Should you build or buy Library Management System (ILS)?

Library Management System (ILS) software runs a library's daily operations: cataloging using MARC records, circulation and holds, acquisitions and vendor orders, and patron account management. It is the system a librarian uses to check an item in, and the one a patron's search relies on, whether the library is a single branch or part of a shared consortium catalog.

You can run Koha, Evergreen or FOLIO yourself, and Stanford, Leipzig and Georgia PINES do, but a library gains no edge from operating its own catalog; the commercial vendors and the open-source hosting providers are both selling the service envelope around the same workflows. Decide on who operates and supports the platform, not on whether open source can.

Build it, buy it, or bridge?

⚒ Build it
✓ Buy it
➔ Bridge
Cost shape
No license fee; the cost shifts to systems staff, servers, upgrades, and any paid migration help.
OCLC WMS and Alma run tens of thousands of dollars a year; Destiny prices per site.
Self-host the catalog core and buy migration or support services only when you need them.
Time to value
Koha, Evergreen, and FOLIO are mature and installable now, but a full migration still takes months.
A vendor-hosted rollout is turnkey, with onboarding and data migration handled by the vendor's team.
Migrate onto open-source software with a paid migration partner, then operate it yourself after.
Differentiation captured
You control the catalog stack, but cataloging and circulation are identical at every library.
The vendor's service envelope is the product; the underlying workflow is the same either way.
Own the catalog data and workflow; let a hosting provider carry uptime and upgrades.
AI feasibility today
AI-assisted coding lowers the bar for customizing FOLIO or Koha modules without a large developer team.
Vendors are adding AI-driven discovery and metadata enrichment on top of the same cataloging core.
Run the open-source core yourself and add an AI discovery layer independent of who hosts it.
Who it fits
Institutions or consortia with existing systems staff who already value owning their infrastructure.
Libraries wanting one bill and one support line instead of owning servers and upgrade cycles.
A consortium with technical staff for the catalog core that still buys support for migrations.

When building makes sense

Running your own ILS is a mainstream choice, not a fringe one. Stanford's 2023 migration to FOLIO was its first new library services platform in over 20 years, and it self-hosted the system after moving bibliographic and holdings data for more than 12 million library items. Leipzig University Library has run self-hosted FOLIO since 2020, and Georgia's PINES consortium operates Evergreen for a large network of public libraries. Koha, Evergreen, and FOLIO cover cataloging, circulation, acquisitions, and holds at the same depth as the commercial platforms, and a consortium with systems staff can run any of them without a vendor's name on the contract. Stanford still paid IndexData for specialized data migration and load testing, which is the honest cost most self-hosting libraries carry: the software is free, and the migration and ongoing operation are not. This path suits an institution or consortium that already has the systems staff and wants to keep that expertise in house.

When buying makes sense

Buying makes sense once you weigh what a library actually gains from running its own catalog server: nothing that shows up to a patron. Ex Libris Alma, OCLC's WMS, SirsiDynix Symphony, and Follett Destiny bundle hosting, upgrades, support, and in some cases a knowledge base or discovery layer on top of the same cataloging and circulation workflow the open-source systems run. A single library without dedicated IT staff gets a working system on day one instead of a systems administrator's ongoing project. The open-source path still requires someone competent to run servers, apply upgrades, and answer for outages, and Stanford's own migration leaned on outside help for the hard parts even with an existing systems team. If your library has no one to own that role, a vendor's service envelope is the cheaper decision once staff time is counted.

The desk read

Build-versus-buy analysis for Library Management System (ILS) is being written. In the meantime, the framework that drives every B4 call is on the B4 Index page.

Representative vendors Ex Libris AlmaFolio + 5 more, listed in the full index

Frequently asked

Is Koha, Evergreen, or FOLIO actually used by real libraries, or is it mostly a hobby project?

They run production catalogs at scale. Stanford self-hosts FOLIO after migrating over 12 million library items, Leipzig University Library has run it since 2020, and Georgia's PINES consortium operates Evergreen across a large public library network. These are not test deployments.

When does buying a commercial ILS make more sense than open source?

When your library has no systems staff to own servers, upgrades, and outages. Alma, WMS, Symphony, and Destiny bundle that operational burden into the subscription, which is worth paying for once you count the staff time a self-hosted system would otherwise require.

When does self-hosting an open-source ILS make sense?

When your institution or consortium already has systems staff who can run servers and handle upgrades, and values keeping that capability in house rather than paying a vendor's service fee for it every year.

Does self-hosting actually save money?

It removes the software license, not the operating cost. Systems staff, infrastructure, and any paid migration or load-testing help still have to come from somewhere, and even a well-resourced consortium can lean on outside specialists to get a large migration right.

The B4 Index scores every software category on two axes, strategic differentiation and AI feasibility, to classify it Build, Buy, Bridge, or Beware. See the full methodology.