ACCURATE

A Data Reconciliation Platform For Public Health Reporting
ACCURATE is a data reconciliation platform designed to help public health teams identify and resolve discrepancies between CDC and STLT case data. I led UX from discovery through MVP, translating fragmented systems and workflows into a structured, scalable product experience.

Role
Senior UX / Service Designer

Scope
Discovery → MVP

Environment
CDC, public health data systems

Product Walkthrough
A short walkthrough of the core reconciliation experience, including report setup, data comparison, and discrepancy review.

This walkthrough highlights the end-to-end workflow designed to help users move from raw data ingestion to identifying and resolving discrepancies across systems.

Why This Project Mattered
Public health teams are responsible for reconciling case data between state systems and the CDC. This process is critical for accurate reporting, but the underlying workflows are fragmented, inconsistent, and often manual.

Differences in how data is structured, grouped, and reported across systems create significant friction. Users must spend time identifying mismatches, validating records, and resolving discrepancies without a clear or unified workflow.

ACCURATE was created to bring structure and clarity to this process, enabling users to more efficiently understand and act on differences in their data.

The Challenge
This was not a traditional UI problem. The challenge was defining a product within a complex and ambiguous ecosystem.

Key complexities included:

• Multiple data sources with inconsistent structures
• Variation in how conditions are grouped and reported
• High reliance on manual reconciliation workflows
• The need to support both flexibility and repeatability
• Constraints around PHI and data visibility

The work required translating a fragmented system into a coherent experience while aligning stakeholders across product, engineering, and public health domains.

My Role
I led UX across the full lifecycle of the product, from early discovery through MVP definition and design.

My responsibilities included:

• Leading discovery and synthesizing findings across stakeholders
• Defining end-to-end workflows for reconciliation
• Structuring the product experience and interaction model
• Designing and prototyping core features and flows
• Collaborating closely with product, engineering, and CDC stakeholders
• Helping align the team around a clear, actionable product direction

Approach

Discovery

Conducted stakeholder interviews and analyzed existing workflows to understand how reconciliation was performed across systems.

Synthesized findings into key pain points, focusing on inconsistencies in data structure, workflow fragmentation, and lack of clarity in identifying discrepancies.

Framing

Defined the core problem space and established a structured approach to reconciliation.

Aligned stakeholders around key concepts such as:

• staged reconciliation workflow
• standardized discrepancy types
• clear separation of data views

Alpha

Close collaboration with CDC and STLT stakeholders was critical during Alpha to ensure the workflows reflected real-world reconciliation needs.

We worked iteratively with stakeholders to validate assumptions, review early outputs, and refine how data was ingested, compared, and interpreted. This continuous feedback loop helped mitigate risk, align on core concepts, and ensure the foundation we were building would scale effectively into MVP.

MVP

Evolved the Alpha into a scalable, production-ready solution.

Focused on:

• Direct database integration with NBS
• Structured, end-to-end reconciliation workflows
• Improved discrepancy identification and categorization
• Enhanced usability and clarity across complex data interactions

Refined the experience based on stakeholder feedback and technical constraints, simplifying interactions while preserving flexibility for real-world use.

Key Design Decisions
The complexity of reconciling public health data required more than interface design. It required structuring how the work itself happens.

These key design decisions were driven by real world workflows, data constraints, and stakeholder input. They translate a fragmented process into a clear, repeatable system that supports both analysis and operational use.

Structured reconciliation workflow

• Designed reconciliation as a clear, staged process to guide users from data ingestion through discrepancy resolution.
• This reduced ambiguity and made complex workflows more predictable and repeatable.

Compared vs. rendered data

• Separated “compared data” (core reconciliation) from “rendered data” (supporting context).
• This allowed users to focus on discrepancies while still having access to the detail needed for troubleshooting.

Flexible but repeatable reporting

• Balanced one-time report creation with scheduling capabilities.
• Users could define a report once and reuse it, supporting both ad hoc analysis and recurring workflows.

Condition management by year

• Introduced condition management tied to reporting year, allowing users to align with how data is actually grouped and reported in practice.

PHI-aware experience

• Designed for full data visibility within active sessions while supporting limited views outside of them.
• This ensured usability without compromising data handling constraints.

Product Highlights
Report dashboard

A centralized view of all reports, including scheduled and manual runs, allowing users to track status and quickly access results.

Reconciliation workflow

A structured workflow that guides users from data ingestion through comparison and resolution, reducing ambiguity and making complex reconciliation tasks more predictable.

Discrepancy review

Discrepancies are surfaced and categorized in context, allowing users to quickly understand issues and focus on resolution without leaving the workflow.

Scheduling and automation

Scheduling transforms reconciliation from a one-time task into a repeatable process, enabling teams to maintain data quality over time.

Data definition support

Integrated access to field definitions and mappings to help users understand discrepancies without leaving the workflow.

Outcome
ACCURATE established a clear and structured approach to data reconciliation within a complex public health ecosystem.

The work:

• Defined the product direction for MVP
• Translated fragmented workflows into a cohesive experience
• Improved clarity around identifying and resolving discrepancies
• Helped align stakeholders across product, engineering, and CDC teams

Reflection
This project was shaped not only by technical complexity, but also by real-world constraints around timing, access, and coordination.

During an administration transition, access to STLT stakeholders became more limited, requiring the team to rely on partial inputs while still moving forward. At the same time, several participating states expressed a need to begin reconciliation earlier, which accelerated the timeline and led to the introduction of an Alpha release to support immediate use.

To mitigate these gaps, we worked closely with subject matter experts on the team to validate assumptions and ensure the workflows reflected real-world practices. This helped maintain momentum while reducing the risk of misalignment.

These conditions required a more adaptive approach to design, balancing incomplete information with the need to deliver meaningful progress. Close collaboration across product, engineering, and stakeholders was critical to maintaining alignment throughout the process.

The experience also reinforced the importance of working effectively within a cross-functional team. Designing in this environment required understanding how different disciplines approach problems and adjusting communication and workflows to support shared outcomes.

Ultimately, these constraints helped shape a more grounded and practical solution, ensuring the product aligned with operational needs while remaining flexible enough to evolve.