Somewhere, a student searches for a book that her university does not own. A librarian finds a copy two states away, sends a request and, before long, the book begins its trip. Elsewhere, an archivist describes a century-old photograph once, then makes that record visible to searchers far beyond the reading room. These ordinary acts depend on agreement: what an item is called, who holds it, and how one institution can ask another for it. OCLC is the organization that turned that agreement into machinery.
Based in Dublin, Ohio, OCLC is a nonprofit technology cooperative for libraries. Its best-known creation is WorldCat, the shared database that tells systems what libraries own. Around that data, OCLC sells cloud management software, cataloging tools, discovery services, remote-access software, digital-collection hosting, analytics and the rails for interlibrary loan. The initials once meant Ohio College Library Center. The geography changed; the habit of pooling work did not.
This makes OCLC an odd and revealing company. It has approximately 1,400 employees and reported $262.7 million in library-services revenue for fiscal 2025, yet it has no shareholders and no venture rounds to narrate. Member institutions pay for shared services. OCLC aims for a modest surplus and reinvests in systems, research and programs. It behaves like an enterprise software provider while carrying the public purpose of a cooperative.
The product is the agreement
In 1967, 54 Ohio college and university libraries founded the cooperative under the leadership of Frederick G. Kilgour. The immediate problem was gloriously unglamorous: libraries were separately cataloging many of the same books. In 1971, librarians at Ohio University entered the first records into OCLC’s online union catalog. Once a good record existed, another library could reuse it rather than begin again at a typewriter.
The savings sound small until they meet scale. OCLC says members copy-catalog 95 percent of their items from WorldCat, saving an average of about 10 minutes per title. Reuse also improves consistency. A standardized title, author, edition and subject trail makes collections easier to search and libraries easier to connect. The obscure metadata job becomes a productivity system.
That loop is OCLC’s moat. Every participating library adds holdings or improves records. Publishers and aggregators send metadata. OCLC’s specialists perform authority work, deduplication and quality control. Better data makes the services more useful, which gives institutions more reason to participate. A conventional vendor can build a capable library system. Reproducing decades of cooperative records, relationships and professional trust is harder.
“Because what is known must be shared.”OCLC’s operating idea, condensed to seven words
A stack for the whole library
WorldCat is the foundation, not the whole house. WorldShare Management Services handles acquisitions, circulation, electronic resources, discovery and analytics in one cloud platform. WorldCat Discovery gives patrons a search interface across local and global collections. WorldShare Interlibrary Loan and Tipasa connect a request to possible lenders and manage the transaction. EZproxy authenticates people who need licensed databases from outside the building. CONTENTdm publishes digital collections. Choreo Insights helps institutions compare and shape what they hold.
Then there is the specialized layer. Record Manager and Connexion serve catalogers. Dewey services maintain the classification system familiar from shelf labels. WorldCat.org gives the public a free route into billions of library-held items. APIs let developers put holdings and bibliographic data into other applications. WebJunction offers training and community programs for library staff. OCLC Research studies subjects from collective collections to responsible metadata.
The 2024 acquisition of CloudLibrary pushed the portfolio further into a public library’s digital front door. The service delivers ebooks, audiobooks, magazines, newspapers, comics and streaming media. At acquisition, nearly 500 libraries in 20 countries used it. The move placed OCLC more directly against consumer-facing digital-lending platforms while giving CloudLibrary a path into OCLC’s infrastructure and member network.
Who buys it, and why
OCLC’s customers range from tiny rural public libraries to universities, medical centers, law firms, government agencies, museums and national libraries. Their constraints differ, but the recurring problems are recognizable: duplicated work, fragmented systems, inconsistent records, hard-to-find collections, expensive digital licenses and users who expect search and delivery to feel as immediate as the commercial web.
A small library gains systems and expertise it could not sensibly build alone. A research university gains a global view of holdings, which can inform whether to buy, preserve or withdraw a title. A consortium can coordinate collections across campuses. A patron gains a trail from a search result to a nearby copy. Publishers and technology partners gain standardized pathways into library workflows.
OCLC’s revenue model follows the work. Libraries subscribe to software and services, often on annual contracts; pricing varies by product, region and institution. Members share costs through use. The organization also maintains an investment portfolio, which helps fund operations alongside service revenue. There are no equity investors expecting an exit. The tension is instead between affordability, maintenance and the cost of modernizing infrastructure that must remain available.
OCLC does not merely sell a catalog. It sells participation in a catalog that other institutions keep making more valuable.
Competing on shared context
The library software market has serious alternatives. Clarivate’s Ex Libris Alma and Primo, EBSCO’s FOLIO and discovery service, SirsiDynix Symphony, Innovative’s Sierra and the open-source Koha ecosystem all answer parts of the management problem. OverDrive is a major alternative to CloudLibrary. Some competitors also integrate WorldCat data, which is telling: OCLC can compete at the application layer and remain useful as infrastructure underneath.
Its difference is not that every OCLC screen or workflow is automatically better. The advantage is context. WorldShare Management Services starts with WorldCat. Resource sharing can see who owns what. Analytics can compare a local shelf with collective holdings. Discovery can move from a global record to a local copy. The pieces speak a common bibliographic language because the network predates the products built upon it.
Partnerships extend that context beyond libraries. Google can direct a web searcher toward a nearby catalog. Goodreads, Wikipedia and Google Books can surface “find in a library” routes. Publishers and aggregators including EBSCO, ProQuest, Gale, Elsevier and Wiley contribute metadata. OCLC says it works with more than 400 partners. Cooperation here is less a press-release flourish than a distribution strategy: put library collections where curiosity begins.
AI enters the catalog room
In late 2025, OCLC added AI suggestions for Dewey and Library of Congress classification numbers and Library of Congress subject headings to Record Manager and Connexion. Catalogers can accept or ignore them, and libraries can decide which suggestions appear. OCLC reported that pilot users saved as much as 20 minutes per title. The human-in-the-loop design fits the institution: use a vast record corpus to reduce repetitive analysis, but leave professional judgment in the chair.
The company also applies machine learning to duplicate detection and resource-sharing routes. Its systems can consider recent lender activity to send requests where fulfillment is more likely, while an Express program connects libraries able to deliver articles and digital resources quickly. More than two million copy requests had been filled through Express by July 2025.
Linked data may be the deeper shift. Traditional library records were designed for library systems; entities and durable identifiers allow people, places, works and concepts to connect across the web. During fiscal 2025, OCLC says it added more than 400 million linked-data entities to WorldCat records. The goal is not to discard the cataloging practices libraries depend on, but to give that work a life outside the catalog.
The quiet advantage
OCLC’s workplace mirrors its market position: librarians beside engineers, researchers beside product teams. The organization talks about purpose, learning and respect; Computerworld has recognized it among its Best Places to Work in IT 17 times. That culture is not incidental. Library systems carry records in thousands of languages and scripts, preserve contested histories and mediate access to public knowledge. The work rewards people who can take both uptime and description seriously.
Recent growth is steady rather than theatrical. Forty-eight libraries joined WorldShare Management Services in the 12 months reported in June 2026. The University of Medellin selected the platform in July. OCLC launched an Italian Discovery Catalog, partnered with Irish research libraries on national discovery, and used WorldCat data to build an America’s 250-Year Bookshelf. These are variations on its old proposition: one institution’s collection becomes more useful when it can be understood beside others.
That proposition has survived mainframes, CD-ROMs, the web, cloud software and now generative AI. OCLC fits between database company, SaaS provider, research institute and civic utility. Its customers can choose other interfaces and systems, but the cooperative’s deep asset is the accumulated decision to share. The next time an unlikely book appears in a local search or travels across a state line, the machine will have done its work without asking to be noticed.