Should I post my paper as a preprint?

The case for posting is mostly about time. Peer review takes months, and during those months your finding does not exist as far as anyone else is concerned. A preprint makes it citable immediately, timestamps your priority if someone is working on the same question, and lets people who would use the result actually use it.

The speed is literal. bioRxiv, PeerJ Preprints, and the quantitative biology section of arXiv normally post a submission publicly within 24 hours, and the timestamp cannot be edited afterwards (Bourne et al., 2017, p. 2). The scale is not marginal either. arXiv started in 1991 for physics and had passed a million preprints by 2017, covering mathematics, computer science, and quantitative biology as well (Bourne et al., 2017, p. 1). A concrete case of what fast posting buys you: when a May 2016 claim that cell-phone radiation raised cancer rates in animals started circulating, the NIH released its data and internal reviews as a preprint within a day, and that preprint was downloaded 148,000 times in a little over five months (Bourne et al., 2017, pp. 3-4).

For doctoral candidates there is a second benefit that matters more than it should have to. Job and funding applications are read against a publication list, and a preprint is something concrete to point at while a paper sits in review. Several funders now accept them explicitly.

The funders named as accepting preprints in applications or reports include the NIH, the Wellcome Trust, and the Simons Foundation, with the Human Frontier Science Program, the UK Medical Research Council, the Helmsley Charitable Trust, and the Canadian Institutes of Health Research allowing or encouraging them (Bourne et al., 2017, p. 4). Treatment across funders is variable, so check the specific call rather than assuming. The same source notes that preprints can add work for reviewers even while giving them a more informed picture of your output (Bourne et al., 2017, p. 3).

The reasons not to post are specific rather than general. A target journal that treats preprints as prior publication, which still exists in a few fields. A pending patent, where public disclosure can affect the application. Findings with public health or safety implications that should be reviewed before they circulate. Data-sharing agreements that restrict what can be released and when.

There is also a quality consideration worth taking seriously. A preprint is permanent and public with your name on it. Post work you would be willing to defend, not a draft you intend to fix after reviewers point at it.

How much you lose by skipping review first is genuinely contested. One argument is that peer review catches errors and improves papers, and that the public exposure of a preprint already discourages people from posting weak work (Bourne et al., 2017, p. 3). Against that, studies comparing arXiv and bioRxiv preprints with their final published versions found few substantive changes and similar reporting quality, though the authors are careful to say the value peer review adds is still hard to measure at scale (Tennant and Ross-Hellauer, 2020, p. 6). Neither position tells you to skip labelling. Mark the manuscript as a preprint that has not been peer reviewed, ask people who know the area to read it, and update the record when the journal version exists.

Will a journal still accept it afterwards?

Almost all will, and a shrinking minority will not. Most publishers now explicitly permit preprint posting and many encourage it. Check the specific journal's policy before posting, since a small number of journals in some fields still treat a preprint as prior publication.

Publisher policy databases list preprint permissions per journal, and the journal's own author instructions are the definitive source. Check the version too, since some journals permit the submitted manuscript but not the accepted one.

The Sherpa/Romeo database showed very few journals treating a preprint as prior publication, and life-science journals were moving the other way, adding preprint-friendly policies and direct submission routes from bioRxiv (Bourne et al., 2017, p. 4). Screening on the server side is light. bioRxiv's check is described as cursory, and it declines preprints that could damage human health, so the judgment about sensitive material is yours before you upload (Bourne et al., 2017, p. 5). Norms also differ by community, and services built to add review on top of preprints have had uneven uptake, including Academic Karma, which shut down in April 2019 (Tennant and Ross-Hellauer, 2020, p. 9).

If the journal requires it, disclose the preprint at submission and give the identifier. Concealing one is a much larger problem than posting one.

Do preprints count as publications?

Not as peer-reviewed publications, and they do count for some purposes. Many funders accept preprints in grant applications and progress reports, and they are citable with a permanent identifier. On a CV, list them clearly as preprints rather than mixing them with reviewed articles.

Field norms vary. In some areas a preprint carries substantial weight and is read widely before the reviewed version appears. In others it is barely noticed, and its main value is the timestamp.

Update the record when the paper is published. Most servers let you link the published version, and doing so stops the preprint circulating as the current text after it has been superseded.

One practical note for co-authored work: get everyone's agreement before posting. A preprint is a public release, and a co-author who wanted to submit first has a reasonable objection you cannot undo afterwards.