Redesigned DataCite's Harvesting Services from the ground up.
DataCite existing Harvester Services are currently underutilized. Given the fact that users of such services are typically willing to pay for them, we recognized an untapped revenue opportunity if we could boost service utilization.
We initiated the project with a focus group study involving representatives from 10 major organizations. This exploratory approach allowed us to understand their needs in terms of harvesting services and their willingness to pay for such services.
A design sprint was subsequently conducted to generate a comprehensive initial prototype for the new harvesting service. This sprint involved around 7 internal collaborators from different departments and was facilitated virtually using Miro due to the pandemic. The aim was to ensure that the service redesign would meet the needs of all organizational corners. To foster a long-term vision, we slightly altered the design sprint's activity order, positioning the mapping activity as a secondary step.
I prepared a bespoke questionnaire to guide participants during the "expert interviews" activity of the design sprint, which led to 83 "how might we" notes. Over the next three days, I guided the participants through each stage of the design sprint, focusing significant effort on the mapping exercise and sketching day preparation.
Two top-voted sketches were selected for further prototyping in Figma, followed by an Expert Walkthrough validation with eight users to collect further feedback.
Both XML and JSON need to be provided for different use cases
Monthly and daily harvesting are of the essence
Most cases will need only metadata but citation data could packaged separately
A tiered business model is possible with a Premium service for for-profits
Tight integration with the DataCite CRM will be needed
For-profits would prefer snapshots due to the high demands on frequency
1. Define a tiered approach for the harvesting service, including a basic service for users requiring only citation metadata and a premium service for others.
2. Modify the tech stack and the user profile service to accommodate the new premium tier users.
3. Scale the infrastructure to address the increased needs for higher frequency harvesting.
The findings also scoped the design work sufficiently to start working on the MVP earlier and faster
The implementation of these recommendations led to the establishment of relationships with four potential paying users of the harvesting service.
The introduction of the snapshot feature allowed dozens of users to take advantage of the service right away.
When we were involved in the design sprint for the DataCite harvesting service, the "How Might We" question technique supported creativity while focusing on the problem to solve for the user.
-- Britta Dreyer, Head of PID and Metadata Services, Technische Informationsbibliothek (TIB)
