Interactive Fraud Analysis Tool
Sector
Finance
Duration
6 Months
My Role
UX Researcher & Designer
The Challange
Develop an interactive visualization tool to empower investigators in uncovering fraud, transitioning from an existing pen-and-paper process to a digital platform.
!
Due to the highly confidential and classified nature of this project, I have omitted client-related information and had to obscure some of the tool's functionalities, which prevents me from showcasing the highest quality and end results of the design.
01
Problem Definition
The existing pen-and-paper process used by investigators to uncover fraud was time-consuming and inefficient, making it difficult to identify connections and relationships between data nodes.
There was a need to develop a digital, interactive visualization tool that could streamline the investigative process, allowing investigators to see the big picture while easily zooming in on specific data points for detailed analysis. The tool needed to enhance the efficiency of investigators by enabling them to quickly identify and save critical connections, as well as facilitating the review of past investigations.
Additionally, the challenge included ensuring investigator satisfaction by transitioning them smoothly to the new platform and providing adequate training to onboard them successfully.
02
Discovery Phase
-
Conducted client workshops to explore investigative methods, both physical and digital tools, business objectives, and success metrics.
-
Facilitated brainstorming sessions with stakeholders and users to generate ideas and align on goals.
-
Performed user interviews to identify key needs and pain points.
-
Conducted competitive research and analysis to understand industry standards and opportunities.
-
Reframed the challenge based on research insights to ensure a user-centered approach.
03
Ideation
-
Defined layout and information architecture, prioritizing users' needs and motivations.
-
Drafted multiple user flows, including the optimal path, error scenarios, and edge cases.
-
Created paper prototypes and conducted user testing to refine and validate concepts.
-
Leveraged the bank's existing design system to save time and resources in both design and development by utilizing pre-existing components and patterns.
04
Design
-
Designed wireframes and a comprehensive sitemap to establish the foundational structure and flow of the tool.
-
Validated wireframes and sitemap with stakeholders, ensuring intuitive usability and optimizing the user experience for efficiency and time-saving.
-
Refined wireframes and sitemap based on stakeholder feedback, continuously improving the design for clarity and functionality.
-
Collaborated closely with developers and data analysts to align the design with the platform's technical constraints and underlying data structure.
-
Developed a high-fidelity, clickable prototype to define and test navigation, page interactions, and component behavior, ensuring a seamless user experience.
Detailed Design Decisions & Screen Functionality
Search Function
At first, the search function was made to only initiate the investigation for one type of data. This was the developers and business analysts assumption. After user interviews it became clear that each investigator (user) has their own unique way of working and in order for them to adopt the tool we have to facilitate that.
Interactive Node Exploration
Clicking on an icon reveals the full information of that data node in a slide-in window on the left side of the screen, causing the entire network to shift right to remain fully visible—a feature implemented based on user interview insights. The selected node changes color to highlight it within the network. Additionally, zoom in/out buttons allow users to navigate between different levels of relationships, from 1st degree to 3rd degree, offering more control over the network's depth and detail.
Network Visualization
The interactive fraud detection tool visually represents complex information systems and social networks, using arrows as nodes to indicate relationships and icons to represent different types of edges. The direction of each arrow shows the direction of the relationship. Each edge type has a distinct icon, helping investigators quickly understand the network's structure. Users can click on any arrow to reveal more information and add additional layers to the network. The thickness of the arrows reflects the volume or strength of the relationship, with labels providing specific details about the connections.
Visualization of Relationship Dynamics
Clicking on a relation brings up a popup window with detailed information specific to that relationship. The distinction between edges and relations lies in the level of detail stored in these nodes and user preferences. User interviews revealed that while edges are the primary focus of investigation, users spend most of their time exploring relations. When examining an edge, they want to maintain a 'big-picture' view, but when diving into a relation, they prefer to concentrate solely on that detail. To accommodate this, users can customize the information displayed, and the popup window includes a visualization of the specific relation to maintain context.
05
Delivery
-
Finalized the layout and information architecture with a strong focus on user needs and motivations.
-
Developed comprehensive user flows, covering the optimal path as well as potential error scenarios and edge cases.
-
Produced paper prototypes and conducted user testing to refine and validate the design concepts before final implementation.
06
Outcome
-
Delivered a clean, intuitive, and interactive tool that exceeded the expectations of both stakeholders and users.
-
Successfully reduced investigators' work time by approximately 30%, while also significantly cutting down on the use of physical resources like paper and pens.
-
The final product featured a user-friendly interface, streamlined navigation, and robust data visualization capabilities, making it easier for investigators to uncover fraud and review past investigations.
-
Achieved high user adoption rates, thanks to the tool's ability to cater to a wide range of user needs and preferences.
07
Learnings
-
Gained valuable experience in data visualization, enhancing the ability to translate complex data into clear, actionable insights.
-
Recognized the importance of continuous user engagement throughout the design and development process to ensure the final product meets diverse user requirements.
-
Reaffirmed that a focus on user-centered design and thorough validation can significantly improve the adoption and success of new digital tools.
08
Feedback
Jochem, Product Owner
"In this data-science focused project Miriam succeeded in translating complicated requests into simple designs. She created a detailed user map and creative solutions to facilitate different user requirements. Her very good communication style leads to trust with both team members and investigators, and enables us to move fast."
Rosa, end user
Hi Miriam,
The investigation tool look very nice, Michael and I think that they would make our job much faster and easier!