Examples
Worked examples
- Is an instance
A postdoctoral researcher runs their own name through Publish or Perish against Google Scholar to pull a current h-index and total-citation count for a grant application's biosketch.
- Is an instance
A librarian uses Publish or Perish to run the same author search against both Scopus and Web of Science to show a faculty committee how citation counts differ by source before the committee sets a policy on which source it will accept.
Counter-examples
Looks similar, but isn't
- Not an instance
Treating a Publish or Perish h-index as a fixed, authoritative figure that will match the number shown on a researcher's native Scopus or Web of Science author profile -- the software recalculates from whichever source was queried at query time, and different sources return different underlying citation counts.
- Not an instance
Using Publish or Perish as a system of record for an institution's publication list -- it produces an on-demand snapshot, not a maintained, persistent institutional record; that role belongs to a CRIS such as Pure or Elements, or to a repository, not to this software.
Editorial commentary
The most useful thing to know about Publish or Perish before you cite a number it produced: the program contains no citation data. It is a query client. Every figure it shows you was fetched from somebody else’s database seconds earlier and recomputed on the spot, which is why the same author can come out with three different h-indexes in one afternoon depending on which source was selected. Publish or Perish is not a bibliographic database, and reporting a figure from it without naming the source and the date of the query is reporting a number nobody can reproduce.
What it is and where it came from
Publish or Perish is a free desktop program written by Anne-Wil Harzing. Its release timeline, published on Harzing’s own site, runs: first public release 17 October 2006; version 2 on 5 April 2007; version 3 on 1 June 2010; version 4 on 28 February 2013; version 5 on 28 October 2016; version 6 on 20 October 2017; version 7 on 2 September 2019, the first release with both Windows and macOS editions; and version 8 on 1 November 2021, which remains the current major release as at 8 August 2026. Point updates within version 8 continue to ship, so a specific build number should be read off your own installation rather than quoted from a reference page. The software is distributed as Windows and macOS applications, and the Windows edition can be installed and used on GNU/Linux with the aid of a suitable emulator.
Download and use are free (gratis) and Harzing states they will remain so; development is funded by optional donations and described as a volunteering effort ongoing since 2006. Harzing’s requested citation form for the software is Harzing, A.W. (2007) Publish or Perish, available from https://harzing.com/resources/publish-or-perish.
The data sources, and what each one actually costs you
This is the part most summaries get wrong, because the access conditions differ source by source. As documented in the version 8 manual’s data-sources page (checked 8 August 2026):
- Google Scholar – freely available.
- Google Scholar Profile – freely available, but the documentation notes that from April 2025 onwards author profile searches require a Google account login.
- Crossref – freely available. See Crossref DOI for what those records contain.
- PubMed – freely available.
- OpenAlex – freely available; the manual’s own note flags it as having had limited search options when support was added.
- Lens.org – requires an API key from Lens.org.
- Semantic Scholar – requires a free API key from Semantic Scholar.
- Scopus – requires a free API key from Elsevier, on top of your institution’s Scopus entitlement.
- Web of Science – requires a Clarivate subscription, typically provided by your organisation. See Web of Science.
- External data import – result sets can be imported from Web of Science, RefMan, EndNote and other formats instead of queried live.
Microsoft Academic was a supported source in earlier versions; Microsoft discontinued that service after 31 December 2021 and it is no longer a live option. Harzing’s documentation is explicit that the program is an interface to external providers and cannot guarantee their continued availability, which is the honest framing: the tool outlives its sources.
Why two runs disagree
Publish or Perish computes paper counts, total citations, the h-index, the g-index and related measures from whatever result set came back. Three things move that result set. The source matters most: Google Scholar indexes theses, preprints, working papers and conference material that Scopus and Web of Science exclude, so Google Scholar figures for the same person are routinely the highest of the three, and Google Scholar Metrics reflects that same wider net at journal level. The query matters next: a plain name search collects namesakes and misses works published under a different name form, which is the failure a curated Google Scholar Citations profile or an ORCID-linked record exists to prevent. And time matters: citation counts only go up, so a figure is worth exactly as much as its timestamp.
Because the software makes it trivial to compute an h-index from an unfiltered result set, it also makes it trivial to compute an inflated one – see h-index inflation and self-citation for the mechanisms that do the inflating. Publish or Perish applies no correction for either, and no correction for retracted or duplicated records.
Reporting a Publish or Perish figure defensibly
State four things together: the metric, the data source queried, the exact query used, and the date of the query. “h-index 24 (Publish or Perish 8, Google Scholar, author-name query, 8 August 2026)” is a claim a reader can check. “h-index 24” on its own is not. For anything that has to persist – an institutional publication list, an assessment return, an annual report – a live query snapshot is the wrong instrument entirely; that role belongs to a maintained CRIS or repository record, which keeps provenance the software never had.
Frequently Asked Questions
Is Publish or Perish a citation database?
No. Publish or Perish holds no citation data of its own – it is a query client that fetches results from other providers, such as Google Scholar, Crossref, PubMed, Scopus or Web of Science, each time you run a search, then calculates metrics like the h-index from whatever comes back.
Is Publish or Perish free to use?
Yes. Download and use of Publish or Perish are free, and Anne-Wil Harzing states they will remain so; development is funded by optional donations rather than licence fees.
Why do two Publish or Perish searches for the same author give different h-index results?
Because the figure depends on which source was queried, the exact query used, and when it was run. Google Scholar indexes theses, preprints and conference material that Scopus and Web of Science exclude, so it typically returns the highest counts of the three, and a plain name search can also pull in namesakes’ work or miss papers published under a different name form.
Which databases can Publish or Perish search?
As of version 8, it can query Google Scholar, a Google Scholar Profile, Crossref, PubMed, OpenAlex, Lens.org, Semantic Scholar, Scopus and Web of Science, or import result sets from Web of Science, RefMan and EndNote files instead of searching live. Access conditions differ by source – several require a separate API key or an institutional subscription.
Does Publish or Perish correct for self-citation or h-index inflation?
No. Publish or Perish computes metrics from whatever result set a query returns and applies no correction for self-citation, h-index inflation, or retracted and duplicated records.
How should you cite a metric that Publish or Perish produced?
State the metric, the data source queried, the exact query used, and the date of the query together – for example, “h-index 24 (Publish or Perish 8, Google Scholar, author-name query, 8 August 2026).” A bare number on its own cannot be checked or reproduced by anyone else.
Machine-readable encodings
Use in your systems
<role vocab="credit"
vocab-identifier="https://casrai.org/dictionary/"
vocab-term="Publish or Perish (Harzing’s citation-analysis software)"
vocab-term-identifier="https://casrai.org/dictionary/term/publish-or-perish-software" />{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"@id": "https://casrai.org/dictionary/term/publish-or-perish-software",
"name": "Publish or Perish (Harzing’s citation-analysis software)",
"identifier": "https://casrai.org/dictionary/term/publish-or-perish-software",
"description": "Publish or Perish is a free desktop program, created by Anne-Wil Harzing and first released in October 2006, that queries academic data sources (as of version 8: Google Scholar, Google Scholar Profile, Crossref, PubMed, OpenAlex, Lens.org, Scopus, Semantic Scholar, and Web of Science) for a given author, journal, or search term, then calculates citation-based metrics such as the h-index and g-index from whatever results are returned. It holds no citation data itself and is not a database, persistent identifier registry, or CRIS -- it is a query-and-calculate client for other providers' data, so results vary by which source was queried and change over time as underlying citation counts change.",
"inDefinedTermSet": "https://casrai.org/dictionary/domain/research-info-systems#set",
"url": "https://casrai.org/dictionary/term/publish-or-perish-software",
"sameAs": [],
"license": "https://creativecommons.org/licenses/by/4.0/",
"publisher": {
"@id": "https://casrai.org/#organization"
},
"dateModified": "2026-08-15T04:26:52",
"inLanguage": "en"
}
