HIGHLIGHT
AI-Powered Finance & Campaign Management Tool - Lead UX Designer
Cut management time by 76%. Reduced error 30%. Approval of invoices 60% faster.
Problem
Finance teams were spending 5 days on average to approve invoices, with high error rates and frequent support tickets. Campaign management required extensive manual oversight, consuming significant time. The existing workflow lacked AI-powered automation, clear status visibility, and efficient decision-making tools. How might we leverage AI to redesign the invoice approval and campaign management processes to reduce processing time while improving accuracy?
What I did
- •
Designed an AI-powered campaign management tool that cut management time by 76% through intelligent automation and predictive insights
- •
Redesigned the end-to-end invoice workflow from submission to approval, reducing processing time from 5 days to 1 day
- •
Implemented AI-driven automated validation and smart routing that reduced manual review time by 60%
- •
Created AI-powered campaign timeline visualization with predictive analytics for campaign performance and fee management
- •
Designed intelligent status indicators and progress tracking that improved visibility across approval chains and campaign lifecycles
- •
Built streamlined decision-making interfaces with AI-generated contextual information and one-click approval actions
- •
Established AI-enhanced error prevention patterns including automated validation checks, confirmation steps, and audit trails
- •
Conducted user research with finance teams to understand pain points and optimize workflows with AI capabilities
My key contribution
Owned the end-to-end redesign of the AI-powered finance and campaign management tool, reducing campaign management time by 76% and invoice processing time by 80% while improving accuracy and user satisfaction.
Highlight:
I designed an AI-powered campaign management tool that cut management time by 76% and redesigned the invoice workflow from the ground up, implementing AI-driven automated validation and smart routing that reduced manual review time by 60%. By designing intelligent status indicators and streamlined decision-making interfaces with AI-generated insights, I enabled finance teams to process invoices 5× faster while reducing errors by 30%.
Results & Impact
- ✅
76% reduction in campaign management time through AI-powered automation
- ✅
80% reduction in invoice approval time (5 days → 1 day)
- ✅
30% reduction in invoice processing errors
- ✅
60% decrease in pending invoices
- ✅
45% improvement in user satisfaction scores
- ✅
Significant drop in support tickets related to invoice processing and campaign management
Overview
The Finance department at Booking.com processes thousands of invoices monthly across multiple teams and approval chains, while managing complex partner campaigns and fee structures. Finance teams were spending 5 days on average to approve invoices, with high error rates and frequent support tickets creating operational bottlenecks. Campaign management required extensive manual oversight, consuming significant time and resources. The existing workflow lacked AI-powered automation, clear status visibility, and efficient decision-making tools, leading to delays in vendor payments and increased manual work. In a fast-paced environment where timely invoice processing and campaign management directly impact vendor relationships and cash flow, the department needed an AI-driven solution that could handle high volume while reducing errors and processing time.
To comply with my non-disclosure agreement, I have omitted and obfuscated confidential information in this case study. All information in this case study is my own and does not necessarily reflect the views of Booking.com.
How might we leverage AI to redesign the invoice approval and campaign management processes to reduce processing time while improving accuracy?
How I structured the problem space to guide design decisions
- •
speed through AI-powered automation and intelligent routing
- •
accuracy through AI-driven validation and error prevention
- •
visibility through clear status indicators and progress tracking
Using user research and workflow analysis, I identified that the key was leveraging AI and machine learning to automate routine validations and campaign management tasks while providing clear decision-making tools for exceptions, rather than requiring manual review of every invoice and campaign.
Information Architecture
Content coming soon...
Designing modular, scalable components that integrate across the product ecosystem
Approach
Benefits
- •Reduces technical debt through reusable components
- •Enables faster iterations and scalability
- •Creates enterprise-scale solutions
Navigating complexity through cross-functional collaboration
I worked closely with finance teams to understand their workflows and pain points, with product managers to prioritize automation features, and with engineers to design validation rules and routing logic. I facilitated user sessions to gather feedback on existing tools and observed user behavior to identify optimization opportunities. Regular syncs with stakeholders ensured the redesign met business requirements while improving user experience.
Challenges and trade-offs
This project required balancing automation with flexibility, ensuring accuracy while maintaining speed.
Key Challenges and Solutions
Complex Approval Chains
Different invoices required different approval paths based on amount, department, and other factors. I designed a flexible routing system with clear rules and exceptions handling, enabling automated routing for standard cases while allowing manual override when needed.
Error Prevention Without Slowing Down
We needed to prevent errors without adding friction to the approval process. I implemented smart validation that ran in the background, surfacing issues proactively with clear explanations and suggested fixes, rather than blocking users with error messages.
Visibility Across Large Teams
Finance teams needed to see invoice status across multiple approvers and departments. I designed a unified dashboard with filtering and search capabilities that provided visibility without overwhelming users with information.
What I learnt
Redesigning the invoice workflow taught me that automation and user experience go hand in hand—automation shouldn't mean removing human judgment, but rather enabling it with better tools and information. By designing clear status indicators and contextual decision-making interfaces, we enabled users to make faster, more accurate decisions. The key was understanding that workflow design is about reducing cognitive load and eliminating friction, not just adding features. This experience reinforced the importance of user research in enterprise tools, where understanding actual workflows and pain points is crucial for meaningful improvements.
Feedback
Software Engineer
"Julia greatly improved the Dynamic Contract Override project by consistently operating at a senior level of ownership and strategic execution. She demonstrated proactive stakeholder management by organizing user sessions to gather insights and leading detailed discussions to ensure full team alignment on DCO features. She established a strategic, multi-step iterative feedback process, moving from collecting feedback on existing tools to observing user behavior on the new setup. This methodical approach led to crucial strategic insights, specifically identifying the difference between the override and stack concepts and the fundamental shift to purpose-based contracts—knowledge that will greatly improve user understanding. Julia took ownership of technical clarity by preparing a comprehensive design document that served as a valuable guide. Her commitment to cross-functional excellence was evident in sharing the final draft for collective review to minimize confusion. Furthermore, she provided critical team enablement by always being available to assist with design needs, directly helping the team move faster in UI development."