Two US universities have recently taken different approaches to implementing discovery tools. Library directors Mary M. Somerville and Kristin Antelman reveal some of their experiences
Over the past few decades there has been a huge shift towards information in electronic form and this has placed new demands on libraries. In particular there is a significant need for tools that help library users to find relevant information from the huge body of content available. And to be truly effective, there is a need to search across many resources, in other words to search on a web scale.
Our two libraries – at North Carolina State University (NCSU) and the University of Colorado, USA (UCD) – have separately been looking at the issue of web-scale discovery. In particular, we have been looking at organisational readiness for web-scale discovery service implementation.
Over 10 years, NCSU Libraries has re-invented its workplace, a process that culminated in the opening of a new facility – the Hunt Library – in January 2013. The UCD Library has undergone a similar process over the past five years. Although we have used different approaches, we have both discovered that organisational transformation helped ensure successful implementation – at both institutions – of the Serials Solutions Summon discovery tool.
The risk of lack of ownership
UCD was careful about implementing Summon, having already failed to implement WorldCat Local. That project failed for a variety of reasons including exclusive project management, which meant the initiative remained ‘siloed.’ Lacking the support of public services librarians, and with technical services librarians’ contributions absent, the service was not adopted.
Having learned the importance of the whole library team being involved in and taking ownership of the project, the university’s library began its new discovery project, in 2008, with a study of technical services – the resource acquisition, management, and discovery functions of the library. The study, carried out by R2 Consulting, recommended that the library organisational structure should reflect that the library materials budget is largely expended on electronic resources and serials – in FY13 this was more than 85 per cent of the total budget.
As a result of this recommendation, the acquisitions, e-resources and serials, and metadata services were combined so that the processes associated with ordering, receiving, processing, cataloguing, and other access management could be coordinated into a single workflow. Technical staff were invited to help redesign the processes, procedures, and workflows.
This re-invention of the workplace helped to create a learning organisation. Philosophical elements and information practices supported new values such as ‘taking initiative’, ‘working together’, ‘leading from wherever you are in the organisation’ and ‘making decisions at the lowest appropriate level’. These philosophies helped ensure successful implementation of the web-scale discovery service.
In 2010, a newly-appointed ‘resource discovery task force’ was asked to ‘define, assess, and recommend how to implement the best possible library resource discovery experience for patrons given existing staffing and budget constraints.’ Three of five of the task force members were technical services librarians. The remaining two were recruited from information technology and public services. To encourage organisation-wide learning throughout the process, meeting minutes and other documents were routinely posted on the library’s intranet.
A weighted decision matrix was then developed to evaluate web-scale discovery solutions, with the ‘Big Five’ vendors being invited to present demonstrations. Following this, based on weighted decision matrix results, three products were selected for usability testing. Morae software captured ‘voice out loud’ observations on decision-making during practice searches. Recorded sessions were then shared with all interested library staff, and their comments were collected, before final data analysis and product recommendation.
In April 2011, the team unanimously agreed that Serials Solutions Summon was the best choice for UCD. Following procurement, a Summon technical services implementation team began building Summon as the ‘place to go for full text’. In keeping with the library’s workplace learning and shared leadership philosophy, frequent communication updates were sent to all library staff, providing information about progress, and delivering insights on the nuances of the discovery service experience. Staff also gained implementation and customisation ideas through the Summon customer listserv, Summon Suggestions.
Taking things slowly
This produced an internal/soft launch in November 2011, a beta launch in March 2012, and a home page launch in August 2012. The whole process took 23 months, during which technical services engaged all interested library staff to ensure adoption and promotion of the service.
With the launch of Summon on the website, web services staff members in the library information technology department immediately initiated refinements in the ‘single search box’ user experience. For instance, a programmer developed a link in Summon to our Prospector resource sharing network. Regular usability testing ensures continuous improvement of the user experience. The whole process demonstrated that realising the full potential of a discovery service engages an entire organisation – further reducing organisational silos and simultaneously eliminating content silos.
A different approach
For its part, NCSU took a different path but arrived at some of the same outcomes.
NCSU Libraries have a culture of innovation and risk-taking. The official vision statement is ‘The Libraries: NC State’s Competitive Advantage’ but the informal motto is ‘embrace ambiguity’. People are hired with that mindset.
The libraries began a transformation process in 2003 after several years of struggling with the impact of increasingly electronic collections, which at the time were mostly journals and databases. At that time there were no vendor products to support electronic resources management (ERM). As the constellation of Access databases orbiting the ILS was getting unwieldy, the libraries decided to build their own ERM system in-house.
While Summon now supports discovery of electronic collections, the ERM that the library developed – called Ematrix – supports the effective management of those collections and is still being developed today.
That initial project laid the organisational and cultural groundwork for subsequent major organisation-wide projects such as implementing a new catalogue front end driven by the commercial search software Endeca in 2006, Summon in 2009, and now the Kuali OLE and GOKb community-source library management software projects. NCSU is one of nine academic libraries working together under the auspices of the Kuali Foundation to build a community-source ‘open library environment’ to support library business processes and services. The Global Open Knowledgebase (GOKb) project is a collaboration between Kuali OLE and Jisc to develop a community-maintained open knowledge base to support management of electronic collections.
The Ematrix project brought the library’s culture of risk-taking and collaboration to technical services. It showed that technical services do not have to follow a traditional path; they can chart a custom path. It also brought together stakeholders from across the organisation to engage with the question of what it meant to manage the electronic collection and prompted creation of new feedback loops across departments.
Such developments challenged the culture in technical services. In particular, the project developed technical services librarians’ leadership skills and developed liaison roles in technical services that later emerged as e-resources librarian positions. The project also caused them to analyse how and why they do the work they were doing, and forced technical services librarians to learn to think about problems more like technology librarians.
What’s more, technical services librarians analysed workflows using business process modelling techniques. They analysed the data they worked with, and relationships among that data, and learned how to talk to programmers to get the functionality they needed.
The project demonstrated that technical services can move very fast. It also established the acquisitions librarian mindset – their ‘get it done’ attitude, and practicality – as the dominant culture in technical services.
The tool brings together resources in the UCD library
The Ematrix project played a role in the reorganisation and reenvisioning of technical services. The aim of the reorganisation was to realign the organisation to support predominantly-electronic collections.
The library now has a single technical services department, acquisitions and discovery, which consists of three units: serials; monographs; and data, projects and partnerships. Each unit is headed by a librarian and also has an e-resources librarian.
Resilient workplace culture
All of this helped when, in 2009, the library implemented Summon. The library needed the strength of culture to manage forces specific to that project that threatened to derail it.
Comparative product assessment was not conducted because, at that time, Summon was the only solution that met the library’s two requirements: being able to search article metadata without authentication, and offering suitable APIs. Summon’s approach resonated with students, in particular the ‘Google-like single search’.
True to NCSU library’s culture, senior management supported early adoption of the product. However, they engaged too deeply with the specifics of implementation. In any culture where administrators are stepping into such specifics, the outcome is tension that can make project leads and implementation teams feel that their expertise is being trumped by politics.
This problem was solved over several months when thought leaders with high organisation-wide credibility, and guided by a savvy department head, put forward an alternative approach in a politically-strategic manner. This approach was successful, in part, because the conversation was refocused on the importance of ‘library search’ as opposed to ‘collections search’. It was also important to demonstrate the continuing value in the pre-Summon search tool, QuickSearch, using data and user feedback about its use. The logical outcome was to build Summon into QuickSearch.
Summon implementations take one of three approaches: a single integrated search where local catalogue and digital collections records are loaded into the tool; side-by-side search with the catalogue, where books appear in one frame and articles in another; or ‘Bento box’ compartmentalised results. NCSU uses the Bento-box approach, which was an organic evolution from the QuickSearch approach.
NCSU’s philosophy is that there is value in the silos (articles, books and media, library website, databases, journals, and more search options) and that showing those silos, and enabling people to jump off into them, results in a better search experience. Now users click through to a link on the search results page 90 per cent of the time (up from an initial rate of 72 per cent and after several redesigns).
Web-scale discovery service partnerships
Unlike at UCD, the NCSU Summon implementation was led by public services staff with digital library and IT staff doing the technical integration. Technical services were involved mostly in a knowledge base migration from SFX to 360 Link and now describe the tool as "just another database" (i.e., an index to the journal literature that supports OpenURL linking). However, both institutions have learned that it is critical to partner with vendors and other libraries that extend beyond your own organisation.
NCSU feels that it is now a highly-successful participant in the Kuali OLE and GOKb projects due to its organisational readiness through these projects, and especially Ematrix.
What we see is that such gains, once firmly established in both culture and staff expectations, are not easily lost. A strong culture pointed toward the needs of the future enables an organisation to be inherently resilient and self-replicating.
Mary M. Somerville is university librarian and library director at University of Colorado Denver, USA. Kristin Antelman is associate director for the digital library at North Carolina State University, USA. This article is based on a joint presentation given at the recent UKSG meeting held in Bournemouth, UK
Somerville, M. M., Schader, B. J., & Sack, J. R. (2012). Improving the discoverability of scholarly content in the Twenty-First Century: Collaboration opportunities for librarians, publishers, and vendors. White Paper commissioned by SAGE. www.sagepub.com/repository/binaries/librarian/DiscoverabilityWhitePaper/
Somerville, M. M., & Conrad, L. Y. (2013). Discoverability challenges and collaboration opportunities within the scholarly communications ecosystem: A SAGE white paper update. Collaborative Librarianship, 5(1), 29-41. collaborativelibrarianship.org/
OLE project: kuali.org/ole
GOKb project: gokb.org