Two hours a fortnight with your advisor. Quester is for the other thirteen days.
A dissertation is the longest piece of unsupervised work most people ever attempt. Five to seven years, several hundred sources, no syllabus, and an advisor who, however good, has about two hours for you a fortnight. Quester is the workspace built for that gap, on patterns drawn from decades of getting students across the finish line.
Nobody fails a PhD because they were disorganised.
People fail, or take nine years, for a small set of specific reasons, and every one of them is invisible from the inside while it is happening.
The literature review is a list, not a synthesis
You have read ninety papers and can describe each one. What you cannot do is say how they speak to each other, because that relationship only ever existed in your head and it is degrading, which is why more reviews come back for major revision over this than over anything else.
The question quietly drifts
The question you are answering in year three is not the one you defended in your proposal, and nobody caught the moment it changed, including you. Sometimes that is growth, and sometimes it is a retreat from something hard that you find out about at the viva.
The corpus goes stale without a sound
The papers everyone cites are the papers you collect, and the papers everyone cites are old. Reviews drift a decade behind the field by accident, and the person who notices is usually sitting on your committee.
It stops feeling like it is moving
This is the one that ends doctorates, because six months of real work feels from the inside like six months of nothing when nothing exists that shows the accumulation. The feeling is almost always false, and almost never checkable.
Each of those has a surface pointed at it.
These are not features looking for a use, since the structure exists because these are the failures worth engineering against.
It reads your library from a distance
The Map takes your corpus and tells you what it actually looks like: how much predates 2015, whether it is 90% journal articles, which countries the authors sit in, which concepts recur that you never named. You cannot see your own selection bias by reading harder.
It shows you where synthesis is missing
The Connections graph carries a synthesis-health reading: where your sources hang together through a note, a question or a model, and where they are still a pile. A thin region is not a failure so much as the next paragraph of your review, already located.
It keeps the question honest
Every revision of every research question is kept, so you can put what you asked in month four beside what you are asking now, while questions get scored against your own design rubric and the weak dimension is named, which gives you a next move rather than a grade.
It keeps the receipts
The Journal logs what you did as you do it, across sources, notes, questions and models, and you can export it, so when the work stops feeling like it is moving you get evidence instead of encouragement, which works considerably better.
An advisor that has read your project.
Not a chat window with your PDFs attached. It opens knowing your sources, your notes, your questions, your models and what you did last month, so the conversation starts at "three new sources arrived; do they fit your questions?" instead of at nothing.
Scope it to five specific papers and ask whether they support the claim you are about to make. Turn its pushback up to adversarial before a committee meeting and let it argue the counter-case. Give it hard rules, like never suggest quantitative methods for this project or always flag a source older than 2015, and those rules beat every other setting.
It will not draft your chapters, invent a citation, flatter you, or assert a conclusion your data cannot support. Ask it to and it declines in a sentence and says why.
That refusal is the product rather than a limitation of it, because in two years you have to defend this work in front of people who will ask where every part of it came from, and a tool that happily writes your literature review is not on your side.
Turn the AI off. The workspace still works.
The advisor is a per-project switch, and off means genuinely absent, with no observations, no prompts and no cards. What stays is the structured workspace: your library, your notes with real links to the sources that prompted them, your questions and their history, your models, your bibliography in any citation style, and the record of everything you did.
Plenty of people work this way for a year and then switch the advisor on, while plenty never do, and either way it is a supported way to use Quester rather than a degraded one, which also gives you a straight answer when someone asks what you are using.
Ten minutes, then it is yours.
Export your Zotero library to BibTeX or RIS and drop the file in. Hundreds of sources arrive in one step with the metadata already right, and Mendeley and EndNote work the same way. Write down what you are actually trying to find out, in rough words. Your advisor reads what you brought and tells you one specific thing about it.
After that it becomes a workspace you return to instead of a tool you evaluate. Everything you make exports in standard formats, and it stays yours after the defence. A fair number of people discover at that point that they never stopped using it.
One honest limit: the free plan holds seventy-five sources. That's room for a proposal or a chapter's full literature, not the two-hundred-odd a finished dissertation carries. You will meet that ceiling, and a meter shows it coming. Nothing you have written gets held hostage when you do.
Bring one paper.
The workspace starts where you are, mid-proposal, mid-fieldwork or on day one. Free to begin, with no time limit and no card.
Start free: bring one paper