That was the topic discussed recently by OCLC Research Library Partners metadata managers, initiated by Jennifer Baxmeyer of Princeton, Erin Grant of the University of Washington, and John Riemer of University of California, Los Angeles. Traditionally, the most common measure of cataloger productivity statistics on the number of records produced and time spent cataloging. As cataloging and metadata librarians become more involved in other activities that are not easily quantifiable (e.g., participating in linked data or similar projects), the problem of measuring productivity and success becomes more difficult.
Cataloging statistics have never captured the impact of cataloging on the end-user; they give no indication of how the bibliographic descriptions in the catalog actually contribute to a user’s success in finding the appropriate resource. Moreover, they add a perverse incentive for catalogers to work only on the easiest titles or problems, rather than on those with the biggest impact or best use of catalogers’ expertise, as those are the ones that will maximize statistics. Finally, since more cataloging data is available from outside sources, and records are often loaded in large batches into the library’s catalog, the number of titles added may not accurately reflect a single person’s contribution to the library’s catalog.
Our discussions focused on communicating the successes and challenges of metadata specialists with the rest of one’s institution and how metadata contributes to the division’s and organization’s strategic goals—especially “stories.” Strategic goals may include “foster discovery and use” of the library’s resources and “enriching the user experience,” for which good metadata is critical. Discovery layers often highlight flaws in the metadata.
Key to demonstrating the value of metadata is integrating metadata specialists into other activities undertaken by other units. Examples of such integration include participating in collection usage and weeding projects, moving collections, convening conversations between different groups such as digital collections, and publicizing who the metadata experts are who can provide metadata consultancy—offering “metadata as a service.” It’s important that other units in the library realize that they need a metadata expert before they’ve proceeded too far with a project that should have had one. Metadata managers have discovered many needs for consulting services, such as content analysis and taxonomy for public-facing web pages. Actively contributing to other units’ activities reinforces the importance of metadata.
Institutions use a variety of methods to communicate their work and completion of project milestones, including: on-line newsletters, blog posts, social media, tie-ins to other library activities such as exhibits, consultations, articles in professional literature, brown-bag lunches, workshops and other events, staff presentations, reporting out highlights from professional conferences as it reflects on one’s own work.
Although most metadata managers keep statistics (and may be required to do so) the impact of metadata goes beyond record numbers. We agree that focusing on outcomes and the impact of metadata on user discovery is important, but these outcomes are hard to measure or document. Metadata departments that have strong senior management support feel their work is valued. Managers may find it challenging to evaluate the work of individuals who do different types of work. Statistics do not adequately represent the breadth of ways materials are described for all the various types of materials collected, but they may be useful as a baseline to identify anomalies, trends, the effect of changed workflows, and the extent of backlogs rather than as a ruler to measure productivity. Knowing what remains to be cataloged is more important than what has been cataloged, and that number can be used to justify adding or retaining positions.
Metadata specialists try to manage the quality of batch-loaded records from vendors but can feel like it’s a “losing battle.” We noted that the distinction between “copy” and “original” cataloging is losing its meaning; the distinction between batch-processing and records you have to hand-touch is becoming more meaningful. This distinction is blurred when so much more metadata is generated by machines that needs to be remediated later. Meanwhile, more attention is spent on creating metadata for hidden or distinct collections and unpublished materials.
Part of the challenge is engulfed by ongoing concern about the position of libraries: is the library in the line of sight for the people we want to use it? Discovery happens elsewhere, and we’ve failed as a profession to get our content into the place people are looking for it, although our metadata can help to do this in a linked data environment. In Australia, discovery often happens first through Trove, a portal that aggregates content from libraries, museums, archives, repositories, and other collecting organizations around Australia. The absence of library linked data services today may be considered as a “temporary inconvenience” while the infrastructure evolves. RDA (Resource Description and Access) attempts to create a standard that is encoding-schema neutral, but vendors aren’t supporting it yet, and libraries aren’t demanding it.
Metadata managers need to balance allocation of staff to “traditional cataloging activities” with more exploratory R&D projects that do not directly relate to getting more metadata into catalogs, such as linked data projects and exploring Wikidata and ISNI. What’s needed is a culture shift, from pride around production alone to valuing opportunities to learn and explore new approaches. Metadata specialists need to understand that improving all metadata is more important than any individual’s productivity numbers. This culture shift requires buy-in from administrators to support training programs for staff to learn new ways of doing things and to view metadata specialists as more than just “production machines.” Metadata managers faced with staff reductions while still being expected to maintain production levels can justify allocating staff time for R&D or “play time” to explore such questions as: What can we stop doing? What is the one thing you learned that we all need to do more of? What do you need to move forward? What open source software could help us do the work more efficiently? Managers need to set goals for success not based only on numbers.
Part of this culture shift is to out-source or train support staff to create metadata for the “easier stuff” and mandate that catalogers only do what well-trained humans can do. Scope the materials requiring metadata that support staff or students can handle, providing templates where possible. If you take away all the easier materials, what’s often left are metadata that requires expertise in languages or formats and in describing (and disambiguating) persons, organizations, and other entities.
To encourage the culture shift among metadata specialists to change their mindsets about how they work and stimulate interest in learning opportunities, metadata managers have used several approaches:
- Identify who on your team has the aptitude to pick up new skills. At one institution, the staff member shared what she learned and the whole unit became “lively” because she brought her colleagues along. It created appreciation for “continuous learning” and staff presented at national conferences what they were doing.
- Convene group discussions to look at problem metadata and come up with solutions, encouraging staff to move forward together. Staff less interested in new skills can pick up some of the production from those learning new skills and producing less.
- Launch “reading clubs” where staff all read an article and respond to three discussion questions to inspire metadata specialists to think about broader metadata issues outside of their daily work. (One of the readings was my 2017 blog post on “New skills for metadata management.”)
- Hold weekly group “video-viewing brown-bag lunches” for staff on new developments such as linked data so staff can together “watch and learn.”
- Participate in multi-institutional projects.
- Encourage participation in professional conferences and standards development.
Discussants valued sharing ideas about facilitating the cultural aspects of shifting priorities and disseminating the value of metadata throughout the library and the institution.
Karen Smith-Yoshimura, senior program officer, topics related to creating and managing metadata with a focus on large research libraries and multilingual requirements. Karen retired from OCLC November 2020.