Evaluating, Re-Designing, and Implementing Costing Solutions in Oracle EPM Cloud SaaS across 8 Projects with Global Clients
Oracle EPM / Management Reporting / Data Analysis
Project Snapshot
-
Worked on research, design, and implementation for 8 enterprise finance transformation projects across healthcare, finance, gaming, and multiple Fortune 500 clients.
-
Designed and developed modern Oracle EPM Cloud systems to replace complex legacy costing tools focused on usability, performance, and scale. Conducted user research, prototyping, and role-specific design to reduce friction, boost adoption, and cut process time
-
Awarded 8 awards across my 3 years in the practice due to excellence in my work with these projects and clients.
-
Achieved 60–90% faster costing, 80–90% reduction in rule complexity, and greater stakeholder alignment through iterative build and testing.
Client/Firm
Deloitte
Role
FP&A Consultant
Timeline
2 to 11 months
Team Size
2 to 8
Overview
PROJECT OVERVIEW
During large-scale digital finance transformations, companies often struggle to move away from complex, deeply customized legacy systems and adopt agile, cloud solutions. These legacy systems, often over 10 years old, are tightly woven into the users' day-to-day operations, which makes change challenging. Clients seek our guidance in navigating user resistance, compliance, and aligning teams working across different systems and expectations.
At Deloitte, I worked on 8 such projects across several industries. Each project centered around understanding the client's existing processes and designing and implementing Enterprise Performance Management solutions to modernize key processes like costing, planning, budgeting, and financial reporting.
Specifics of these projects are confidential and cannot be disclosed. However, I have provided a basic overview of work I have done as part of these projects. All assets displayed here-in are generic representations.
CLIENT OVERVIEW
The clients I worked with included one of the largest nonprofit healthcare systems in the US, 2 Fortune 500 companies, a leading UK healthcare provider, as well as a major gaming company. Each came with their own context, business priorities, and legacy infrastructure, giving me the opportunity to work with users and tailor systems that balanced their needs with complex enterprise demands.
MY ROLE
I worked as a cross-functional consultant on teams ranging from 2 to 8 members, collaborating with Senior Consultants, Managers, and Directors. My roles across the projects spanned conducting user research, facilitating workshops as a technical SME, gathering requirements, designing and configuring Oracle EPM systems, driving testing through SIT and UAT sessions, and supporting post-go-live adoption.
TOOLS & PLATFORMS
🛠️
Core Platforms
Oracle EPM
Integrations
Essbase
🧠
User Research
Miro
Figma
Visio
👨💻
Development
Groovy
SQL
Python
✅
Testing
JIRA
Azure DevOps
Sharepoint
Discovery
Problem Statement
How do we design our cloud solution to seamlessly take over from a complex and custom legacy financial system while improving user experience for stakeholders?
WHY IS THIS A PROBLEM?
For finance and reporting teams, legacy systems often mean long, manual processes with steep learning curves for new staff. Users often have to rely on IT to make even small changes or fix errors, which can be especially painful during time-sensitive periods like month-end close. The legacy tools were technical, cluttered, and not built with user experience in mind.
At the same time, these systems had been around for years. To maintain continuity of user experience, maintaining familiarity with existing costing processes was vital, which made it very important to have a meticulous understanding of user journeys and incorporate it into thoughtful experience design. Based on user feedback, "Clicks, not code" became a fundamental motto for us as we started the process on these projects.
3-5
Days to Process
Costing processes often took several days to complete due to manual reconciliations, fragmented tools, and delays in validation.
12+
Manual File Loads
For most clients, each cycle required over a dozen manual data loads , increasing the risk of errors and delaying processing.
$100k+
in Estimated Delays
Clients consistently flagged, the slow legacy process leading to delayed reporting, missed charges, or revenue leakage
RESEARCH
We approached discovery through two parallel tracks: user-centered research and system-centered analysis. On the user side, we conducted several avenues of primary and secondary research, such as exploratory sessions, workshops, and process walkthroughs with stakeholders directly involved in the costing process. These helped us understand day-to-day workflows, pain points, and expectations from the new system.
Primary Research
👥
User Workshop
Sessions with users to understand workflows, pain points, and costing needs
🔄
System Workflow
Tracing current processes to map costing logic and user-system interactions stepwise
📓
Documentation
Reviewing documentation, SOPs, and training guides for background and structure
💻
Code / Queries
Analyzing SQL rules and legacy scripts to extract embedded planning and costing logic
In parallel, we carried out a technical deep dive into the legacy platform. This included reviewing business rules, SQL queries, reports, and available documentation. To supplement this, we referenced internal SMEs, Oracle’s knowledge base, and domain standards to evaluate best practices and identify opportunities for system improvement.
Secondary Research
🌍
User Workshop
Research through Oracle's knowledge base regarding implementation
🧠
Internal SME
Insights from functional leads and prior implementations in the industry
📚
Documentation
Industry and compliance references relevant to finance/reporting
🔎
Benchmarking
General benchmarking from public case studies and UX examples
Synthesizing findings from both tracks allowed us to map current-state user journeys, define the system's functional gaps, and translate real-world needs into actionable, technically feasible requirements. This foundation informed both our design strategy and the implementation roadmap.
Key Insights
1
Key Insight 01
Legacy systems are expensive, have high op-ex and cap-ex costs and often require dedicated infrastructure and operational IT staff
2
Key Insight 02
Users disliked the time-taking process of having to contact IT to debug the code-heavy system, especially when they were under a time-crunch to complete month-end financial close activities
3
Key Insight 03
The UI was confusing and too technical for majority of non-technical users and they required help performing simple tasks such as extract reports
4
Key Insight 04
Due to the complicated code-based nature of the Costing system, performing what-if analysis and testing out alternative methodologies was a time and effort-consuming process
5
Key Insight 05
Complicated calculation logic often required hundreds or thousands of individual rules to execute
REFINING REQUIREMENTS
Using the insights gathered through primary and secondary research, we created a comprehensive set of detailed requirements and user journey maps for key user roles across the costing process. These artifacts were accompanied by flow diagrams, data mapping sheets, and system integration models to visualize the end-to-end user experience and technical dependencies.
The core needs identified from this became the basis of the final requirement documentation which typically included hundreds of individual functional and non-functional requirements, organized by user role and system module. Each requirement was validated through stakeholder review sessions and translated into user stories within our delivery framework. These stories then became the basis for wireframes, test scenarios, configuration logic, and sprint-level development planning. I was involved in drafting several of these as a Consultant with a wide range of skills in design, development, project management and planning.
Core Needs
1
Cloud Based System
A cloud based SaaS solution which would lower expenses and eliminate the need for companies to purchase software and hardware or hire additional IT staff with no costly infrastructure to support.
2
Clicks not Code
The solution must have user-friendliness at its core and make use of out-of-the-box Oracle EPM wherever possible to allow users to run the costing process, troubleshoot simple data issues, and configure rules without requiring additional IT support.
3
Role Specific UI
The solution must have user-friendliness at its core and make use of out-of-the-box Oracle EPM wherever possible to allow users to run the costing process, troubleshoot simple data issues, and configure rules without requiring additional IT support.
Design
PROCESS MAPPING
We created Visio-based system flows to map current-state costing processes, identifying users, touchpoints, and data movement as well as parallelly identifying approaches to implement them in the newer system. These diagrams helped surface inefficiencies, misaligned responsibilities, and where manual workarounds were baked into the system. They also created a shared foundation for collaboration across business and technical teams.
ROLE BASED UX DESIGN
One of the clearest insights from our research was that different user groups had vastly different needs. A costing SME needed access to allocation rules and logic. A financial analyst mostly cared about accurate reporting. End users wanted simplicity. Trying to serve all of them with the same interface would have defeated the purpose of the modernization of the legacy system.
Grouping our users into key user groups represented by personas, we used Oracle EPM’s navigation flow capabilities to design multiple UIs tailored to specific roles. This meant users only saw the modules, integrations, reports, and tools relevant to their workflow, reducing times on task and helping them move faster. For example, costing SMEs had quick access to model configuration, while end users were guided straight to dashboards and exports.
AFFINITY MAPPING & PRIORITIZATION
After collecting qualitative feedback from early sessions, we organized recurring themes using affinity mapping. This helped us prioritize what mattered most to users and align it with technical feasibility. It directly fed into sprint planning by identifying which user stories should go into early builds versus future iterations. It also helped frame design decisions in a user-centered way, even when stakeholder priorities differed.
PROTOTYPING & VALIDATION
Our users, who were mostly finance professionals, were deeply familiar with spreadsheets. To meet them where they were, we created our mid-fidelity prototypes in Excel. These included layout structures, data flows, and even working formulas. This approach made testing intuitive. Users could explore the prototype using tools they already trusted, which accelerated validation and improved feedback quality.
Once we had buy-in on layout and logic, we recreated the experience inside Oracle EPM’s demo environment. These high-fidelity prototypes used real navigation flows, forms, and rule structures, simulating the complete experience from data entry to reporting.
We conducted live walkthroughs with stakeholders, power users, and directors, which helped align expectations early. Seeing the system in action gave teams confidence and turned abstract ideas into tangible workflows and significantly reduced the risk of last-minute surprises
Once prototypes were in place, we ran demo sessions with stakeholders, super users, and directors to walk them through the full solution. These walkthroughs gave stakeholders confidence that what we were building was practical, usable, and aligned with system capabilities. It also reduced the number of surprises during the actual implementation.
Usability Testing & System Build
SYSTEM BUILD & CONFIGURATION
After validating the design through prototypes and walkthroughs, we moved into implementation within Oracle EPCM. I worked on configuring rules, building allocation models, integrating source data, and designing validation logic to support seamless operations.
To handle complexity, we used Groovy scripting and structured templates, which allowed us to reduce redundancies and keep the system adaptable to future changes. We built iteratively, where every round of feedback helped us refine logic, solve edge cases, and make sure what we built truly matched how users worked.
USABILITY TESTING
After the initial build, we conducted several rounds of usability testing to validate the system with real users. The goal was to ensure that the costing process not only functioned as expected, but also aligned with how users naturally worked, especially during high-pressure month-end close periods.
Usability Testing Cycle
1
Test Scripts
Users followed task-based scripts simulating their real-world responsibilities, including running models, validating outputs, and troubleshooting data issues.
2
Participants
We started with neutral test users unfamiliar with the system to surface navigation or usability issues. We then tested with actual end users and Costing SMEs to validate logic and uncover role-specific feedback.
3
Environments
All testing was done in a secure lower/beta environment, mirroring the final setup without affecting live operations.
Common issues that surfaced included unexpected data movement and formula discrepancies, which were iteratively addressed. Feedback cycles helped us simplify language, restructure certain screens, and adjust logic order for better user control.
Outcomes
60-90%
faster costing process
Users reported that month-end costing, which previously took multiple days, could now be completed in a fraction of the time. This was due to automated integrations, role-based UI design, and the ability to schedule calculations, validations, and data loads.
~80%
reduction in Rules
By simplifying logic and combining redundant rules, we reduced the number of costing rules significantly, in some cases by over 90%. This led to easier maintenance, improved model performance, and greater flexibility for users testing different scenarios.
These projects shaped my understanding of what effective enterprise UX looks like. I learned that familiarity is not a constraint. It is a bridge. Prototyping in Excel, while unconventional, made users more comfortable, sped up validation, and helped us build trust early. It also showed me how prototypes can go beyond testing. At their core prototypes are alignment tools. When stakeholders see their workflows brought to life, it unlocks better conversations and faster decisions. Most of all, I realized that enterprise design is about balance between the new and the known. That mindset continues to guide me in every project I take on.
