The essential link between publishers, librarians and researchers.

 
Recent

Open access is not broken. Discovery is.

An OCLC report, Unlocking the future of e-resource management, explains how to make OA routes reliable in hybrid environments

OA intent is high; OA success is patchy

When a student hits a paywall, the next step is often predictable. They look for an open access version.

OCLC Research confirms this behaviour. Its Open Access Discovery1 study found that while users actively seek scholarly, peer-reviewed OA publications, they still find it not very easy to discover or access. The problem is not motivation. It is whether OA is operationalised as part of the same end-to-end pipeline as licensed content, so links, versions, and rights information stay consistent wherever users start.

In Article 1 of this series, we argued that metadata is now strategic infrastructure – an asset that requires attention at the highest levels of universities and research institutions. This second article gets practical. If you want OA to work consistently, you have to make three things routine: version clarity, rights clarity, and link priority.

A new report from OCLC, Unlocking the future of e-resource management, builds on earlier work in OCLC’s “Managing e-Resources” series and draws on the OCLC Research led by Ixchel Faniel on open access discovery. It sets out a playbook libraries can run now to reduce friction in discovery and access, especially in hybrid collections where open and licensed content sit side by side. 

All of this and more will be explored in OCLC’s forthcoming webinar: Simplifying e-resource management for the age of OA, AI, and hybrid collections, on Tuesday 2 June – more details at the end of this article.

A quick scenario: the undergrad who just needs the PDF

An undergraduate searches for a seminar reading. They click a result and hit a paywall. They try again and find a repository copy. It downloads, but it is not obvious whether it is the accepted manuscript or the final published version. There is no clear reuse statement. The student is left guessing whether it is safe to cite, safe to share, and even whether it is the right thing.

This is the moment where OA either feels effortless or feels risky. The difference is metadata that carries version and rights signals all the way through to the point of decision.

“Users do look for open versions, but they won’t find them reliably unless identifiers and rights travel consistently across systems.” Ixchel Faniel, Senior Research Scientist, OCLC Research

What “open access” really means in practice

Open access is scholarly content that is free to read online and often free to reuse under stated terms. OA copies can be preprints, author accepted manuscripts, or the version of record. These versions can look similar to users while carrying different signals about peer review, citation, and reuse.

That is why OA discovery is not just about finding a free copy. It is about helping users identify what they are looking at and whether it is the best lawful copy for their purpose.

What breaks in hybrid contexts

Hybrid journals and transformative agreements have made OA and subscription content coexist at the article level. In theory, this expands access. In practice, it can create messy outcomes when metadata is incomplete or inconsistent.

Without rigorous article-level metadata and reliable propagation of license and version information, discovery systems can:

  • Surface duplicates that look like separate items;
  • Hide legitimate open copies behind weaker links; 
  • Route users to a paywalled version even when an OA copy exists; 
  • Fail to help users distinguish between a preprint, accepted manuscript, and version of record; and 
  • Leave users unclear about what they can reuse, share, or cite.

Unfortunately, the reality is that article-level metadata for OA titles is often incomplete. Harvesting and aggregation rules can be opaque. Labels sometimes confuse OA status or version. Closing these gaps turns OA policy into practice by making open versions visible alongside licensed copies, and by ensuring that version and rights signals are clear at the point of need.

A unified OA playbook

The message according to OCLC’s new report, Unlocking the future of e-resource management is straightforward, if you want OA to work reliably for users, you need to treat it as part of the same end-to-end pipeline as licensed content. The report translates that into four moves libraries can implement now:

1) Integrate OA and licensed content at source

If OA is activated through a different route than licensed content, it will drift. Links will break in one place and not another. Coverage will update in one system but not the next. Users see the consequences as dead ends and inconsistent results.

The practical fix is to register OA collections and entitlements in one vendor-neutral knowledge base so holdings and links travel together into discovery and link resolution. 

2) Make version and rights visible where users decide

Users make fast decisions in results lists. They need signals that help them choose confidently.

The report points to metadata that supports this moment of decision, including version information, license and reuse information, peer review context where available, and persistent identifiers that connect versions and copies.

In practice, this means improving how discovery and link resolution present options. If an OA copy is available, it should not be buried. It should be prioritised alongside licensed access routes, with enough context for users to recognise what they are opening.

A common example is prioritising legitimate OA links via integrations such as Unpaywall, so users can reach a usable PDF sooner. The point is not the tool. The point is that OA is treated as a first-class route, and the interface makes the version and reuse context clear.

3) Enforce identifiers at the article level

Persistent identifiers are what make OA discovery scalable. Without them, systems struggle to match, merge, and connect records across platforms.

The report highlights the value of identifiers such as DOI for works, ORCID for authors, and ROR for institutions. At article level, these identifiers support deduplication, version linking, and more reliable analytics. They also help metadata travel across systems without losing meaning.

4) Automate the volatile parts, but keep humans in control

OA coverage changes. Packages change. URLs move. Manual clean-up does not scale.

The report recommends machine-assisted enrichment to reduce duplication and improve link reliability, while keeping librarians in control of policy and exceptions. This matters in hybrid environments where the “best” link can vary by entitlement, version, and user context.

“When metadata, licensing, and access operate from one backbone, you stop firefighting links and start delivering outcomes.” John McCullough, Executive Director, Library Management Services, OCLC

Measuring what matters (beyond usage totals)

If you only look at usage totals, you miss the story of how users got there, or where they gave up.

Unlocking the future of e-resource management recommends combining multiple lenses:

  • COUNTER usage data (via SUSHI where possible) 
  • authentication-based insight that shows friction and failed routes 
  • peer benchmarking and programme alignment to support planning

This matters for OA because OA usage can be invisible in traditional access logs when users do not need to authenticate. It also matters because version confusion and duplicate records can distort cost-per-use and misattribute demand.

Counterpoint: “If we prioritise OA, do we risk lowering quality?”

A common concern is that OA can surface non-peer-reviewed content, especially when preprints are in the mix. The remedy is not to deprioritise OA. It is to clarify the signals users need to evaluate what they are seeing.

The OA discovery findings from OCLC2 points to peer review status, version information, and identifiers as key metadata that helps users decide what is trustworthy and what is appropriate to cite. Better OA discovery is not about pushing users toward one route. It is about giving them clear choices.

What we’ll delve into at the webinar

With Independent library technology consultant Ken Chad moderating, OCLC’s John McCullough and Ixchel Faniel will dig into real-world patterns behind OA discovery failures, how to operationalise version-aware linking, and where libraries get the largest gains from unified activation. Bring your toughest OA edge cases. We will focus on pragmatic fixes you can pilot this summer. 2 June, 14:00 BST/15.00 CEST.

→ Read more: download the e-resources report https://connect.oclc.org/ri-eresources-OA
→ Register now: register for the e-resources webinar
https://connect.oclc.org/ri-eresources-webinar-article2

Key takeaways

  • OA discovery breaks when OA is treated as a side channel. Unify pipelines. 
  • Make version and rights visible at the point of decision to reduce uncertainty and support confident use. 
  • Use identifiers to suppress duplicates, connect versions, and improve routing and reporting. 
  • Use automation to keep links and metadata reliable as content changes. 
  • Combine usage, authentication insight, and peer context to prioritise improvements and demonstrate value.

References:

1.  Faniel, Ixchel M., Brittany Brannon, Lesley A. Langa, Brooke Doyle, and Titia van der Werf. 2024. Improving Open Access Discovery for Academic Library Users. Dublin, Ohio: OCLC Research. https://doi.org/10.25333/4xem-xr80.

2. Faniel, Ixchel M., Brittany Brannon, Lesley A. Langa, Brooke Doyle, and Titia van der Werf. 2024. Improving Open Access Discovery for Academic Library Users. Dublin, Ohio: OCLC Research. https://doi.org/10.25333/4xem-xr80.

 

Back to top