ARE WE SOLVING THE RIGHT PROBLEM?

GE | Strategy | UX

GE Capital’s VP of Human Resources wanted to make opening job postings more efficient. Feedback from HR teams showed that it was difficult to write, track, post, and compare job descriptions, and they often have to do tasks manually.

I led the UX strategy and design to improve the design of legacy Oracle systems. I identified that poor data integrity was the roadblock to improving the user experience. I saved GE over $250,000 when I successfully convinced the VP to make a strategic pivot to prioritize data clean up.

To first understand the pain points, I interviewed 70 HR employees around the world and conducted 20 usability tests on the existing systems. The goal was to identify easy, “quick win” fixes and longer term issues to tackle.

I discovered that complex HR databases did not talk to each other, resulting in errors. Users did not trust the data and each team wrote cheat sheets on paper for themselves. Users had no insight into the overall process — they had no idea whether their new job was posted successfully or who was next in the approval process. This resulted in multiple delays.

I used affinity mapping and grouped all of the user research into themes, like "clear instructions", "filter and search", "improve data inputs", and "connect HR processes". I led a visioning session with 10 in-person and remote stakeholders to prioritize themes against business requirements. This defined the product roadmap, beginning with minimum viable product (MVP). 

Strategy RESULTS

I conducted user research and showed that poor data integrity and other technical issues were the root causes of the poor user experience. I aligned with a cross-department team on a strategy: First address quick wins while investigating how to improve the underlying data and technical issues. I created a 2-year roadmap, which was presented to the VP of HR.

When writing a new job description, users often search for an existing job and edit its description. In the new design, the user can search by keywords using more natural language. They can also see results according to departments. I also designed a way to see results as both a table and a list with an excerpt. This helps users to quickly identify the right description. A compare function with grouped descriptions also alleviates questions of whether one job description is equivalent to a similar one in a different department.

Since job descriptions in a job family (e.g. developer, accountant) are similar, another concept is to store often-repeated phrases so that the user can simply select and edit. Once the user is done editing, they can track the approval and posting of the job using a dashboard and identify when they need to take action.

OVERALL RESULTS

The design concepts inspired the team to think broader than simple fixes and reframed why tackling data integrity was fundamental in ensuring a successful user experience.

I saved the company over $250,000 in my recommendation to strategically prioritize data cleanup.