A preprint is a complete draft of a research paper shared publicly before it has been through journal peer review. Posting one takes an afternoon, costs nothing on the major servers, and makes your work readable, citable, and timestamped months or years before a journal version appears. For many researchers, especially those watching physics or biology from other fields, the whole practice can seem mysterious: is a preprint a real publication, will journals still take the paper, and where would you even post one?
This guide answers those questions in order. It covers what preprints are and are not, why researchers use them, which preprint servers serve which disciplines, how journal policies treat them, the genuine risks, and a simple workflow for posting your first one responsibly.
What a Preprint Is (and What It Is Not)
A preprint is the author’s version of a full manuscript, posted to a public server before peer review. It looks like a finished paper—complete arguments, methods, figures, references—because it is one; what it lacks is a journal’s stamp of evaluation. Most servers screen submissions for scope, plagiarism, and obvious problems, but screening is not review. Readers are expected to bring their own judgment, which is why preprints carry prominent labels noting their status.
Three neighboring terms are worth separating. A working paper is essentially the same object under a different name, common in economics and policy fields and often hosted by a department or institution rather than a dedicated server. A postprint is the peer-reviewed, accepted version of a paper before the publisher’s formatting. And the version of record is the final published article. The preprint sits at the start of that chain, and on most servers it stays permanently available even after the journal version appears, with a link connecting the two.
Why Researchers Post Preprints
The case for preprints is practical rather than ideological, and it holds across disciplines:
- Speed. Journal review commonly takes many months. A preprint makes findings available to colleagues within days of finishing the draft.
- A public timestamp. Posting establishes a dated, citable record of your ideas, which is the strongest low-effort protection against being scooped.
- Early feedback. Readers flag errors, missing citations, and unclear arguments while there is still time to fix them before formal review.
- Free open access. Anyone can read a preprint, including colleagues at institutions without subscriptions and practitioners outside academia.
- A citable object. Most servers assign a DOI or stable identifier, so the work can be cited in grant applications, job materials, and other papers while the journal version is still in review.
- Visibility for early-career researchers. Graduate students and postdocs can point hiring and fellowship committees to complete work rather than a line that says manuscript in preparation.
Norms still vary sharply by field. In physics and much of computer science, posting a preprint is the default first act of publication; in parts of the humanities and in clinical research it remains uncommon or sensitive. Before posting your first one, notice what respected people in your own subfield do, and when in doubt, ask a mentor how a preprint would be read in your discipline.
Major Preprint Servers by Discipline
Preprint culture started in physics and spread unevenly, so the right server depends on your field. These are the established options most researchers encounter:
| Server | Main fields | Worth knowing |
|---|---|---|
| arXiv | Physics, math, computer science, statistics | The original preprint server, running since 1991; posting is standard practice in these fields |
| bioRxiv | Life sciences | Operated by Cold Spring Harbor Laboratory; widely integrated with journal submission systems |
| medRxiv | Medicine and health sciences | Applies extra screening given clinical stakes; findings there are not practice guidance |
| SSRN | Social sciences, economics, law | Long-established home of the working paper culture; owned by Elsevier |
| ChemRxiv | Chemistry | Supported by major chemistry societies |
| PsyArXiv | Psychology | Community-run on Open Science Framework infrastructure |
| OSF Preprints | Multidisciplinary | Hosts many field-specific communities, including education and the humanities |
If your discipline lacks an obvious home, a multidisciplinary server or your university’s institutional repository serves the same purpose: a public, dated, freely readable copy of your manuscript.
Will Journals Still Publish My Paper?
This is the fear that stops most first-time posters, and for the large majority of venues the answer is reassuring: most major publishers explicitly allow submission of work that has appeared as a preprint, and many journals now ask for the preprint link during submission. The old worry traces back to the so-called Ingelfinger rule, a mid-twentieth-century policy against publishing previously reported findings; its strict form has faded, though a minority of journals, particularly in clinical medicine, still take a conservative line.
The practical rule is simple: check the specific journal’s preprint policy before you post, not after. Look on the journal’s website under submission guidelines, or search the Sherpa Romeo database, which summarizes publisher policies on preprints and self-archiving. While you are there, note two details: whether the journal restricts which license you may apply to the preprint, and whether it expects you to update the preprint record with a link to the published version. The broader landscape of who may read which version of your paper is its own topic—our explainer on gold, green, and diamond open access routes maps it, and preprints are the fastest lane of the green route.
The Real Risks (and How to Manage Them)
Preprints are not risk-free, and pretending otherwise does new researchers no favors. The risks are manageable once named:
- Errors go public. A mistake in a preprint is visible to everyone, including future search committees. Treat posting with the same care as submission: complete drafts only, checked by coauthors.
- Misreading by non-experts. Unreviewed health and policy findings can circulate as settled fact. Write your abstract so its limits are explicit, and label preliminary claims as preliminary.
- Version confusion. Readers may cite an early version after the paper has changed. Post updates to the same record rather than a new one, and add the journal link once it exists.
- A small set of incompatible venues. A few journals still decline preprinted work. If your dream journal is one of them, that is a reason to delay posting, not a reason to abandon the practice.
- License lock-in. The license you choose at posting, such as CC BY, is hard to walk back. Understand what it permits before you click. Our primer on who owns your conference paper and academic copyright basics explains the choices in plain terms.
A Simple Workflow for Your First Preprint
- Confirm the target journal’s policy. Five minutes on the journal site or Sherpa Romeo settles whether posting is safe for your publication plans.
- Get every coauthor’s agreement. Posting is a joint decision; surprised coauthors are a bigger professional risk than any server policy.
- Choose the server your field actually reads. Visibility comes from posting where colleagues already browse, not from the server with the nicest interface.
- Prepare a clean, complete PDF. Include all figures and references, spell out limitations, and add a line noting the manuscript has not yet been peer reviewed.
- Pick your license deliberately. More permissive licenses maximize reuse; more restrictive ones preserve options. Either is defensible when chosen on purpose.
- Post, share, and maintain. Announce it to your networks, respond to feedback, upload revised versions to the same record, and link the journal article when it appears.
Preprints, Conference Papers, and Proceedings
Conference work and preprints complement each other naturally. A conference presentation forces a full draft into existence by a fixed date; a preprint server gives that draft a public life afterward. Many researchers present first, absorb the audience’s questions, revise, and then post the improved manuscript—the same maturation path we trace in our post on how working papers grow into publications.
Publication venues attached to conferences add one more consideration. Papers accepted at NIVA’s virtual conferences, for example, can appear in the official conference proceedings, which are published with an ISBN, and authors developing work further can aim at NIVA’s journals. If you intend to route a manuscript through proceedings or a journal, apply the same habit you use with any venue: check the publication’s policy on earlier public versions before you post, and keep each version clearly labeled so readers always find their way to the most current one.
Frequently Asked Questions
A preprint is a complete manuscript posted to a public server before journal peer review. It is freely readable, usually receives a DOI or stable identifier, and is clearly labeled as not yet peer reviewed. Most servers keep it available permanently, linked to the journal version once one exists.
No. Preprint servers screen submissions for scope, plagiarism, and obvious problems, but they do not evaluate the research the way journal reviewers do. Readers should treat preprint findings as provisional, and authors should label preliminary results honestly in the abstract.
Usually yes. Most major publishers accept submissions that exist as preprints, and many journals ask for the preprint link during submission. A minority of journals, especially in clinical fields, remain restrictive, so check the specific journal policy or the Sherpa Romeo database before posting.
List them, but in their own clearly labeled section, separate from peer-reviewed work. Committees generally value preprints as evidence of active scholarship, especially for early-career researchers, and a preprint with a DOI can be cited in grant applications while the journal version is under review.
A conference deadline is still the most reliable way to turn a promising draft into a finished manuscript worth posting. Send NIVA a proposal for its next virtual conference—faculty present for $150, students for $70, membership is free, and your finished paper leaves the event ready for a preprint server or a journal.