Credit Risk Digital
A case study of complex app UX optimization
Credit Risk Digital (CRD) is an enterprise software which digitalizes the workflow of credit risk analysis for credit analyst.
In this project, I lead UX design to support the first release, identify UX issues, conceptualise and design new features for the second release. I collaborated directly with the product head and worked closely with internal risk experts to shape the product and successfully increased the usability.
Learn about user's non-linear, complex activities, design new features to adapt and optimize the current product.
Credit Risk Digital(CRD) is a digitalized support corporate and dealer business.
Unlike generalist Apps for everyday domains, CRD is a complex app that supports specialized users in carrying out broad, unstructured goals in nonlinear workflows.
Learn about analysts’ work processes and sort out their workflow by conducting an internal expert interview.
Credit risk analysis is very specialized work. To get the full picture of the user activity & user behavior, I conducted a few expert interviews. By asking them questions & observing them using the current application for some tasks, I gathered some learning, findings, and identified some issues.
A global credit application consists of several local applications, each of them consists of several decisions from dozens of deciders.
The local credit applications process is supported by the current platform. However, global applications is a lot more complicated than the local ones because local ones are only one of many steps in global applications(fig. 1 &2) to make a global application possible, some extra functionalities should be adapted.
1. Losing track due to long task flow
Users often face long waits and frequent interruptions to their work. Complex data analyses can run for hours, if not days, Even without unplanned interruptions, the complexity and variability in subtasks executed to achieve a goal requires users to hold a lot of information in working memory during tasks, which can be easily lost and causes low efficiency.
"Sometimes it’s difficult for me to memorize where I left after returning to CRD."
- Analyst A
2. Deficiency of primary dashboard
In the current version of CRD, there are two categories of customers, but only one category is accessible on the primary dashboard. Besides, I found that although the progress monitoring of CA is needed, the only progress bar is hidden in the group dashboard, which is two layers away from the primary dashboard.
"I wish that for some status checking I don't have to go all the way to the application page just for some update."
- Analyst B
With the help of the expert interviews, the root cause of the UX problem was identified.
The current issues are identified as follow:
Due to the constraints, the old information architecture must be kept. To optimize the usability, I decide to focus on enhancing the personal dashboard, to make the pages in a deeper layer more accessible.
Before
In current CRD main dashboard, the widgets are limited to only three and the customer widget only contain customers in one level, through there are two customer levels at the moment.
After
After adjust the IA, I made customers in all level easier to access from customer widget, besides, I introduced credit application progress status widget, which helps the user to monitor the application status without going all the way through application board.
Instead of having only local customers in the widget, the new design allow customers and customer groups in all level accessble, which save analyst many time for identify and check customer relationships.
Credit application widget made all the credit application which is relevant to the user transparent and up to date.
When several tabs in different category open, new tab design provides better visual hierachy for tabs from different levels.
Too many open tabs causes high cognitive load and makes certain pages hard to find. To solve the issue I reduced the number of opened tabs by opening customer files within the same customer.
Key screen findability significantly increased, analysts' workflow was simplified and work efficiency increased for all test participants.