Research method: from sources to outline
The most dangerous moment in a thesis isn't the final blank page: it's the research phase that stretches on with no end, where every article you read cites three more, and the feeling of "not knowing enough yet to start writing" grows rather than shrinks. Research isn't a phase that precedes writing: the two need to be interwoven from the start.
The research question before the bibliography
Before piling up sources, you need a question — even provisional, even imperfect — that the thesis will try to answer. Without it, every text seems relevant, because there's no criterion for deciding what is and what isn't.
The question sharpens as you read, but it has to exist in rough form from day one: "how does X affect Y in context Z" is already enough to orient your first reading and discard what, however interesting, doesn't serve this work.
Read by topic, not by source
The classic mistake in note-taking is organising notes by source: one summary per article, in reading order. When the time comes to write, you're left with twenty summaries and no structure.
The more useful method is to take notes by topic: create from the start the thematic headings that will become the thesis's sections — "definitions of the concept," "positions in favour," "criticisms," "case studies" — and place every quote and every idea under the right heading, with the precise reference to the source. By the end of the reading, the outline is already largely built.
Distinguish primary and secondary sources from the start
Every discipline has its distinction between primary sources (the data, the original texts, the works being discussed) and secondary ones (those who wrote about them afterward). Knowing what weight to give each category keeps you from building a whole argument on a second-hand reading when the direct source was available.
Noting, next to each reference, whether it's primary or secondary and how authoritative it is in your discipline, is a few seconds of work that pays off in the writing phase, when you have to decide what to cite to support a point.
The research that kills the writing
There's a point where continuing to read no longer adds clarity but subtracts it: you accumulate contradictory positions without being able to take a direction. The sign is that new readings confirm things you already know, or add nuances that don't change the thesis's structure.
At that point it's worth stopping and writing a first draft of a chapter, even an imperfect one: writing reveals the real gaps in the research — the ones to fill with targeted reading — far better than continuing to read in general does.
Keeping notes, citations and chapters together
The practical problem of research is scattering: notes in one file, quotes in another, chapter drafts in a third, and the exact reference to a source that vanishes just when you need it for the bibliography. Keeping thematic notes next to the chapter structure, with source references linked to the point where you'll use them, cuts the time lost reconstructing where you read a certain thing. In NovLore, the work's chapter-and-section structure lives in the same place as the text, and the AI assistant stays optional: you can use it to rephrase a passage or summarise a source, or not use it at all and keep it only as a working structure.
Frequently asked questions
How much should I read before I start writing the thesis?
Less than the research phase tends to make you believe. A practical rule: when you can write in one sentence the question the thesis answers, and list the three or four main positions in the literature on the topic, you have enough to sketch the outline and start a chapter. The rest of the reading serves to fill specific gaps that surface as you write.
How do I organise research notes?
By topic, not by source. Create the thematic headings that will become the work's sections and place every quote and every idea under the relevant heading, with the full bibliographic reference. That way, when you start writing, the structure is already largely ready.
Is the bibliography built at the end?
No, it's built as you go. Recording each source in full at the moment you read it, instead of putting it off, avoids the final hunt for missing references, which is one of the most common causes of stress in the last weeks before submission.