With the current ILS market constricting into only a few players for academic libraries, how sustainable for a long-term strategy is the current subscription pricing model of 5% annual increases? Especially when a library's ILS consumes a large block of operations budgets, which don't grow at that pace, if at all. Libraries can cut serials and book budgets, but there is little flexibility for an ILS. You either have one or you don't. Do you see a potential “crash” where the same libraries who migrated to Alma or World- Share a few years ago are forced to migrate to open source systems, given the projected declines in higher education in general? Also, what do you recommend as an appropriate ratio between the annual cost of a library's inventory management system and the annual cost of a library's inventory to use as a determining factor of when to jump away from a particular ILS?
The consolidation of the library technology industry has many implications, including availability of systems and their costs. The library technology industry differs from most other business sectors because the customers are non-profits with constrained funding and lengthy budget planning cycles. I have previously mentioned research that shows that libraries generally have choices of multiple viable products in the current business environment, with a more varied slate of choices than in previous times (See Nov/Dec 2020 issue of Library Technology Reports.)
The pricing of products remains a valid concern that likewise warrants further research. Secrecy of pricing hampers the ability to perform this research. Almost all contracts for bigticket technology systems include non-disclosure clauses that preclude the library from sharing the price paid for the product. I see this practice as generally harmful to libraries. They are in a weaker position without the data needed to understand the market value of these products and related services. In the same way that libraries have made progress in removing non-disclosure terms from major contracts for content products, it seems that they could likewise insist on public disclosure of system pricing.
The current business paradigm of major technology products reflects some basic characteristics worth noting.
Subscription pricing. Libraries increasingly acquire technology products through software-as-a-service deployment models paid for through annual subscription pricing. These subscription fees cover all aspects of the product, including hardware, software, continual upgrades, and support services. A true web-based SaaS-based product obviates the need for local server equipment, software installed on user computers, and technical personnel for system administration. The total cost of these factors involved in a technology product housed and managed locally should be a point of reference when calculating the relative budget impact of a SaaS-based product.
Inflationary increases. Affecting all areas of spending, price increases exert tremendous pressure on library budgets. In academic libraries, annual increases in electronic journals, for example, cause sacrifices in other areas, especially monographs. Thus far, open access publication models have not yielded relief to collection budgets. Table 1 illustrates the levels of inflation academic libraries have seen in journal subscriptions, according to surveys and data from EBSCO Information Services.
|Table 1. EBSCO Information Services Serials Price Projections (percentages)|
|Year||Individual Titles||Ejournal Packages|
Flat library budgets. Academic library budgets overall have not kept up with inflation. According to figures derived from statistics published by the Institute of Education Sciences, average total expenditures have been mostly level from 2014-2018, with a total increase across those years of 6.6 percent (Table 2).2 (Data from the 2019 survey has not yet been published.)
|Table 2. Expenditures (billions $), All US Academic Libraries (2000–2018)|
The COVID-19 pandemic has disrupted university budgets. Library allocations, especially for 2020 and 2021, are likely to decline. Libraries will necessarily have to make painful cuts in spending beyond their already constricted budgets. The costs for a library automation system falls within these challenging budget contexts. Are the financial pressures so great that libraries will opt to change to less expensive systems? What are appropriate levels of spending for resource management systems and discovery services relative to total budgets or in proportion to collections spending?
Given the secrecy in pricing, it is difficult to assess how current pricing scales to the broader budget constraints. Some anecdotal information can be gleaned from public sources. A systematic study would be needed to posit spending trends with a high level of confidence. Based on a very small sample, I currently estimate that an academic library might spend 2 to 3 percent on these systems relative to its total collection budget. A library with a $10 million collections budget might spend $200,000 to $300,000 on its resource management and discovery services. Additional data from more libraries could be used to more fully articulate the current trends on spending for these products. The costs of a resource management system usually represent less than 1 percent of the library's overall budget. The cost of these products is a small part of a library's overall budget scenario compared to collections expenditures and personnel.
Personnel expenditures have seen less growth in recent years than other categories. Most libraries operate with fewer personnel than in previous times. A reduced workforce means that libraries must operate more efficiently than ever. They rely on their resource management systems more than ever to automate routine tasks. Library collections have become more complex than ever, with electronic, digital, and print components, each with its own quirky procurement processes. Traditional subscription-based procurements to electronic resources were already complex. The trend toward increased proportions of open access content has further increased complexity of management. Some of the additional factors requiring attention include tracking open materials available beyond subscriptionbased entitlements and article processing charges.
The increased complexity of collection management and the reduced staffing levels in libraries increases the need for sophisticated technology systems. These factors may argue against changing to a less expensive automation system with reduced capabilities. If a library does consider moving away from a library services platform, it is critical to ensure that any candidate products likewise address electronic and print resources and include the knowledge base and discovery indexes needed to work efficiently.
Open source systems may or may not be less expensive than proprietary products offered through subscription pricing. Again, there is limited data available to assess the cost of operating open source ILS or LSP products relative to proprietary products. Although the software itself is provided without a direct fee, the total cost of operation over time are likely to be similar. The cost of a subscription to a comprehensive SaaS-based library services platform would need to be compared to all the components of operating an open source service. In most cases these cost components would include a service contract for hosting, product maintenance, and support, as well as a subscription to a discovery service and e-resource knowledge base. Operating an open source product will not necessarily involve additional local personnel, unless the library opts for a self-managed and hosted implementation.
Libraries will also need to consider switching costs as they evaluate whether changing systems will result in budget relief. Some of these costs include data extraction and other exit services from the incumbent vendor, data migration and loading services to the new provider, other installation and set-up fees, as well as vendor-provided or in-house training. It often takes a library a year to return to the same level of productivity as achieved on its previous system, even when the new system has more sophisticated capabilities. The cost of switching and the intensive work required under the best of circumstances have led to libraries' retaining their systems for long periods, on average 12 or more years. During times of difficult budgets, the organizations or administrations funding libraries may also resist requests to fund a new system unless substantial savings can be projected.
Under extreme budget scenarios, libraries may also negotiate with their vendors for price concessions. Vendors may be willing to soften pricing rather than risk losing an account. Subscription contracts are usually based on a set of factors in place at the beginning of the subscription term. Factors may include institutional FTE, the number of library personnel, collection size, and transaction volume. The pricing for subsequent years tends to be determined by inflationary increases on the first-year base price. If these factors diminish, the library may be able to rationalize subscription fee adjustments. It is interesting to note that Biblionix adjusts its subscription fees for its Apollo customers annually and those that see lower use of their systems also see reductions in costs.3
Related questions apply to metadata management. Many libraries allocate similar costs to bibliographic services and resource sharing services as they do for their integrated library systems or library services platforms.
All these factors lead me to suggest that there will not be substantial defections in the short term away from comprehensive products such as OCLC WorldShare Management Services and Ex Libris Alma. A more likely trend would be a higher proportion of academic libraries moving away from legacy integrated library systems and opting for open source products, especially FOLIO and Koha. Data from the annual Library Automation Perceptions Survey suggests that as a driver of FOLIO and Koha increases among academic libraries. Those libraries that have implemented Alma and WMS show little interest in changing systems. It will be interesting to see if the results of the 2020 survey now underway, to be published in February 2021, show any new trends in this regard.
- Library Technology Guides, “Number of Contracts Made for Automation Systems by Year,” https://librarytechnol ogy.org/products/procurements.
- Library Technology Guides, “Statistics published by IPEDS for All US Academic Libraries,” https://librarytech nology.org/libraries/nces/trends.
- Library Technology Guides, “Biblionix fee reductions and COVID-19,” https://librarytechnology.org/pr/25709.