16 September 2026 · 7 min read
Ask a student how long their thesis will take and they will estimate writing time. Ask a supervisor and they will tell you how many rounds of comments each chapter needs. The second number is the one that decides the submission date, and it is almost entirely outside your control — which is exactly why the schedule has to be built around it.
Take your supervisor's realistic turnaround — not the one they promised, the one they have actually managed so far — and multiply. A fortnight per round, three rounds on a chapter, and that chapter occupies six weeks of calendar time however fast you write. Five chapters handled strictly one after another does not fit in a semester, and that is the whole problem in one line.
The single most useful habit is having something to work on while a chapter sits with your supervisor. Students who send chapter two and then wait lose two weeks per round; students who send chapter two and start the instrument, or the data entry, or the reference list, lose nothing.
You cannot make a supervisor faster, but you can change what you hand them. Send a chapter that has already been proofread, formatted to the department's template and checked for similarity, and the comments come back about the argument. Send a rough draft and the comments come back about typography — a whole round spent on something you could have fixed yourself.
Most students reading this are not planning a thesis, they are rescuing one. The order of recovery matters: secure the data first, because nothing else can be finished without it, then write results, then methodology, then discussion, then the literature review, and the introduction and abstract last. The literature review is the chapter students start with and the one that can most safely be finished late.
Tell your supervisor you are behind before they work it out. A supervisor who knows in week eight will help you cut scope to something finishable; the same supervisor told in week fifteen has nothing left to offer but an extension request.
Put the similarity and AI checks in the plan as dated tasks, chapter by chapter as chapters settle, plus one on the assembled document. A report that arrives the night before submission tells you about a problem you no longer have time to fix, which is the same as not running it.
Related:
Send your deadline and where you actually are today, and we will tell you plainly whether the plan fits — including when it does not.