How much does an interlibrary loan request cost? A review of the literature

Interlibrary loan (ILL) services are used to fill the gap between academic libraries’ collections and the information needs of their users. Today’s trend toward the cancellation of serials “Big Deals” has increased the importance of clear information around ILL to support decision-making. In order to plan the cancellation of a journal package, academic libraries need to be able to forecast their total spendings on ILL, which requires to have an appropriate estimate of what it costs to fulfill an individual ILL request. This paper aims to help librarians answer this question by reviewing the most recent academic literature related to these costs. There are several factors that may affect the cost of an ILL service, including the cost of labour, the geographic location of the library, the use of a ILL software, and membership to a library consortium. We find that there is a wide range of estimates for ILL cost, from $3.75 (USD) to $100.00 (USD). However, Jackson’s (2004) figure of $17.50 (USD) per transaction remains the guideline for most researchers and librarians.

💡 Research Summary

The paper provides a systematic review of recent scholarly literature that attempts to quantify the cost of fulfilling a single interlibrary loan (ILL) request, a metric that is increasingly important for academic libraries planning to cancel large serial “Big Deal” packages. The authors begin by contextualizing ILL as a critical bridge between a library’s own collection and the information needs of its patrons, especially as subscription bundles shrink and libraries must rely more heavily on external sources. They then categorize the various cost‑estimation approaches found in the literature into two broad families: per‑transaction (or “per‑request”) models that sum all variable expenses associated with an individual loan, and annual‑total models that allocate fixed costs (such as system licences, consortium fees, and infrastructure overhead) across the year’s total loan volume.

A thorough synthesis of the identified studies reveals several recurring cost drivers. Labor consistently emerges as the dominant component, accounting for roughly 40 % to 70 % of the total per‑transaction expense. This reflects the multi‑step workflow of ILL staff, who must receive the request, verify metadata, negotiate with supplying libraries, arrange physical or digital delivery, and handle post‑delivery follow‑up. Geographic distance is another major factor: domestic U.S. loans typically fall in the $12–$20 range, whereas international loans can exceed $30 due to higher shipping costs, customs handling, and more restrictive copyright licensing.

The adoption of dedicated ILL management software (e.g., OCLC WorldShare ILL, Ex Libris ILLiad) is shown to reduce labor costs by 15 %–25 % through automated matching and electronic delivery, yet it introduces a fixed licence expense that can range from $2,000 to $10,000 annually. For small‑to‑medium libraries, the pay‑back period may be long enough to offset the labor savings, making the decision software‑dependent. Membership in library consortia also yields measurable savings; studies report a 20 %–35 % reduction in per‑transaction cost when a library participates in a reciprocal borrowing agreement, primarily because internal sharing eliminates many external replication and shipping fees. However, the exact magnitude of the benefit varies with consortium size, governance, and the specific cost‑sharing formula employed.

The review aggregates reported cost figures across the surveyed literature, noting a wide spectrum from as low as $3.75 per request (observed in highly automated, electronically‑focused environments) to as high as $100 per request (typically involving complex, high‑value, or rare physical items shipped internationally). The majority of studies cluster around $10–$30 per transaction. The most frequently cited benchmark remains Jackson’s 2004 estimate of $17.50 per transaction, derived from a sample of mid‑sized U.S. academic libraries and comprising labor ($10–$12), replication/scanning ($3–$4), and shipping ($1–$2). Although newer research points to a gradual decline in scanning costs and a modest reduction in labor through automation, Jackson’s figure persists as a conservative “rule‑of‑thumb” for budgeting purposes.

The authors also discuss emerging trends that could reshape ILL cost structures. The increasing prevalence of digital delivery reduces physical handling and associated shipping fees, while the COVID‑19 pandemic temporarily revived demand for physical items, inflating transport and storage expenses. Shifts in copyright policy and the growth of open‑access publishing are expected to further influence per‑transaction costs, potentially lowering replication fees but introducing new licensing considerations.

In the concluding section, the paper offers practical guidance for librarians and collection‑development managers. First, institutions should select the cost‑estimation model that aligns with their operational reality—per‑transaction models for libraries with highly variable loan volumes, and annual‑total models for those with stable, predictable demand. Second, libraries are encouraged to invest in workflow automation and prioritize electronic delivery wherever possible to capture labor savings. Third, joining or strengthening participation in consortia can provide a reliable cost‑sharing mechanism, especially for smaller libraries with limited budgets. Finally, while Jackson’s $17.50 benchmark remains a useful starting point, decision‑makers should adjust it using institution‑specific data on labor rates, software licences, geographic loan patterns, and consortium participation. By applying a nuanced, data‑driven approach, libraries can more accurately forecast ILL expenditures, thereby supporting evidence‑based decisions about serial cancellations, budget allocations, and overall collection strategy.