
Fair Isaac Corporation, Inc. (Project Sponsor)
Overview
FICO (Fair Isaac Company) is a decision analytics company with offices around the world. For the last 25 years, FICO has worked with some of the world’s largest credit card companies to screen transactions and highlight suspicious activity using a software product called Falcon. FICO is in the process of developing a new platform for Falcon to provide innovative new features.
The focus of this design project was on improving the experience of how managers and senior level workers are able to view their current fraud strategy is working and take actions to improve that strategy if needed. The fraud strategy at FICO, Inc is comprised of both people and artificially intelligent fraud detection models.
Scope: 4 months
My role:
I independently ideated and designed this dashboard data visualization project.
Janine's Typical Workday
Janine wants to be able to run different scenarios that allow her to quickly respond to changing customer and analyst behavior, as well as respond to changes in results as new data becomes available.
Current Pain Points
• Difficult to get reports to see the “big picture”
• Hard to find the data she needs
• Reports don’t provide visuals to communicate clearly to upper management
Janine's Perfect Day
Instead of having to react to operational challenges in a knee-jerk fashion, she is able to proactively run scenarios that consider likely challenges.
The UX Design Goals:
Goal #1 – Design the information architecture and overall experience of reviewing operational and statistical data.
-
Filtering the data — Customize the dashboard by using prompts such as date range and customer dimensions.
-
Search — Identify items in the database that match the search criteria.
-
Drilling down on the data — Search deeper into the dashboard data by clicking on links embedded in the dashboard content.
-
Print or Export to PDF.
Goal #2 – Introduce a configurable dashboard based on the role of the logged in user, the Team Leader/Fraud Manager: Janine.
-
Analyst performance:
-
Aggregate by team level and by a specific analyst
-
How fast they are working cases
-
How many cases had a status and what that status was- open/active/closed/ and how many per hour/day
-
How fast they are working a queue and which queue
-
Case Type, case level
-
-
Queue and queue performance:
-
How fast the queues are being worked
-
How long they are staying in “active” state before moving to “closed”
-
-
Case level, case type
-
Closed cases fraud/not fraud- confirmed by customer
-
-
Model performance:
-
How well are the models performing (e.g. false positive rate)
-
The Result
1. An object-action conceptual model that communicates a framework of the dashboard's menu organization.
2. An object-attribute conceptual model that lists attributes associated with every object found in the dashboard experience.
3. A prioritization matrix that defined the predicted frequency that users would complete specific tasks on the user interface.
4. A high-fidelity prototype on Axure RP.
