Home Find grey literature and policy

Find grey literature and policy

Learn how to use Overton to find grey literature and policy documents for research and to support evidence synthesis reviews.

By Euan

• 4 articles

How does Overton calculate relevance?

Overton’s full-text search function is powered by Elasticsearch. The default relevance scoring algorithm used by Elasticsearch is BM25, which is a version of the TF-IDF (Term Frequency – Inverse Document Frequency) model. TF-IDF gives higher relevancy scores to words that appear often in one document but are not common in other documents. Term frequency (TF)— The more documents that contain a search term in the field that we are searching, the less important that term is or how often does the term appear in this document. **Inverse document frequency (IDF) **— The more times that a search term appears in the field we are searching in a document, the more relevant that document is or how often does the term appear in all documents in the collection. Overton’s boost settings Users of Elasticsearch can also adjust the relevancy score by using boost settings. Boosts prioritise specific terms or fields within a document for where a search term can be found. Our current relevancy score calculation is: _score = boost * idf * tf Overton’s boost settings help us fine-tune our search results to what is particularly relevant for our users. Our boost settings (These are current as of July 2024. We may be tweak the from time to time). - Boost is 1 by default - Boost is 20 if the match is in title or translated_title - Boost is 10 if the match is in the snippet - Boost is 3 if in PDF title - Boost is 5 if match is in “other_identifiers” e.g. the source’s internal identifier If you have further questions about our search or relevance scoring, please contact [email protected]

Last updated on Jun 25, 2026

Search using Topics

Overton currently has over 650,000 topics connected to policy documents. Topics can be useful when trying to locate policy documents and other grey literature for research purposes. To use topics in your search, start off with a refined keyword search to start – this will help narrow down the topics found in your search results to ones that will most likely be relevant to you. To build an effective search query, try using our Advanced Query Builder. You can learn all about our Advanced Search here. If you are working with topics specifically for your research, using Overton’s REST API will allow you to see more than the filter view. Email [email protected] to learn about API access. Overrepresented topics We also identify “overrepresented” or “unusually common” topics that appear more frequently in specific contexts than in the general policy document collection. For example, when viewing an individual’s policy citations, the blue topics shown are those that appear more frequently in documents citing their work compared to all policy documents. The topics filter does have a limit, meaning smaller or more niche topics may not appear unless you first search for related keywords to bring them into the filtered results. Workarounds to find specific topics To access smaller topics that don’t appear in the main filter: 1. Perform a keyword search for their topic of interest 2. Examine the topics filter after the search results appear 3. Look for additional relevant topics that now show up in the sidebar 4. Add the desired topic filter, then remove the original keyword search For example, after searching for the keyword “resilience,” the topics filter reveals additional options like “urban resilience” and “resilience (engineering and construction)” that weren’t visible in the initial topic list. Combining topics Currently, topics cannot be combined using AND logic. To search for multiple topics: - Use keyword strategies with proximity syntax instead of topic filters - Export results from separate topic searches and find overlapping Overton IDs using Excel or scripts - Search for the relevant Overton IDs to see and combine related topics using the topics filter

Last updated on Jun 25, 2026

Searching for similar policy documents

Overton supports semantic searching of our policy documents. Semantic searching focuses on meaning rather than exact keyword matching. How does it work? Our semantic search is powered by our AI generated document descriptions which provide an overview of the document and the themes contained within it. Find out more about how we generate our documents descriptions here. The results of a semantic search will be documents that have a conceptual similarity to your search terms rather than just keyword matching. Search syntax In order to conduct a semantic search you will need to prepend your search with similar:. For example, when conducting a search looking for documents that relate to the impacts of climate change on mental health the search might look like: similar: The document discusses the mental health impacts of climate change and provides guidance on how to maintain mental well-being during this challenging time. It highlights the various stressors associated with climate change, including extreme weather events, heatwaves, and droughts, which can lead to anxiety, depression, and post-traumatic stress disorder. This will return results for policy documents where the AI document descriptions contain themes related to the mental health impacts of climate change.

Last updated on Jun 25, 2026

Using Overton’s Advanced Search

Learn how our advanced search query builder works and which search strategies you can use in Overton Index. Our advanced search query builder guides you to create effective, precise searches. It helps you build complex queries, apply search strategies as they function in Overton Index, and refine your search with the query visualiser. Overton’s advanced search query builder Using the query builder This step-by-step guide introduces the query builder and shows you how to construct a simple search query for policy documents. Directory of advanced search strategies Boolean operators - Operators like AND, OR and NOT combine search terms and control how terms relate within a query. - Select operators from the dropdown that appears after a search term is entered. Boolean operators example Example: "covid-19" AND protection NOT masks This query finds documents that contain both “covid-19” and “protection” but do not contain the word “masks” within the full-text. Parentheses - Use parentheses to group terms and control logical order. - Search terms inside of the brackets are evaluated first as a separate expression. - Add parentheses by clicking on the ‘Add group’ or ‘Add nested group’ option below the query line. Nested groups example search Example: (governance AND ("data science" OR "artificial intelligence")) OR "AI ethics“ This query finds documents containing governance and (data science or artificial intelligence), or contain the exact phrase AI ethics. Phrase search - Search terms that are not contained within quotation marks will be searched for as individual words. - For example:income tax will match documents where there are instances of income and tax but not necessarily adjacent. - Find documents where your words appear together in a phrase by selecting ‘exact phrase’ from the dropdown box to the right of the query line. - A phrase or keyword search searches the full text of the document, including the title, translated title, and abstract (if available). It does not include document summaries, topics, or other fields. Example exact phrase search Example:"income tax" This query finds documents where “income tax” appears as a phrase with these words in this exact order within the full-text. Proximity search - Search for words occurring near each using the ~N operator. - “N” is how many words are allowed in between each part of your phrase. Example exact phrase search Example:"data science"~1 This query finds documents containing the words “data” and “science” next to each other or with one word between them where “data” appears first and “science” appears second. In this example, “data science” and “data and science” would match, but “data indicates that science” and “science of data” wouldn’t. Semantic search - Semantic searching focuses on meaning rather than exact keyword matching. - Using the similar search will find documents that share semantic similarity with the text you have used in your search. - Run a semantic search by selecting “Similar” from the “Search within” dropdown box at the top of query builder and entering at least 2 lines of text into the query box. - You can use Overton’s AI generated document summaries to search for other similar documents – just copy and paste the summary into the query builder and run a search. Example Similar search Example:The document discusses the mental health impacts of climate change and provides guidance on how to maintain mental well-being during this challenging time. It highlights the various stressors associated with climate change, including extreme weather events, heatwaves, and droughts, which can lead to anxiety, depression, and post-traumatic stress disorder. This query finds documents that will share similar context and content. Learn more about ‘Searching for Similar Policy Documents.’ Special characters and accents - Overton is sensitive to diacritics, accents or special characters. - Searching Overton for “nino” will return different results than a search for “niño”. - When conducting a search for an individual in the People search, we recommend including special characters. Example exact phrase search Example:“Bernhard Schölkopf” rather than “Bernhard Scholkopf“ Specific field searches for policy documents - Users can search within specific fields such as full-text, title, abstract, domain, policy organisation author or similar. - Users can only search within one field per query. Multi-field searching is not yet supported. - The field is selected by clicking on the ‘Search within’ dropdown near the top of the query builder. Example exact phrase search Explanation of search fields for policy documents - **Full-text:**searches within the full-text of policy documents including all the other fields. - Title: searches within the title only. - Domain: will return documents that cite specific domains. - Policy author: will find outputs from a specific organisational policy author. - ID: A site-specific identifier assigned to output like a catalog number or DOI (eg. JR127882 for a policy document authored by the Joint Research Centre ). We only collect these for certain sources. - Similar: Semantic search, focusing on meaning rather than exact keyword matching. Search is conducted using our AI generated document summaries. Additionally, users can search within the title and abstracts of policy documents simultaneously. - This field search requires manual entry in the query builder using the prepend abstract: before the search terms. - The limitation to this search as many policy documents do not have author provided abstracts. Example exact phrase search Specific field searches for scholarly articles - Users can search within specific fields such as author, title, abstract, ID or within all these fields simultaneously using ‘All fields.’ - Users can only select one field option per query. - The field is selected by clicking on the ‘Search within’ dropdown near the top of the query builder. Example exact phrase search Explanation of search fields for scholarly articles - All fields: searches within all available metadata fields for scholarly articles including title, abstract, authors, IDs and more. - Title: searches for words within the title of scholarly articles only. - Abstract: will find articles with the words within the title or abstract of scholarly works. - Author: finds scholarly outputs authored by a specific person - ID: searches for specific identifiers including DOIs, PubMed IDs, ORCIDs, or grant IDs. Tagged: booleanphraseproximityquery buildersearchsearch strategiessimilarwildcard Was this article helpful? ** Yes ** No Related Articles - Search using Topics - Searching for similar policy documents - How does Overton calculate relevance? - Summary reports for policy documents Leave a Comment Cancel You must be logged in to post a comment.

Last updated on Jun 25, 2026