. Either way, the doctorate you think you are doing is provisional until a panel says otherwise.

Four names, one gate

The single biggest reason candidates cannot find guidance is that they are searching for the wrong word. Here is the same milestone at five UK institutions, taken from their published regulations:

There is also “transfer viva”, which is what a great many departments call it in conversation regardless of what the regulations say. If you have been told you have a transfer viva and your university's website has nothing under that name, search for confirmation, progression, upgrade and progress review before you conclude the guidance does not exist.

Most candidates never read the criteria

This is the part that does real damage. The panel is not forming a general impression. At York, a candidate is “assessed against the relevant University progression criteria by a progression panel”. At Sussex, the assessors' job is to determine whether you have “met the University criteria for progression relevant to your stage of registration”. At Loughborough the criteria sit in a numbered paragraph of a named regulation. Birkbeck publishes a list of numbered criteria and then says the expectation is that the written submission plus the upgrade viva together demonstrate that these criteria have been met.

So there is a document. It is usually public. Most candidates have never opened it, and prepare instead for the viva they imagine. Find yours before you write a word of the submission, and if your school sets additional local requirements — Sheffield explicitly leaves the minimum written submission to schools and faculties — ask for those in writing too.

The outcomes are not pass and fail

Candidates brace for a binary that mostly does not exist. The published outcome sets are graduated, and the middle of the range is where most difficulty lands.

Sheffield's Confirmation Review can be deferred “pending a second attempt within 18 months of the start of the student's study, either with or without a further viva”. Birkbeck allows a second attempt, and a candidate may be referred to resubmit the written work only, or to sit a second oral, or both; where a first attempt is unsuccessful the candidate “must be provided with feedback detailing any additional requirements and support necessary”. Sussex gives a second opportunity within three months of formal notification. York's panel may recommend progression, a second and final opportunity, transfer to an MPhil, or withdrawal.

And then the detail almost nobody knows, which is worth reading twice. York's policy states that at the first attempt a candidate “is not obliged to accept a recommendation for transfer or withdrawal and may instead opt for a second and final opportunity” to meet the criteria. If a panel recommends dropping to an MPhil, that may be a recommendation you can decline in favour of trying again. Check your own institution's wording — but check it, rather than assuming the first outcome is the last word.

The flip side is real and should be said: transfer to MPhil has consequences beyond the title. Sheffield notes that candidates transferred to MPhil have their time limit reduced to three years, and that sponsored candidates are responsible for telling their sponsor, “as this is likely to have an impact on the conditions of their sponsorship”. For an international candidate, registration changes can reach further still. That is a question for your institution's international student team, in writing, early.

Who is in the room, and who is not

The panel is deliberately not your supervisor. Sheffield requires an impartial panel of at least two academics “who have not had any previous close association with the student's research project”. York's progression panel is “independent of a PGR's supervisor(s)”. Birkbeck's panel is two academic staff with the relevant subject expertise, and — a detail worth noticing — its members are ineligible to examine the eventual thesis.

Your supervisor's role is usually to write a report, not to advocate in the room. Birkbeck requires a written report from the principal supervisor in advance, including a plan and timetable for the remaining work. Sussex has the main supervisor submit a confidential report on progress against the criteria. Where supervisors may attend at all it tends to be as a silent observer, at the candidate's request and with permission, withdrawing before the decision.

The practical consequence: this is the first time a stranger reads your work cold. Everything your supervisor fills in from memory, a panel will ask you to say out loud.

What the panel is actually testing

Not a finding. A plan, and your capacity to execute it. Birkbeck says the assessment “should not be confined to the research material presented by the student but should also consider the student's demonstrable academic and generic skills”. In practice, four things:

1. Is the question doctoral, and is it bounded? Birkbeck expects a research question or hypothesis, while acknowledging these evolve.
2. Does the design answer that question? Methodology and, where relevant, ethical approval. A literature review plus methodology can be the substance of a first-year upgrade where fieldwork comes later.
3. Is the remaining work deliverable in the time left? Hence the timetable in the supervisor's report. A plan with no slack reads as a plan that has not been thought about.
4. Can you defend choices as choices? Why this design and not the obvious alternative, and what it costs you.

That last one is where candidates who know their material still come unstuck, and it is pure technique. The questions examiners ask guide applies directly, as long as you remember the register is different: at the final viva you defend what you did, at the upgrade you defend what you propose to do.

Six weeks out

• Week 6: Obtain the written criteria and any school-level submission requirements. Read them against your draft and list the gaps.
• Week 5: Find out the panel's composition and the format — presentation or no presentation, length, whether questions come from the submission only.
• Week 4: Write the timetable for the remaining work yourself, before your supervisor writes theirs. Argue the two together.
• Week 3: Prepare a sixty-second, a three-minute and a ten-minute account of the project. You will need whichever one they ask for.
• Week 2: Rehearse the five hardest questions about your design with someone outside your project who will not fill in your gaps for you.
• Week 1: Re-read your own submission as a stranger would. Mark every claim you could not defend if pressed twice.

About this post

Related guidance

Sources

Primary sources and UK institutional guidance only. No competing commercial page is cited or paraphrased anywhere in this post. Every external link was checked on 9 October 2026.