Searching grey literature
The evidence that never became a journal article is not a rounding error. In some fields it is most of the evidence.
What is grey literature and how do you search it?
Grey literature is research produced outside commercial publishing: government and NGO reports, theses, conference proceedings, working papers, standards, trial registrations and technical documents. It is searched through targeted sources rather than one index — repositories, agency websites, registries and specialist databases — and each source, date and search term must be logged, because none of it is reproducible by default.
Why it matters
Studies with null or unfavourable results are published less often and later than positive ones. Because grey literature is not filtered by editorial appetite, it holds a disproportionate share of that missing evidence — which is why excluding it does not merely make a review less complete, it makes it biased in a predictable direction.
In applied fields the argument is even simpler. Policy evaluations, engineering standards, clinical guidelines, industry technical reports and government statistics are frequently the only place a question is answered at all.
What counts as grey literature
| Type | Typical source |
|---|---|
| Government and agency reports | Department and agency websites, national repositories |
| NGO and think-tank reports | Organisation websites, OpenGrey successors, policy aggregators |
| Theses and dissertations | University repositories, EThOS, DART-Europe, ProQuest, NDLTD |
| Conference proceedings and abstracts | Society websites, conference sites, Scopus where indexed |
| Working papers and preprints | SSRN, RePEc, NBER, institutional series, preprint servers |
| Clinical trial registrations and results | ClinicalTrials.gov, WHO ICTRP, EU CTR |
| Standards and technical specifications | ISO, IEC, ASTM, IEEE, national standards bodies |
| Patents | Espacenet, Google Patents, Lens.org, USPTO |
| Datasets and data papers | Zenodo, Dryad, figshare, domain repositories |
How to search it reproducibly
There is no equivalent of a single database query. A grey-literature search is a set of targeted searches, and its credibility comes entirely from the log.
- List the organisations that would produce evidence on your question — agencies, regulators, funders, professional bodies, major NGOs. Write the list before searching.
- Search each organisation site directly, and record the URL, the date and the terms used.
- Search the general repositories and aggregators: CORE, BASE, OpenAIRE, and the national or subject repository for your field.
- Search trial and study registries where relevant, including for registered studies that never reported.
- Run a bounded web search, and fix a stopping rule in advance — the first hundred results, or the first ten pages — because an unbounded web search cannot be reported or repeated.
- Log everything: source, URL, date, search terms, and number of items screened. This log is the whole reproducibility claim.
Fix the stopping rule before you start, not when you get bored. First 100 results, sorted by relevance, on this date is reportable. I looked until nothing seemed useful is not.
Appraising it
Grey literature has not been peer reviewed, and its producers often have a direct interest in the finding. That does not make it unusable — it makes appraisal explicit rather than delegated.
- Who produced it, who funded it, and what outcome would suit them.
- Whether the method is reported in enough detail to be assessed at all.
- Whether the data behind it is available.
- Whether it has been superseded by a later version — reports are revised silently far more often than papers are.
- Whether anything independent corroborates the central claim.
The AACODS checklist — Authority, Accuracy, Coverage, Objectivity, Date, Significance — is a standard way to record this, and it takes a few minutes per document.
Reporting it
In a PRISMA 2020 flow diagram, grey literature found outside databases belongs in the other methods column, reported separately from database records. Name each source searched, with dates. A review reporting grey literature searched as a single line has not reported it.
Questions
- Is grey literature peer reviewed?
- Usually not, though some categories have their own quality processes — standards go through committee consensus, and government reports are often externally reviewed. Treat the absence of journal peer review as a reason to appraise explicitly rather than a reason to exclude.
- Should a systematic review include grey literature?
- In most fields, yes. Cochrane and JBI guidance both recommend searching beyond bibliographic databases, principally because omitting unpublished and null results biases the synthesis toward positive findings.
- How do I cite a report with no DOI?
- Give the author or issuing organisation, the year, the full title, the report or series number if there is one, the publisher, and a URL with the date you accessed it. Archive the page in a service such as the Wayback Machine, because agency URLs break often.
- What replaced OpenGrey?
- OpenGrey closed in 2020 and its records were folded into other services. In practice the working substitutes are CORE, BASE, OpenAIRE, national repositories, and direct searching of the organisations that publish in your field.
Related guides
- Finding theses and dissertations — How do you find theses and dissertations?
- Building a systematic review search strategy — How do you build a systematic review search strategy?
- Finding and using preprints — Where can I find preprints?
- The PRISMA flow diagram, explained — What is a PRISMA flow diagram?
Last updated 2026-08-25. Part of the litscout literature search guides.