Automating Budgeting Workflows with AI
Capstone Project
— SEP 2025 - PRESENT —
OVERVIEW
—
PFF, LLC is a consulting firm that develops tools to help nonprofits, local governments, and small agencies manage their budgeting and financial planning more effectively.
InCEP Platform
The consulting firm partners with the InCEP (Integrated Capital and Expenditure Planning) platform, which allows organizations to plan, forecast, and report budgets using data-driven insights.
Emergency Missions
Searching Mission Assignments
Creating New Mission Assignments
Creating Cost Estimates
The InCEP platform presents a clunky and time-consuming experience, creating an opportunity for AI-driven automation.
EMPATHIZE
—
Competitive Analysis
We started by conducting a competitive analysis to understand existing solutions in the space and how InCEP compares to them. For this, we analyzed popular budget forecasting platforms, including Euna Solutions, OneStream, OpenGov, and Cube Software.
Some insights include:
Competitors like Cube and OneStream are moving toward purpose-built AI agents that handle specific financial tasks, not generic chatbots. There is an opportunity to eliminate the manual burden of the legacy InCEP platform rather than simply adding an AI layer on top of it.
OneStream, despite being the most feature-rich platform, suffers from steep learning curves. If the tool requires extensive training, it will face adoption resistance.
Across all competitors, the strongest dashboard patterns share top-level KPIs at a glance with drill-down detail below.
No AI-powered budgeting tool exists specifically for emergency management organizations.
Understanding InCEP Workflow
Led call-centre agent interviews using a custom Excel guide outlining key steps across six core workflows
Past Interviews
Interviewing end users to assess RC perception and understand user journey
DEFINE
—
User Personas
Affinity mapping to extract user behaviors and pain points
User Stories
IDEATE
—
Design Goals
Reduce Visual Clutter
How might we create a clearer, more focused homepage that helps users find what they need?
Simplify Navigation
How might we help users locate licensing checklists and key resources faster?
Improve Search
How might we help users find specific information without feeling overwhelmed?
Clarify Language
How might we make NMLS terminology accessible to first-time small business users?
Keep Content Current
How might we ensure users always have access to accurate, up-to-date information?
AI Use Case Selection
How might we create a clearer, more focused homepage that helps users find what they need?
Task-Based Information Architecture
Removed competing content sections like NMLS News, Agency News, and Popular Links.
Persistent sidebar navigation.
Reduced cognitive load and resources organized by user type and task.
How might we help users locate licensing checklists and key resources faster?
Elevated Checklist Visibility & Persistent Navigation
Checklists moved from buried links to primary placement on homepage.
Breadcrumb wayfinding allows users to see their location and navigate back easily.
Multiple access points.
How might we help users find specific information without feeling overwhelmed?
Global Persistent Search
Persistent, prominent search bar accessible from every page.
Contextual search within sections allows users to search within their current area for more relevant results.
How might we make NMLS terminology accessible to first-time small business users?
Integrated Glossary Access
NMLS Glossary surfaced on homepage as well as internal sections, previously buried deep in documentation.
Glossary sits alongside related policy content, so users can reference terms while completing tasks.
How might we ensure users always have access to accurate, up-to-date information?
Dynamic Content Management
Replaced static PDFs with XML-based content.
Content editors can push changes instantly without regenerating and re-uploading entire documents.
Broken links, duplicates, and outdated pages are gone.
PROTOTYPE
—
Design System
REFLECTION
—
Impact
Visibility
Synthesized and presented user research to VPs and Product Managers, bridging the gap between user insights and strategic product decisions.
Validation
Socialized the redesigned Resource Center with call center teams (the same users who informed our research) resulting in strong positive reception.
UX perception
Helped establish UX credibility within the business unit, showcasing how research-driven design leads to measurable product improvements.
Impact Stats
50%
reduction in navigation time across 10 common searches
~ 600 K
industry users impacted (state and federal Resource Center)
6
journey maps delivered, covering the most critical licensing workflows
Reflection
I completed my internship in August 2025 and was extended through the Fall semester. Working at the intersection of financial regulation and UX has taught me more than I can easily summarize, but here are the key takeaways:
Empathy is key: Conducting user interviews revealed just how much is at stake in this industry. Every piece of instructional text, every button placement carries weight that can cost people their livelihoods. Designing with user empathy has never felt more essential.
Data drives buy in, but stories create connection: Senior stakeholders need the numbers, user counts, error rates, conversion metrics. But user stories are what make those numbers resonate. Learning to weave both qualitative and quantitative methods into my journey maps has been one of the most rewarding parts of this work.
Pivot often: Real-world projects are rarely linear. We started with a vague goal of improving self-service for small company users, paused, and shifted to understanding their pain points. That detour uncovered insights that shaped key decisions and informed tree testing for the new NMLS Resource Center information architecture.
Small discoveries snowball: This research also helped us improve support for call center agents. A deep dive into ServiceNow call center data revealed untapped potential for UX improvements we hadn't anticipated.
