Case study: Expanding functionality to an evolving AI data platform
―Enterpriseㅤ―Product and UXㅤ―Web
IBM’s Cloud Pak for Data (CPD) is an AI-powered data analytics product that aims to transform data and reduce complexity by having everything you need in one convenient platform. Out of the box, CPD is already powerful, being able to connect to data with built-in governance, automate workflows, and having solutions already packaged with the product. However, where CPD truly shines is in its capability to expand its features and integrate with native and third-party services via the services catalog.
Recently, IBM has been making a big push towards putting design first with their initiatives, like IBM design thinking and their Carbon design system. With this shift, products across all of IBM has been making a move to update its then current system to Carbon, including Cloud Pak for Data. This meant all of the associated products like the product I was working on, the services catalog, needed to make that change too.
The overall process of this project was fairly iterative, as we were always looking back at customer and stakeholder needs and regularly updating the product as needs evolved.
Audit and user definition
My teammate and I started with an audit of the existing product, as what was implemented at that time was created without design resources. During our auditing phase, we wanted to review the users journey through the current experience. We ran design exercises, reached out to our stakeholders, went through documentation, and interviewed internal users (data scientists) to highlight the needs and pain points of our user.
Through this collaboration, we were able to identify where we could make improvements as well as opportunities for new features to enhance the user experience.
Breaking it down to build it back up
The catalog has two main parts: the tiles, consisting of the services and its contents, and the interface, consisting of features that assist the user in finding the products that they’re looking for. While conducting user interviews, we came across an essential insight:
the catalog was a toolbox, and it was our job to bridge the gap between the problem and the solution.
This statement would become the basis which our solutions would revolve around.
Mapping and scoping
After iterating on potential solutions and features, we wanted to properly scope the project given our timeline and set our priorities.
“How valuable are these features and when do our users need them?”
“How do these solutions fit into the larger scope of the product? How will it grow with the product?”
“What will the catalog look like when we are done? How about after that?”
A big idea, but more like a north star
During our ideation phase, we wanted to set a vision for what the services catalog could be, given that there weren’t certain dependencies or constraints. This was an opportunity to explore different solutions outside of what was deemed possible in our situation, from which we could scale back to fit in the scope of our project. Many of the ideas we formed here would be seen in the final product.
Implementation and solutions
