IGeLU/ELUNA Analytics WG
Discussions about an Analytics group began just before IGeLU 2015. Alma’s Product Working Group was forming an Analytics subgroup, and Ex Libris’s announcement of Oracle OBIEE as the analytics tool for hosted Primo suggested that a joint group would be useful. Skills in OBIEE would obviously be transferable across the 2 products. John Krug volunteered to get the group started and a round of recruitment began. The first remote meeting took place in April 2016.
Since then (coordinator / deputy coordinator)
- Q1 2016 – John Krug, Lancaster University (UK) / Margaret Briand Wolfe, Boston College (USA)
- Q3 2016 – Margaret Briand Wolfe / Joan Kolarik, The Weizmann Institute of Science (Israel)
- Q2 2020 – Joan Kolarik / Davin Pate, University of Texas at Dallas (USA)
Analytics, the activity of collecting, structuring and combining data into relevant and actionable information for organizations, is more than reporting. Analytics focuses on finding the added value that data, such as assets and their patterns of use, can offer an organization, providing insight into their business and highlighting new opportunities in support of decisions or predictions. As Library processes develop and become more complex, the ability to analyze the resulting data has never been more crucial.
The IGeLU/ELUNA Analytics group is not part of any single Product Working Group because the analytics data is created in multiple applications and some of those applications have their own analytics interface, as well. The group’s cross-product view allows this group to cover broader topics from outside the perspective of any single product. Specific areas of examination include data extraction, consistency, processing and enrichment, visualization and APIs. Integration of other tools, general analytics, data warehousing are also relevant to the group.
A fundamental principle of this group is the Ex Libris users must benefit from openly specified, documented, and understood product data models, analytics data models, and processes for transforming product data into analytics data. Therefore, all issues pertaining to how product data is extracted, enriched, stored, accessed, and used are pertinent topics for the Analytics group. Whenever possible, members of this group also participates in ongoing development and planning.
Tasks and Responsibilities
- maintain a ‘key goals’ document
- maintain an open use cases and scenarios document
- discuss and promote the various uses of analytics in Ex Libris products
- discuss and promote the use of analytics in ‘evidence-based-librarianship’
- help ensure all relevant product data is available to use within the associated analytics products (for example, all relevant Alma data is available to the OBIEE Alma instance)
- help ensure data created using Ex Libris products is readily available, ideally via APIs and through publishing / batch export routines so as to enable the use of special processing, such as institutional data warehouses, big data tools, visualization tools, and programming languages
- encourage and provide some support for groups (geographical or interest-based) to hold analytics skills development days such as those that have taken place in the UK (Alma Analytics) and UNSW for sites in Sydney (Alma Analytics) and through the Ex Libris Northeast Users Group (Tableau).
- maintain close links with the LOD, Interoperability, and Content groups and their communities
- coordinate common problems and enhancement requests for the NERS process
- maintain a repository on Google Drive for sharing information, tips & tricks, discussion, etc
- maintain a mailing list for announcements and discussion
|Member & Role||Site||Location|
|Davin Pate (Coordinator – ELUNA)||University of Texas at Dallas||USA|
|Maya Amichal (Deputy Coordinator – IGeLU)||Weizmann Institute of Science||Israel|
|Amy Allenspach||University of Sydney||Australia|
|Jennifer Bazeley||Yale University||USA|
|Michael Johnson||University of Minnesota||USA|
|Kirsty Lee||State University of New York, New Platz||USA|
|Allison Powers||Harvard University||USA|
|Emily Singley||Boston College||USA|
|Rebecca “Becky” Smith||Memorial University of Newfoundland||Canada|
|Alice Tippit||Northwestern University||USA|
|Sam Tyler||Reading University||UK|
|Tony Wait||Macquarie University||Australia|
|Alma||Leon Krautheusen||Freie Universität Berlin|
|Primo – IGeLU||Paul Harding||University of York, UK|
|Primo – ELUNA||Anna Subramaniam||Purdue University|
|Linked Open Data||Laura Akerman||Emory University|
|Interoperability||Mehmet Celik||KU Leuven|
|IGeLU SC Liaison||Wei Zuan||University of Manitoba|
|ELUNA SC Liaison||Hong Ma||Loyola University Chicago|
A common theme for users of Analytics is that it is difficult to understand, in many circumstances, what exactly the data presented in Analytics products means. Everybody has the experience of not being sure how what is presented in Analytics is derived from what is being presented in Alma/Primo or other source products. The data in Analytics is not directly the data from Alma/Primo, etc. but there has been an ETL (Extract, Transform, Load) from Alma/Primo, etc. to OBIEE Analytics or other analytics products.
We believe there are benefits for all product users in having the product data models, the Analytics data models and the transform process openly specified and understood. It would make it more likely that problems arising will be accurately diagnosed. We believe this also benefits Ex Libris in that Analytics will be better understood, be more effective and users will be able to help find issues and improve the quality of the service that is delivered for the community.
This goal is held in common with the other groups. We entrust Ex Libris with our data, its storage, manipulation and security. It is our data and we wish to maintain and extend access to it whether through batch processes or APIs. The community wants to use this data with external data warehouses and visualization tools, not just within Ex Libris products.
It is our on-going job to collect and analyze evidence that identifies:
- problems/bugs found in the data in Analytics
- incorrect labeling or mapping from products to Analytics
- data not mapped at all from products to Analytics
- difficulty in understanding what is available via Analytics due to incorrect or incomplete documentation
We work to present this evidence in a unified format and persuade Ex Libris to devote resources to address the issues in a timely manner. The majority of the evidence is based on group research, Analytics issues documented in product lists, support cases, and issues raised at ELUNA and IGeLU. Once evidence is been gathered it is presented to the ELUNA and IGeLU steering committees as well as the Product Working Groups for comments and then presented to Ex Libris for discussion and with a goal of getting it on the roadmap towards resolution.
We are also putting together a survey. The survey will be repeated every 2 years to monitor progress and changes on issues over time.
Education and Advocacy
We will work with Ex Libris to improve and extend documentation and training materials within the community about the ETL process between Alma and Primo and Analytics and build a more complete thesaurus of the data available in analytics and where it derives from in Alma and Primo. There is also a need for a common language so that elements are called the same thing in Analytics as they are called in Alma and Primo. It is often hard to find data in Analytics when it has been renamed to something else. There has already been some community generated material and this should be collected, preserved and built upon.
We have started a Discussion List so SIWG members and other interested users can share the work they are doing and post questions, concerns, answers.
Meetings and Communications
A monthly electronic meeting takes place on the third Tuesday of each month.
Open discussion list for Analytics SIWG: firstname.lastname@example.org
All lists and subscription: https://exlibrisusers.org/postorius/lists/