Blind spots, gaps, and unexpected traffic: A (brief) history of the transition to Google Tag Manager and new approaches to improved MD-SOAR analytics

Abstract

The transition to Google Tag Manager (GTM) seemed like an obvious choice for the MD-SOAR shared institutional repository: easier management of custom analytics tags, immediate integration with existing DSpace and Google Analytics infrastructure, and easy customization translating to better item-level statistics for participating campuses. However, now - more than a year post implementation - numerous issues with the GTM approach have been observed. In addition to breaking existing custom development that was implemented for recent DSpace releases, GTM was revealed to miscount metrics related to bitstream downloads and inbound web traffic from indexed search engines.

This poster will provide an overview of problems observed with the Google Tag Manager implementation for the MD-SOAR DSpace instance and the custom tag development necessary to adequately address these problems. Furthermore, this poster will provide an overview of a newly formed partnership with the RAMP initiative headquartered at the University of Montana, a collaboration that hopes to support the development of a novel analytics approach that more accurately reflects platform and bitstream use. A comparatively assessment of gathered metrics will be provided.