z Operational Insights: Fine-tuning the mainframe

In 2015, I joined IBM as a UX designer for z Systems mainframe IT service management sector. As a very large, 100-year-old tech company, IBM,  was radically changing from the inside out.

The challenge and problem space

In the past ten years, large enterprises such as airlines and financial banks have experienced intense pressure to modernize their mainframes to adopt cloud service models and graphic interfaces. For decades many mainframe system programmers and admins used 3270 green screens to manage their environments. These tools were cumbersome to use and required deep domain knowledge of CLI commands and troubleshooting through runbooks. As well, many companies needed to overcome a widening age and skill gap. 

IBM and its mainframe were caught between its legacy in engineering and the need for a more human-centric approach to how products were researched, developed, and built. 

The solution

My product, z Operational Insights, was z Systems first Cloud SaaS offering designed to address the pressures the industry faced by turning log data into actionable, understandable reports for users of all skills.

Designing for multiple learning curves

Throughout user interviews, I repeatedly heard people expressing frustration on two pieces:

1. Combing through 3270 screes to understand and pinpoint where problems occurred

2. Having to consult a runbook to determine which actions to take.

The insight reports needed to address these concerns. I wanted to strike a good balance between data visualizations showing problem areas, the text clearly explaining in layman terms what was happening, and providing clear takeaways to improve performance. So, I pitched the idea of using a  “narrative style” report to improve the clarity, quality, and usability of the reports. I lead the visual design direction and overall experience of zOI.

Using simple, clear, and concise language

Throughout my design process, I took into consideration the different domain knowledge and learning curves of our users. Working with a content designer, we kept explanations short but informative and paired them with data visualizations to highlight where problem areas occurred.

Each inside report included a scannable list of recommended actions a user could take along with additional resources to consult.

Building a component a library

At the time z Operational Insights was developed, there was no centralized design system and many products in z Systems were using outdated frameworks. So, I designed and built a small library of reusable components with my dev team throughout the design and development process.

Establishing trust through branding

Along with the narrative style, I spearheaded the zOI's branding direction to be based on the geometric panels of a z13 mainframe. This work became a grassroots project I lead with a few other visual designs in IBM Z. More can be found in my IBM Z - Reimagining the mainframe case study.