Tag: PhD tools

Best AI Research Assistants for PhD Students in 2026: Discovery and Appraisal Tools Compared

Two distinct jobs hide inside “AI research assistant”, and buying the wrong one is how doctoral researchers end up paying a subscription they stop opening by December. Discovery tools help you find what to read, usually by walking the citation graph. Appraisal and extraction tools help you do something structured with papers you have already found. This comparison covers both, judged on what happens at 300 sources rather than at the onboarding demo. Table first.

Tool	Job	Index / scale (publisher’s own figure)	2026 price	Doctoral verdict
Semantic Scholar	Baseline search	“over 200 million academic papers”	Free (Allen Institute for AI)	The floor everyone should be standing on
ResearchRabbit	Citation-graph discovery	“over 310 million academic papers”; “1,000,000+ researchers”	No pricing page published	Best free discovery tool for exploratory reading
Litmaps	Discovery + monitoring	Not published as a paper count	Free tier (20 inputs, 2 maps, 100 articles/map); Pro $10/month, $120/year, academic email required	Buy it for the alerts, not the maps
Connected Papers	Single-seed graph visualisation	Not verifiable — see note	Not verifiable — see note	Useful for one job, not a subscription
Elicit	Extraction + screening	“more than 138 million papers”	Free tier; Plus $11/mo ($132/yr); Pro $39/mo ($468/yr); Scale $89/mo ($1,068/yr)	The one paid tool most systematic reviewers should consider
Scite	Citation-context appraisal	“1.6B+ citations”, “300M+ scholarly sources”	No free tier — 7-day trial; Basic $20/mo, Pro $50/mo, billed yearly	Field-dependent; buy only if contested claims are your problem

Prices were read from each vendor’s own pricing page in August 2026 and are shown at the annual rate where one exists. Two verification notes, because a comparison that hides them is not much use. Connected Papers’ site returns “we’re sorry but Connected Papers doesn’t work properly without JavaScript enabled” to a plain fetch, and its pricing page carries no readable content — so nothing about its current tiers is quoted here rather than guessed. Consensus blocked the same check outright. Treat every price you read anywhere, including here, as needing a look at the vendor’s page before you enter a card.

The shortlist, ranked for a doctorate

1. Semantic Scholar — the free baseline

Run by the Allen Institute for AI, Semantic Scholar describes itself as providing “free, AI-driven search and discovery tools, and open resources for the global research community”, indexing over 200 million papers from publisher partnerships, data providers and web crawls. It has no paywall, no tier and no upsell.

Who it suits: everyone, as a starting point. Where it falls short: it is a search engine rather than a workflow — no screening, no extraction tables, no project structure. Use it as the layer beneath whatever else you adopt, and note that its open API is why several tools further down this list exist at all.

2. ResearchRabbit — the best free discovery tool

ResearchRabbit works the way exploratory reading actually works: start from a paper you trust, expand outward through related works, authors and citations, and let collections build as you go. Its own figures are “over 310 million academic papers” and “1,000,000+ researchers worldwide”. Notably, its site publishes no pricing page at all.

Who it suits: anyone in the first eighteen months, or anyone entering an unfamiliar sub-literature. Where it falls short at doctoral scale: discovery without discipline sprawls. A graph will happily hand you four hundred papers, and nothing in the tool tells you which forty matter. Pair it with a hard inclusion rule you wrote down first.

An abstract citation network with clustered nodes representing related academic papers — Citation-graph tools are excellent at showing you a field’s shape and terrible at telling you where to stop. The stopping rule has to come from your protocol, not the visualisation.

3. Litmaps — discovery plus the thing nobody budgets for

Litmaps builds seeded maps of a literature and — the genuinely valuable part over a three-to-four-year candidature — monitors them, pushing alerts as new work appears. The free tier allows up to 20 inputs, 2 maps and 100 articles per map with monthly alerts; Pro is $10 a month or $120 a year with unlimited maps and daily or configurable alerts, and the education rate requires an academic email address.

Who it suits: candidates past the upgrade, whose literature is defined and now needs to stay current until submission. Where it falls short: the free tier’s 100-article cap is small for a doctoral map, so this is effectively a paid tool — but at $120 a year it is the cheapest insurance against the examiner question “were you aware of the 2027 paper on this?”.

4. Connected Papers — one job, done well

Give it a seed paper and it produces a similarity graph of the surrounding literature — a fast orientation to an unfamiliar area, and genuinely useful the week you take on a new chapter. It is not a workflow, does not manage a project, and its current commercial terms could not be verified for this comparison. Use it as an occasional instrument rather than something you subscribe to and forget.

5. Elicit — the extraction engine

Elicit is the tool that does something other than find papers. Its free Basic tier already offers unlimited search across “more than 138 million papers”, unlimited summaries, chat with full-text papers and Zotero import — which is more than most candidates will exhaust. Paid tiers buy structured work: Plus at $11 a month ($132 annually) adds exports to RIS, CSV, BIB, PDF and DOCX plus five extraction columns at a time; Pro at $39 a month ($468 annually) adds a dedicated systematic review workflow that can screen 5,000 papers, twenty columns, and extraction across up to 135 data sources; Scale runs to $89 a month and Enterprise adds screening at 40,000 papers.

Who it suits: anyone building a structured evidence table — a systematic or scoping review, or a methods-comparison chapter where you need the same eight fields from ninety papers. Where it falls short at doctoral scale: extraction accuracy is not verification. Every extracted cell you intend to cite must be checked against the paper, and the checking is not optional overhead — it is the review. Note also that the vendor reserves language about “PRISMA-grade” accuracy for its enterprise tier, which tells you something about how to treat the cheaper ones.

6. Scite — citation context, if that is your problem

Scite’s distinctive claim is that it classifies how a paper has been cited — supporting, contrasting, or merely mentioning — across “1.6B+ citations” drawn from “300M+ scholarly sources”, which lets you see whether a finding you are about to build on has actually been corroborated or quietly contradicted. Basic is $20 a month billed yearly with unlimited assistant use and collections to 1,000 papers; Pro is $50 a month with API access, larger collections and patent, clinical-trial and grant datasets. There is no free tier — only a 7-day trial — and student discounts are handled by referring your institution rather than by presenting a student card.

Who it suits: fields with replication problems, contested effects or a retraction history worth checking. Where it falls short: classification is automated and imperfect, coverage skews to well-indexed STEM literature, and $240 a year is a real fraction of a stipend. If your field’s problem is finding literature rather than adjudicating it, this is the wrong purchase.

What none of them do

Four boundaries worth naming before you attribute powers to any of these tools.

They do not read for you. A summary is a lossy compression of an argument, and the parts it loses — the caveat in the methods, the sample that is not what the abstract implies — are precisely the parts a doctorate is examined on. Anything you cite, you have opened.

They do not confer method. A tool that screens 5,000 records does not make your review systematic. A protocol, an inclusion rule fixed in advance, dual screening where your method requires it and transparent reporting of numbers do that; the software only makes the labour survivable.

They are not complete, and their incompleteness is uneven. Every index here is assembled from partnerships, feeds and crawls, and coverage is markedly better in indexed STEM literature than in humanities monographs, non-English scholarship, grey literature and policy documents. If your field lives in books, these tools are a supplement to your library’s catalogue, not a replacement for it.

They are not your library, your coding software, or your writing environment. References belong in a reference manager; qualitative coding belongs in NVivo, ATLAS.ti or Taguette; the conceptual notes these tools generate belong in your notes system; and none of them is where 80,000 words gets drafted.

A printed journal article densely annotated by hand beside a closed laptop — The stage no tool removes. Discovery software changes how many papers reach this desk; it does not change what has to happen to them here.

One more thing to check before you subscribe

Two questions that cost nothing and change the answer. First, what does your university already hold? Institutional subscriptions to discovery and appraisal platforms are common and badly advertised — your subject librarian knows, and ten minutes with them is the highest-return conversation in this whole comparison. Second, what does your ethics approval and data-management plan permit? Uploading a published paper is unproblematic; uploading your own unpublished chapter, participant data or a collaborator’s manuscript to a third-party service is a different decision, governed by commitments you have already signed.

The recommendation

Start with a free stack and make it prove insufficient before you spend anything: Semantic Scholar for search, ResearchRabbit for citation-graph discovery, Elicit’s free tier for summarising and chatting with papers you have found. That covers the majority of doctoral literature work at zero cost.

Then buy at most one paid tool, chosen by the specific job you cannot otherwise do. Structured extraction across dozens of papers for a systematic or scoping review — Elicit Plus or Pro. Keeping a defined literature current across a four-year candidature — Litmaps Pro at $120 a year. Adjudicating contested findings in a field with a replication problem — Scite. Buying two of these is usually a sign that the underlying problem is an unwritten inclusion rule, which no subscription fixes.

Where the bottleneck actually moves to

Solve discovery and the constraint relocates, quickly and predictably, to writing. Candidates who have found and read three hundred sources do not stall for want of a three-hundred-and-first; they stall converting the material into linear chapters — the failure mapped in our notes-to-chapter workflow, and the reason the introduction chapter is so often the last thing written and the worst thing written.

Tesify is built for that stage: chapter structure, section-by-section drafting and an automatic bibliography that formats your references as you cite them. The reading, the judgement and every sentence remain yours — 100% written by you — which is the only arrangement that survives a viva. It is free to start.

Frequently asked questions

What is the best AI research assistant for a PhD in 2026?

There is no single best, because discovery and extraction are different jobs. For most doctoral researchers the strongest starting configuration is free: Semantic Scholar for search, ResearchRabbit for citation-graph discovery, and Elicit’s free tier for working with papers you have already found.

Is Elicit free?

It has a substantial free tier — unlimited search across more than 138 million papers, unlimited summaries, chat with full-text papers and Zotero import. Paid plans start at $11 a month billed annually and exist chiefly to buy structured extraction and systematic-review screening.

Does Scite have a free plan?

No — only a 7-day trial. Individual plans are $20 and $50 a month billed yearly, and academic discounts run through recommending the tool to your institution rather than through a student rate at checkout.

Connected Papers or ResearchRabbit?

ResearchRabbit for ongoing discovery and collection-building across a candidature; Connected Papers for a fast one-off orientation to an unfamiliar area from a single seed paper. They answer different questions, and the free way to find out is to run both on a paper you know well and see which graph tells you something you did not already know.

Can these tools do my systematic review?

They can carry the labour — search, deduplication support, screening at volume, extraction into tables. They cannot supply the protocol, the inclusion criteria fixed in advance, the second screener your method may require, or the transparent reporting of how many records were excluded and why. The method is yours; the tool is a lever.

Will my examiners object to AI-assisted literature searching?

Searching and screening with software is ordinary research practice and has been since databases replaced card catalogues. What examiners test is whether you know the literature you cite — so the operative rule is that anything in your bibliography is something you have opened and read, regardless of how you found it. Follow your institution’s disclosure requirements for any generative component.

Do these tools hallucinate references?

The retrieval-based ones surface real indexed records rather than inventing them, which is the main argument for using them over a general chatbot for literature work. That is not a guarantee: summaries can still misstate what a paper found, and a citation you have not opened is a risk whatever produced it.

Are they any use in the humanities?

Less, and honestly so. Coverage is built on indexed journal literature, so monograph-based fields, non-English scholarship and archival work are poorly served. Your library catalogue, subject bibliographies and a good subject librarian remain the stronger route.

Should I pay for one of these on a stipend?

Only against a named job the free stack cannot do, and only after checking what your institution already licences. The comparison worth making is not the subscription against zero, but the subscription against the hours it genuinely removes from a specific piece of work.

What about general chatbots for literature review?

Useful for explaining an unfamiliar method or critiquing your own writing; unsuitable for finding sources, because a general model without retrieval will produce plausible references that do not exist. Fabricated citations are found in seconds at doctoral level and are catastrophic when they are.

August 22, 2026

Zotero vs Mendeley vs EndNote for a PhD Thesis: Which One Still Has Your References in Year Four?

Every guide to doctoral tooling tells you to set up a reference manager on day one, and then declines to say which. That includes our own comparison of LaTeX and Word at thesis scale, which twice insists the reference manager is the decision that matters more than the editor without ever naming one. This page settles it.

The choice is not really about features. All three of these will insert a citation, generate a bibliography and change style on demand, and any of them handles a taught-degree essay identically well. What separates them appears only at doctoral duration and doctoral volume: three hundred sources, several gigabytes of PDFs, a supervisor who wants access, and a four-to-eight-year horizon over which your institutional affiliation is not guaranteed to be continuous.

So the operative question is not “which has the best Word plugin”. It is which of these still has your library in year four.

The comparison, at doctoral scale

	Zotero	Mendeley	EndNote
Owner	Digital Scholar (non-profit)	Elsevier	Clarivate
Licence model	Open source, GNU AGPL v3	Proprietary, free account	Proprietary, paid version
Free storage	300 MB	2 GB	Institutional licence dependent
Paid storage / price (2026)	2 GB $20/yr · 6 GB $60/yr · unlimited $120/yr	Tiers not quoted here — see note	One-time purchase, not a storage subscription
Software cost (2026)	Free	Free	Full €310 · Student €170 · Upgrade €140, one-time, incl. tax
Usual doctoral route	Install it yourself	Install it yourself	Institutional site licence
LaTeX / BibTeX	Native and well supported	Supported	Supported
Risk at year four	Storage cost	Vendor dependence	Licence expiry with your registration

Figures were read from each vendor’s own pages in August 2026. Two verification notes, because a comparison that hides them is worth less than one that admits them. Mendeley’s paid storage tiers are not quoted here because they were not stated on the pages checked; the free allowance is advertised plainly as “Claim your 2GB of storage — it’s free” and anything beyond that is a question for Elsevier rather than a number to guess. And EndNote’s pricing is quoted in euros because that is the currency its own purchase page returned; check the figure in your own currency before you buy, and check your institution first, because you may not need to buy at all.

The storage arithmetic nobody does in advance

Here is the single most consequential number in this article. Zotero’s free tier is 300 MB. Paid tiers run 2 GB at $20 a year, 6 GB at $60 a year, and unlimited at $120 a year.

Now do the doctoral arithmetic. A journal article PDF is commonly one to three megabytes; scanned material, supplementary files and book chapters are considerably larger. A candidate who reaches submission with three hundred stored PDFs is therefore looking at somewhere between roughly 300 MB and well over a gigabyte of attachments, and that is before the archival scans, the datasets and the drafts of your own chapters that people habitually park in the same library.

The practical consequence: on Zotero, a PDF-storing doctoral library will outgrow the free tier, usually in the second year. That is not an argument against Zotero. It is an argument for knowing that Zotero costs roughly $20 to $120 a year in practice rather than nothing, and for budgeting it deliberately instead of discovering it as a stalled sync three days before a chapter deadline.

Two ways out, both legitimate. Use Zotero’s linked-file mode, keeping the PDFs in a folder you sync yourself through whatever cloud storage you already have, so Zotero stores only the metadata and the free tier lasts indefinitely. Or pay the $20 tier and stop thinking about it. What does not work is storing everything on the free tier and hoping.

Mendeley’s 2 GB free allowance is the most generous free storage of the three and will carry a great many doctoral libraries to submission without payment. That is a real and underrated advantage, and it is the strongest argument in Mendeley’s favour.

A sync that has stopped because the reference library has outgrown its free storage allowance — The failure is never dramatic. The sync just stops, usually in year two, usually the week something is due.

The licence horizon: the EndNote question

EndNote is different in kind from the other two. It is a one-time software purchase rather than a free account with paid storage: Clarivate’s own purchase page lists a full licence at €310, a student licence at €170 with verification, and an upgrade from EndNote 21 or earlier at €140, all one-time and inclusive of tax, with a 30-day free trial.

But almost no UK doctoral researcher pays that, because EndNote is very commonly provided through an institutional site licence. Which is exactly where the risk lives, and it is a risk specific to doctoral study rather than to taught degrees.

A taught student’s relationship with their institution ends cleanly at graduation, by which point the dissertation is submitted and the library is disposable. A doctoral researcher’s does not. You may spend a continuation or writing-up period in an altered enrolment status — the University of York, for instance, states that candidates in the continuation period are “not formally enrolled at the University”, a position set out with the fee implications in our guide to how long a UK PhD actually takes. You may be rebuilding a thesis for resubmission across a year in which your access has changed. You will almost certainly still be citing the same library while publishing papers from the thesis after the award.

So the question to put to your library or IT service, in writing, before you commit four years of references to EndNote is a narrow one: does our EndNote site licence cover me during continuation or writing-up status, and for how long after my award? The answer is institution-specific, we are not going to guess it for you, and it is a two-minute email that can save you a €310 decision made under duress. Ask the same question about any other licensed software your thesis depends on while you are at it.

The corresponding advantage of Zotero is structural rather than a feature: it is released under the GNU Affero General Public Licence version 3 by a non-profit, so nothing about your access is contingent on your continuing registration at any institution. For a project measured in years rather than terms, that is worth more than most of the feature differences.

Group libraries, and who actually pays for them

Shared libraries are genuinely useful in doctoral work — a group library holding the corpus for a systematic review, or one shared with a supervisor so they can see what you are reading, is a real improvement on emailing PDFs. There is a detail here that surprises people, and Zotero states it plainly: “Group file storage always draws from the storage account of the group owner.”

Read the consequence. If you create the group, every file every collaborator adds is charged against your quota, and a 300 MB free tier will evaporate. If your supervisor creates it, it is charged against theirs. This is not a trick; it is simply a fact worth knowing before you set up a shared corpus and then wonder why your personal library stopped syncing. Where a group library is central to your project, agree who owns it and who is paying for the storage as part of the ordinary business of managing your supervisory team, rather than discovering it later.

The migration you should do now or never

All three export to RIS, and all three read it. Migration is therefore possible at any point, and it is also the single most reliably regretted job in doctoral tooling, because what survives an export is not what you care about.

Bibliographic metadata migrates cleanly. What migrates badly, or not at all, is the accumulated layer that actually represents your reading: PDF annotations and highlights, notes attached to items, tags and colour coding, folder or collection structure, and the “date added” ordering that lets you reconstruct what you were thinking in your second year. Those are stored differently by each application, and an RIS file is not designed to carry them.

Two rules follow. Choose in your first term and then stop revisiting the question — the same displacement-activity pattern that makes candidates re-litigate LaTeX versus Word every few months applies here, and it is most tempting precisely when the writing is going badly. And if you are going to move, move early, while the library is fifty items rather than four hundred and while the annotation layer you would lose is thin. Migrating a reference library during your writing-up year is a way to introduce citation errors under time pressure for no scholarly gain.

Handing back a university card, the moment an institutional software licence ends — Institutional software is borrowed against your registration. A doctorate is one of the few degrees where the registration can end before the work does.

The recommendation

Default to Zotero, and budget $20 a year for storage from the start. It is free software from a non-profit under an open licence, its data is yours in an open format, its BibTeX support is the best of the three if you are in a LaTeX discipline, and nothing about it lapses when your registration does. The 300 MB free tier is its one genuine weakness at doctoral scale and it costs twenty dollars to solve.

Choose Mendeley if the free storage is decisive for you and you are comfortable with an Elsevier account holding your library. Two gigabytes free is the most generous allowance here and it may well carry you to submission at no cost.

Choose EndNote if your department’s workflow is built on it — shared team libraries, an established local template, a supervisor who will exchange libraries with you — and you have confirmed in writing how long your institutional licence covers you. Local convention is a real and legitimate reason, and swimming against it costs more than it looks like it costs.

What matters far more than which of the three you pick is that you pick one in your first term and enter every source as you read it. The most expensive tooling mistake available to a doctoral researcher is not choosing the wrong manager; it is arriving at 250 sources with no bibliographic database at all and rebuilding it by hand in the month you can least afford.

What a reference manager will not do

It stores and formats what you have already found and decided to keep. It does not find the literature — that is the job of the discovery and appraisal tools compared in our roundup of AI research assistants at doctoral scale. It does not code your qualitative data, which belongs in NVivo, ATLAS.ti or Taguette. It does not hold your thinking, which belongs in your notes system. And it does not tell you whether a source you have stored is one you have actually read, which remains the only standard your examiners will apply to your bibliography.

It also does not write the chapters. When the library is in order and the drafting is what has stalled, Tesify holds the chapter structure and formats your bibliography as you cite, so the references land correctly in a document that stays internally consistent across 80,000 words. Every sentence stays 100% written by you, which is the only arrangement that survives a viva. It is free to start.

Frequently asked questions

Which reference manager is best for a PhD?

Zotero for most candidates: free, open source under the AGPL v3, strong BibTeX support, and not tied to your institutional registration. Budget for storage, because the free tier is 300 MB and a PDF-storing doctoral library will exceed that.

Is Zotero really free?

The software is free and open source. Cloud storage is not beyond 300 MB: 2 GB is $20 a year, 6 GB is $60, and unlimited is $120. You can avoid the cost entirely by storing PDFs as linked files in your own cloud folder and letting Zotero sync only the metadata.

How much storage does a doctoral reference library need?

More than 300 MB if you store PDFs. At one to three megabytes per article, three hundred stored papers alone can approach or exceed a gigabyte, before scans, datasets or your own drafts. Plan for a gigabyte or two rather than hoping.

How much free storage does Mendeley give you?

Two gigabytes, advertised on Elsevier’s own page as “Claim your 2GB of storage — it’s free”. That is the most generous free allowance of the three and is enough for a great many doctoral libraries.

How much does EndNote cost?

Clarivate lists a full licence at €310, a student licence at €170 with verification, and an upgrade from EndNote 21 or earlier at €140 — all one-time purchases including tax, with a 30-day trial. Check your institution before buying, since site licences are common.

Will I lose EndNote when I finish my PhD?

If you are using an institutional site licence, possibly, and the timing is the problem: doctoral researchers often need the library during continuation status, during a resubmission period, and while publishing from the thesis after the award. Ask your library in writing what your licence covers and for how long, rather than assuming either way.

Can I move my library from one manager to another?

Yes, via RIS export, which all three support. The metadata transfers reliably; annotations, notes, tags and collection structure often do not. Move early if you are going to move, and never during your writing-up year.

Whose storage is used by a shared group library?

In Zotero, the owner’s: “Group file storage always draws from the storage account of the group owner.” Decide who creates the group before you build a shared corpus, because that decision determines whose quota pays for it.

Which works best with LaTeX?

Zotero, whose BibTeX export is well supported and widely used with Overleaf workflows. All three can produce BibTeX, but if you are writing in LaTeX this is a real point of difference rather than a marginal one.

Does it matter which one my supervisor uses?

Only if you intend to share a library or exchange files directly, in which case matching them removes real friction. For ordinary supervision it does not matter at all, because what they receive is a document with formatted citations, not your database.

Should I switch reference managers during corrections?

No. Corrections run against a deadline and re-generating a bibliography in a new application is a reliable way to introduce citation errors at exactly the wrong moment. Make the corrections where the thesis already lives.

Do I still need a reference manager if I only have 80 sources?

Yes, and the reason is not the count but the changes. A doctoral bibliography gets reordered, restyled and partially rewritten repeatedly across the final year, and every one of those operations is instant in a manager and a manual reformatting job without one.

August 22, 2026

Best Transcription Tools for PhD Research Interviews (2026): Accuracy, Ethics and Cost Compared

Transcription sits at an awkward junction: it is the most automatable drudgery in qualitative research, and it involves shipping your participants’ voices — personal data by definition — to whatever service you picked at midnight. So this comparison runs compliance first, capability second: the fastest transcriber in the world is the wrong choice if your ethics approval and data-management plan never named it.

	Local AI (Whisper-class)	Otter.ai	Your university’s Microsoft 365 tenant	Human transcription services	Doing it yourself
Where audio goes	Nowhere — processed on your machine	Otter’s cloud	Institutional cloud, inside existing agreements	The service’s staff and systems	Nowhere
Cost	Free (open-source)	Free tier 300 min/month; Pro $8.33/month annual (1,200 min); 20% student discount via .edu email	Usually included	Per audio hour; the premium option	Your time: ~4–8 hours per audio hour
Accuracy on clear audio	Strong	Strong	Good	Best, especially accents and crosstalk	Perfect, eventually
Accuracy on messy audio (cafés, dialect, jargon)	Degrades; better with larger models	Degrades	Degrades	Holds up best	Holds up, slowly
Ethics-form friendliness	Highest — no third-party processing	Requires naming a third-party processor	High — often pre-approved infrastructure	Needs confidentiality agreement	Highest
Best for	Most doctoral interview studies	Meetings and low-sensitivity recordings	Institutionally cautious projects	Difficult audio, funded projects	Small N, analytic immersion

Start with the rule, not the tool

Interview recordings are personal data — a voice is identifiable even before the content is — and your handling of them is governed by what your ethics application, participant information sheet and data-management plan actually said. Universities increasingly publish lists of approved transcription routes or require a data-protection assessment before audio leaves institutional systems, and sending recordings to an unapproved consumer cloud service can put you in breach of commitments you signed, whatever the tool’s own privacy page says. The sequence that keeps you safe: name your transcription route in the ethics application; describe it in the information sheet (“recordings will be transcribed using…”); and if your plans change mid-project, amend the approval rather than improvising. If you are before that stage now, write the transcription paragraph today — it is ten minutes that removes the whole class of problem, and the same logic applies to every tool that touches participant data.

The shortlist, ranked for a doctorate

1. Local AI transcription — the new default

Open-source speech models of the Whisper family changed this decision: strong automatic transcription that runs on your own computer, so the audio never leaves your possession. That single property collapses most of the compliance analysis — there is no third-party processor to name, justify or trust — and the price is zero. Costs: you need a reasonably capable machine, a small amount of setup (your university’s research computing team or a colleague has almost certainly done it already), and accuracy still degrades on poor recordings. For a typical doctoral interview study — sensitive-ish data, dozens of hours, no transcription budget — this is the recommendation.

2. Your university’s Microsoft 365 transcription — the institutional route

If your university runs Microsoft 365, transcription inside Word or Teams processes audio within the institutional tenant — infrastructure your university has already contracted for, which is why data-protection teams often point students here first. Accuracy is serviceable rather than stellar, and speaker separation is basic, but “the recording never left university systems” is a sentence that makes ethics reviewers relax. Check your institution’s own guidance for what its licence covers.

3. Otter.ai — capable, for the right recordings

Otter is polished and fast, with live transcription, speaker labelling and a workable free tier (300 minutes a month; Pro at $8.33 a month billed annually raises it to 1,200 minutes, with a 20 per cent student discount via a .edu address). The catch is precisely its cloud nature: for research interviews it is a third-party processor of participants’ personal data, which your approval must cover. Where it fits: low-sensitivity recordings, your own research memos, supervision meetings — and interview studies whose approval explicitly names it.

4. Human transcription services — buy them for the hard cases

Professional transcribers still beat every machine on strong accents, overlapping speech, poor recordings and specialist vocabulary, and offer the choice of intelligent verbatim versus full verbatim done judgementally rather than mechanically. They cost real money per audio hour and add a confidentiality step (a signed agreement, a reputable service, an approval that mentions outsourcing). Rational uses: a funded project, a handful of unusable-by-machine recordings, or Deaf/accessibility workflows.

Audio recorder and consent forms after a research interview — The consent form and the transcription route are one decision: participants agreed to a specific handling of their voices.

Recording well is half the transcription problem

Every tool in the table performs a tier better on good audio, and good audio is mostly free. Use a dedicated recorder or a decent phone app rather than a laptop microphone across the table; put the device nearer the participant than yourself, since their words matter more than your questions; choose the quiet room over the atmospheric café whenever the participant allows; record thirty test seconds and listen back before the interview proper; and for remote interviews, use the platform’s own recording rather than re-recording speaker audio through the air. One more habit that saves projects rather than minutes: duplicate the file to institutional storage before you leave the building or close the call. A machine transcript of a clean recording plus a light correction pass beats a human transcript of a bad one — and the recording is the only stage you can never redo.

The step everyone skips: correction is analysis

Whatever produces your first draft transcript, the checking pass — you, headphones, audio against text — is not optional overhead. Automatic transcripts fail exactly where interviews are most interesting: jargon, names, emotional speech, overlaps. And the correction hours are quietly the first analysis pass; researchers routinely report their best early codes emerging while fixing transcripts. Budget roughly one to two hours per audio hour for correction of machine output — against four to eight for typing from scratch — and log your conventions (how you marked pauses, laughter, redactions) because your methods chapter will need them. From there the transcripts flow into coding — NVivo, ATLAS.ti or Taguette — while the conceptual notes they generate belong in your notes system, not a folder of stray documents.

The recommendation

Run local Whisper-class transcription as your default; use your university’s Microsoft 365 transcription where institutional processing is the path of least resistance; pay humans for the recordings machines cannot handle; and use consumer cloud tools only where your ethics approval names them. In every case, the tool appears in your ethics paperwork before it appears in your workflow.

And when the transcripts are coded and the findings chapter looms, the bottleneck moves back to the writing itself. Tesify structures and drafts the thesis with you — chapters, methods documentation and bibliography in one workspace, 100% written by you — so the months you saved on transcription arrive intact at the examination.

Frequently asked questions

What is the best free transcription tool for research interviews?

Local Whisper-class transcription: free, strong on clear audio, and — because it runs on your own machine — the easiest route to justify in an ethics application. Your university’s included Microsoft 365 transcription is the runner-up.

Can I use Otter.ai for PhD interviews?

Only if your ethics approval and participant information cover a third-party cloud processor — name it, justify it, and check whether your university restricts it. For meetings and non-participant audio it needs no such ceremony.

Is it a GDPR problem to upload interviews to a transcription site?

It is a data-processing decision that must match what participants consented to and what your university permits. Voices are personal data; an unapproved upload can breach your own signed commitments even where the service itself is reputable.

How long does transcription actually take?

Typing from scratch: commonly four to eight hours per audio hour. Machine-first with human correction: the machine minutes plus one to two hours of checking per audio hour. The correction time is real work — and doubles as first-pass analysis.

Do I have to transcribe every interview in full?

Methodologically, it depends on your analysis: some approaches require full verbatim transcripts; others defensibly work from full transcripts of core interviews plus indexed partial transcripts elsewhere. Whatever you choose, state and justify it in the methods chapter.

Should I use intelligent verbatim or full verbatim?

Full verbatim (every um, repair and overlap) where the interaction itself is analysed, as in conversation-analytic work; intelligent verbatim (cleaned for readability, meaning preserved) for most thematic work. The decision belongs to your method, not your transcriber.

How should I store recordings and transcripts?

Exactly as your data-management plan says: institutional storage, not personal devices; recordings and identity keys separated from transcripts; pseudonymisation applied at transcription time; deletion on the schedule you promised participants.

Can AI transcription handle strong accents and dialects?

Less well than clear standard speech — error rates rise, and rise most on precisely the participants whose voices are least represented in training data. Pilot your tool on your hardest expected audio before committing, and budget human help for what fails.

Do I need participants’ consent to use AI transcription?

You need consent that covers your actual processing. The clean practice is to describe the transcription route in the participant information sheet; a vague “recordings will be transcribed” plus a later cloud upload is where problems start.

What should the methods chapter say about transcription?

The route (tool or service, and where processing happened), the verbatim convention, the correction process, anonymisation practice, and the storage arrangements — three or four sentences that jointly demonstrate the data was handled as approved.

August 20, 2026

Obsidian vs Notion for PhD Thesis Notes: Which Survives Four Years? (2026)

Choosing a notes system for a doctorate is not like choosing one for a job you might leave next year. Whatever you pick will hold several thousand notes by submission, must still open instantly in year four, and must give everything back cleanly when the thesis is done. Judge both tools against that horizon, not the onboarding demo. Comparison first, verdict after.

	Obsidian	Notion
Data model	Plain Markdown files on your own disk	Cloud workspace of pages and databases
Price	App free, including commercial use; Sync $4/user/month (annual) is optional — free sync workarounds exist	Free plan; Plus €9.50/member/month — and the Plus plan (1-member) is free for students via a school email
Offline access	Total — it is local files	Limited; built cloud-first
Linking and graph	Core strength: wiki-links, backlinks, graph view	Links exist; databases with properties are the real organising tool
Structured data (reading logs, participant trackers)	Via plugins and properties; less natural	Core strength: databases, views, filters
Longevity of your notes	Markdown readable by any future tool, no export step needed	Export exists (Markdown/CSV) but structure degrades on the way out
Extensibility	Large community plugin ecosystem	Integrations and API; AI features on paid tiers
Failure mode at scale	Plugin sprawl and vault clutter	Sluggish giant databases; structure becomes its own project

The two tools disagree about what a note is

Obsidian treats notes as documents you own: a folder of Markdown files on your machine, linked into a network. Nothing leaves your disk unless you choose sync; the app is free including commercial use, with optional paid services (Sync at $4 a month billed annually, Publish, a voluntary $50/year commercial licence) funding development. For the Zettelkasten-style working pattern — hundreds of small linked notes whose connections surface as arguments — it is the natural home, and the graph of backlinks becomes a map of your literature. The deeper doctoral argument for it is brutal and simple: in year four, your notes are text files. Any tool, any decade, can read them. The export problem cannot happen because there is nothing to export.

Notion treats notes as records in databases: pages with properties, filtered into views. For the structured side of a doctorate it is genuinely better — a reading log with status, method and relevance fields; a participant tracker; a supervision-meeting database that generates your agenda. Students get the Plus plan free (single-member) with a school email, which removes price from the argument. The costs are the mirror image: it is cloud-first, so patchy connectivity on fieldwork is a real constraint; big databases slow down; and while export to Markdown and CSV exists, the relational structure you spent years building does not survive the trip out cleanly. Notion’s AI features sit on limited trials outside the paid tiers — pleasant, but nothing a doctorate depends on.

Index cards linked with thread mapping thesis ideas — The linking pattern is the Obsidian case; the tracking pattern is the Notion case. A doctorate contains both.

The three-year test

Before committing, ask each candidate tool the questions that only matter later. Can you find every note touching one concept in under a minute, once there are three thousand of them? Backlinks and saved database views both pass; folder hierarchies alone do not. Does it work on the train, in the archive basement, at the fieldwork site — or does it quietly assume connectivity? What is the realistic total cost across four years, including the sync or plan you will actually need rather than the tier on the comparison page? If the company changed its pricing or shut down in year three, what exactly would you lose, and how fast could you leave? And — least asked, most predictive — does its working grain match yours, because a tool you resent opening is a tool you will stop opening by Easter of year one. Obsidian and Notion give opposite answers to several of these; that, not feature counts, is the real comparison.

The recommendation

For the thinking layer of a PhD — literature notes, concept development, argument building — use Obsidian. Local files, free without conditions, and structurally incapable of trapping four years of notes. Use Notion where its databases genuinely earn their keep — project tracking, reading logs, supervision admin — on the free student Plus plan. Many researchers run exactly this split happily; if you refuse to run two tools, pick the one matching your dominant work pattern: linking and prose → Obsidian; tracking and structure → Notion.

Two honest caveats. Obsidian’s flexibility is a procrastination hazard — plugin-tinkering and vault-gardening feel like work and are not; adopt a minimal setup and freeze it for a term. And Notion’s databases invite the same failure at a different altitude: a perfect tracking system for reading you are not doing. The tool is the container; the doctorate is the reading and writing.

What neither tool does

Three gaps worth naming before you commit either tool to jobs it cannot do. Neither is a reference manager — Zotero (or your lab’s equivalent) still owns citations and PDFs, and both notes tools link out to it well enough. Neither is a QDA package — if your method needs systematic coding of transcripts, that is NVivo/ATLAS.ti territory, not a notes app. And neither is where the thesis gets written: at 80,000 words with hundreds of references, drafting lives in a proper long-form environment — the trade-offs are in our LaTeX vs Word comparison — or in a workspace built for thesis drafting.

That last hand-off is the one that matters most and gets planned least. Notes become a thesis through a deliberate pipeline: distilled claims move from the vault into a chapter outline, sources attach to claims, and drafting proceeds against that outline rather than against three thousand raw notes. Tesify is built for that stage — chapter structure, drafting and an automatic bibliography in one place, 100% written by you — and the full notes-to-chapter method is in our writing-up workflow guide.

Setting up either tool for a doctorate in one evening

For Obsidian: one vault for the whole doctorate; folders kept to a handful (literature, concepts, methods, admin); one note per source and one per concept; links made at write-time, not in gardening sessions; the vault folder inside your backed-up university storage; and no more than five plugins until the habit is stable. For Notion: one workspace; a source database with the fields you will actually filter by (status, method, chapter relevance); a supervision database whose entries double as agendas; and a weekly view you visit on a schedule. In both: write the first fifty literature notes before customising anything further — real content teaches you what structure you need better than any template gallery.

Frequently asked questions

Is Obsidian really free for a funded PhD researcher?

Yes — the app is free without limits for personal and commercial use alike; the commercial licence is optional support, and the paid products (Sync, Publish) are add-on services, not unlocks. Syncing through your university cloud storage instead of paid Sync is common and free.

Is Notion free for PhD students?

Notion’s education offer gives students the Plus plan free as a single-member workspace via an eligible school email — and the eligibility list extends well beyond .edu domains. AI features remain limited-trial outside paid tiers.

Which is better for a Zettelkasten?

Obsidian, without much contest — atomic linked notes, backlinks and the local graph are its native model. Notion can imitate the pattern with relations, but you are working against its grain.

Can I use both without the setup eating my week?

Yes, and the stable split is by function: Obsidian for thinking (literature and concept notes), Notion for tracking (reading log, supervision admin, project boards). The one rule that keeps it sane: each kind of content lives in exactly one place.

What happens to my notes when the PhD ends?

Obsidian: nothing — they are Markdown files you already possess. Notion: export the workspace (Markdown/CSV) and expect the database structure to flatten. Whichever you choose, rehearse the exit once in year one so the end of the doctorate holds no surprises.

Which handles PDFs and annotations better?

Neither should be your PDF library — that is the reference manager’s job. Keep PDFs and highlights in Zotero and write the intellectual note (claim, evidence, relevance to your argument) in your notes tool, linked by citekey.

Is Obsidian’s graph view actually useful or just pretty?

Mostly pretty at whole-vault scale; genuinely useful filtered to a cluster — the local graph around one concept surfaces connections and orphaned notes you had forgotten. Treat it as an occasional audit tool, not a workspace.

How do these tools handle collaboration with supervisors?

Notion is far better for shared artefacts — a meeting database or progress board your supervisors can see. Obsidian is fundamentally single-player; researchers share exports, not vaults. Most supervision workflows need only a shared document anyway.

What about Roam, Logseq and the other networked-notes tools?

The criteria transfer directly: who holds the files, what exports, what it costs across four years, and which working pattern it serves. Obsidian and Notion sit at the two poles — local-linked versus cloud-structured — which is exactly why this comparison generalises.

Do I need any of this to finish a PhD?

No — doctorates were finished with index cards, and some still are. A notes system pays off in retrieval: the moment in year three when you need every note touching one concept, in seconds. If your current system already does that, keep it and spend the energy on writing.

August 18, 2026