Somewhere between nine and eighteen months in, most doctoral programmes make you stop and justify the project formally. The name varies — confirmation of candidature, upgrade from MPhil to PhD, transfer, the qualifying or comprehensive exam — and so do the mechanics. The underlying purpose does not.
The panel is answering two questions on the institution's behalf. Is there a doctoral contribution in this project? And can this particular candidate complete it with the time and resources left? Everything they ask serves one of those two.
What the milestone is for
It is easy to read confirmation as a hurdle designed to catch you out. It is more useful, and more accurate, to read it as the last cheap opportunity to change course.
A project that is too broad, or whose data are unobtainable, or whose question has already been answered, is a problem at any stage. Discovered at confirmation, it costs a redesign. Discovered in year four, it costs the degree. The panel exists to catch that while it is still affordable, which is why they push hardest on scope and feasibility rather than on the depth of your reading.
This reframing matters practically: a panel that interrogates your design is doing its job, and a candidate who treats hard questions as attacks tends to perform worse than one who treats them as free consultancy.
What panels actually assess
Across institutions the criteria converge on five things.
- A researchable question. Specific, bounded, and answerable with data you can realistically obtain. This carries more weight than anything else in the document.
- A defensible gap. Not "nobody has studied X in my country", but a genuine limitation, tension, or unresolved disagreement in the existing work that your study addresses.
- A method that matches the question. Panels are quick to spot a method chosen because it is familiar rather than because it fits.
- Feasibility. Access to participants or data, ethics, equipment, skills, and above all time. This is where most conditions get attached.
- Evidence you can do the work. Usually a written chapter or two and the way you handle questions — whether you can defend a choice, and equally whether you can concede a fair point.
Preparing the document
Formats differ — anything from a ten-page proposal to two full draft chapters — but the reader is the same: a panel member who is not in your sub-field, reading it the evening before. Write for them.
- State the question by the end of page two. If a reader has to hunt for it, that is itself a finding about the project.
- Make the gap an argument, not a claim. Show the disagreement in the literature rather than asserting that a gap exists.
- Justify the method against the question, explicitly. One paragraph saying why this design and not the obvious alternative is worth pages of description.
- Include a real timeline with dated phases, including ethics. A plan that leaves four months for data collection you cannot possibly complete in four months invites exactly the questions you least want.
- Name your risks yourself. A short section on what could go wrong and what you would do about it is the single most effective thing you can add. It converts the panel's biggest objection into evidence of your judgement.
The presentation and the questions
Most confirmations involve a short talk followed by questioning. The talk is rarely where candidates come unstuck; the questions are.
Prepare a crisp answer to each of these, because some version of all of them will come up: What is your question, in one sentence? Why does it matter? Why this method? What will you do if you cannot recruit enough participants? What is the contribution? What will you cut if you run out of time?
Two behaviours distinguish strong performances. The first is conceding well: when a panel member identifies a genuine weakness, "that is a fair point — my thinking is X, but I take the concern and would address it by Y" reads as judgement. Defending an indefensible position reads as inflexibility, which is the trait panels worry about most in a candidate who still has years of supervision ahead.
The second is knowing your own limits. "I do not know, but here is how I would find out" is a strong answer at this stage. You are not expected to have finished the project. You are expected to know what you do not yet know.
The outcomes and what they mean
Terminology varies, but the outcomes map onto four categories.
- Confirmed. Proceed as proposed. Less common than candidates expect, because panels almost always have something to add.
- Confirmed with conditions. By far the most common. Revise the question, narrow the scope, add a pilot, resubmit a chapter. This is a normal, successful outcome, and reading it as a failure is a mistake.
- Deferred or resubmit. The panel is not yet satisfied and wants a substantially revised submission, usually within a few months. Serious, and recoverable — but engage with the feedback rather than re-arguing.
- Not confirmed. Rare, and usually the end point of problems that were visible long before. Often results in transfer to a different award rather than withdrawal.
Whatever the outcome, get the conditions in writing and confirm your reading of them with your supervisor. Ambiguity about what was required is a recurring source of trouble at the next milestone.
Using it properly afterwards
The panel has just given you a detailed, independent, free review of your project by people who are not invested in it. That is rarer than it sounds — you may not get another until the viva.
Write the feedback up within a couple of days, while you still remember the questions rather than just the verdict. Sort it into what you will act on, what you will note as a limitation, and what you disagree with and why. Take that list to your next supervision meeting. Candidates who do this tend to arrive at the viva having already answered the hard questions once.