Librarians have been actively examining overlap in research database coverage for many years in an attempt to provide the widest possible coverage of the published literature in the most fiscally responsible manner possible. As library budgets become more stretched by ever-increasing journal subscription costs and decreasing institutional budgets, database comparison articles become extremely useful in assisting with decisions on possible database cancellations/purchases.
In 2018, librarians at the University of Maryland were asked to justify purchase of agricultural databases given the free availability of Google Scholar. The result was the following article:
This guide is intended to be a summary of database comparisons, including a collection of links to database comparisons published within the last 5-7 years. It is not a comprehensive listing.
Where possible, data directly from database providers is provided. If not available from that source, research study data is provided instead. It should be noted that each research study has different methodologies and results vary.
Also, some databases are periodically re-indexed to adjust and improve the search results. This can dramatically change numbers of publications found and prevent searches from being 100% repeatable with any accuracy. For example, Google Scholar in a May 2025 report preprint was listed as containing 399M records. In an August update provided in a Blue Sky discussion thread, by the author of the May preprint, GS was listed as containing 364M records.
For librarians and academic researchers, the point is not to use whichever database has the highest numbers. It’s to figure out which database provides the best results (and in some cases the most comprehensive results) for their searches. Raw numbers of records or journals are insufficient alone. It is essential to know the strengths and weaknesses of each database, along with what is (or is not) included in the database. An additional challenge is that databases that work well for the sciences (where scholarship is primarily journal article focused) are known to not perform nearly as well for humanities and social sciences disciplines (where scholarship is a mix of journal articles, book chapters, books, policy documents, and publications in non-English languages).
