Project Highlights
Data Science
Workers’ Compensation Underwriting ML Quote App
PROBLEM / PURPOSE
The existing quote model used for workers’ compensation underwriting was ineffective at predicting risk—especially for new clients or clients with volatile claim histories—resulting in limited utility for pricing strategy.
KEY ACHIEVEMENTS / IMPACT
Delivered a highly improved model that accurately predicts claims for both established and new businesses. Enabled the Pricing Team to stratify risk and assign more accurate underwriting rates. Model performs well for expected claims under $100K and clearly flags high-risk clients over that threshold. App still in active use.
Sales Opportunity Predictive Forecasting
PROBLEM / PURPOSE
The Sales Team relied on qualitative labeling and Salesforce Einstein scores to predict opportunity closures, both of which were not sufficiently accurate.
KEY ACHIEVEMENTS / IMPACT
Reduced reliance on inaccurate manual or Einstein-based forecasts. Achieved 90–95% accuracy in predicting whether an opportunity would close in the current or following month, significantly improving forecasting reliability for both Sales and executive leadership.
Client Contact Clustering & Segmentation
PROBLEM / PURPOSE
Client-facing contacts varied widely in behavior, influence, and satisfaction, but there was no structured way to segment or understand them. This limited the company’s ability to target referrals, gather testimonials, or identify service issues.
KEY ACHIEVEMENTS / IMPACT
Enabled Account Managers to prioritize promoter-type contacts for referrals/testimonials and recognize patterns in dissatisfied contact types. Provided foundational logic for future segmentation and service strategy improvements.
Client Satisfaction Prediction Model
PROBLEM / PURPOSE
The company needed a way to proactively identify clients at risk of dissatisfaction or termination, and to highlight those likely to provide referrals. Existing feedback (e.g., NPS) came too late for proactive action.
KEY ACHIEVEMENTS / IMPACT
Enabled weekly satisfaction prediction to proactively engage at-risk clients and strengthen referral outreach. Improved visibility into client health, supported client success workflows, and provided a foundation for ongoing model refinement.
Client Survival Analysis & Break-Even Modeling
PROBLEM / PURPOSE
The company lacked visibility into when and why clients terminate, how long they remain profitable, and how break-even timing differs across client types. This limited the ability to strategically allocate resources and optimize service models.
KEY ACHIEVEMENTS / IMPACT
Revealed client lifecycle patterns and profitability dynamics across key segments. Enabled leadership to better prioritize client types for retention, product tailoring, and resource allocation.
COVID-Era Monthly Profitability Forecasting
PROBLEM / PURPOSE
During the onset of COVID-19, leadership needed to understand how client shrinkage and rate relief would impact month-by-month profitability. Uncertainty in economic conditions made traditional forecasting unreliable.
KEY ACHIEVEMENTS / IMPACT
Enabled leadership to plan for pandemic-related financial disruption with confidence. Provided flexible modeling that allowed for scenario adjustments. Forecast remained highly accurate despite market volatility.
Data Analysis
Data Job Trends Analysis from LinkedIn Postings
PROBLEM / PURPOSE
Job seekers and hiring managers often lack clear, up-to-date insight into demand trends and skill requirements across data roles. Publicly available reports are often too high-level or outdated.
KEY ACHIEVEMENTS / IMPACT
Delivered two-part public analysis with data-backed insights into the data job market. Demonstrated the ability to independently gather, clean, analyze, and communicate complex information clearly. Article supports career planning and job market navigation.
Client Profitability Report
PROBLEM / PURPOSE
The Pricing Team had no way to holistically assess a client’s profitability when Account Managers escalated repricing requests. Without visibility into billing, claims, or insurance-related costs, repricing decisions lacked data-driven support.
KEY ACHIEVEMENTS / IMPACT
Data-enabled the Pricing Team to assess clients holistically and make informed, targeted repricing decisions. Introduced benchmarking and per-stream profitability analysis, improving precision and defensibility of pricing strategies.
Responsiveness KPI Dashboard (Calls & Cases)
PROBLEM / PURPOSE
Leadership needed centralized visibility into responsiveness across client-facing channels, including phone and case activity. Metrics existed in separate systems and lacked consistency or trend reporting.
KEY ACHIEVEMENTS / IMPACT
Delivered a unified dashboard used across the organization for tracking responsiveness KPIs. Enabled visibility into both near-real-time metrics and historical trends. Supported operational monitoring and performance review across departments.
Revenue Forecast Variance Root Cause Analysis
PROBLEM / PURPOSE
The company experienced a large Q1 revenue shortfall despite having a record sales year the prior year. Leadership needed to understand the disconnect between forecasted and actual revenue.
KEY ACHIEVEMENTS / IMPACT
Revealed the underlying cause of a major financial variance. Drove awareness that Pricing strategy shifts must be accompanied by forecasting models. Informed executive-level discussions and influenced cross-team accountability in future pricing changes.
Client Health Claims Benchmarking Report
PROBLEM / PURPOSE
Larger clients wanted insights on employee wellness trends and healthcare usage to support initiatives like flu shots and chronic condition management. The company lacked a scalable, data-driven product to provide value in this space.
KEY ACHIEVEMENTS / IMPACT
Produced an innovative, HIPAA-compliant client reporting concept aimed at increasing client satisfaction and creating a unique product offering. Although the product didn’t launch due to a role transition, the framework remains a potential foundation for future white-glove client programs.
Sales Funnel & Performance Reporting with Enhanced Metrics
PROBLEM / PURPOSE
The Sales Team lacked clear visibility into pipeline health and conversion performance. Salesforce reporting limitations restricted analytical depth and metric innovation.
KEY ACHIEVEMENTS / IMPACT
Enabled the Sales Team to make data-driven decisions based on newly introduced and more actionable metrics. Improved transparency across the sales funnel and supported better targeting, coaching, and strategic adjustments.
Analytics Engineering / Data Architecture
Voice Call Self-Service Reporting via Google Sheets
PROBLEM / PURPOSE
SSRS and Tableau reporting created a bottleneck, as most employees lacked the access or skills to work directly with Snowflake data. One manager responsible for phone queue oversight had to manually download call data from the phone system and analyze it in Excel—a process that took several hours each time and delayed timely insights.
KEY ACHIEVEMENTS / IMPACT
Replaced multi-hour manual Excel-based call reporting with a scheduled, self-updating dashboard accessible to business users. Proved the viability of Coefficient-based reporting, paving the way for wider adoption of self-service data tools.
Email Text Cleaner for Communication Mining
PROBLEM / PURPOSE
Emails stored in Salesforce were difficult to analyze due to reply chains, signatures, legal disclaimers, and other noise that obscured meaningful communication patterns.
KEY ACHIEVEMENTS / IMPACT
Created a reusable cleaned email dataset that became the foundation for multiple downstream analytics and ML projects, significantly improving signal clarity for communication-based insights.
Self-Service Data Object Suite
PROBLEM / PURPOSE
Reliance on SSRS and Tableau created a bottleneck for report creation, since only the Analytics Team had Snowflake access or technical skills. As a result, many valid data needs from other teams went unmet or were manually handled in Excel/Google Sheets.
KEY ACHIEVEMENTS / IMPACT
Delivered robust, scalable data architecture and self-service products that preemptively addressed cross-functional reporting needs. Greatly reduced reliance on Analytics Team and improved access to trusted data sources. Strong user adoption and positive feedback across departments.
Business Minutes Between Timestamps Calculation
PROBLEM / PURPOSE
Multiple reports required accurate calculations of business hours between two timestamps, but no consistent or scalable method existed across teams.
KEY ACHIEVEMENTS / IMPACT
Created a durable and widely adopted solution for calculating business time across reports. Eliminated inconsistency and manual workarounds, becoming the go-to method for timestamp-based metrics across teams.
Automated HRIS Document Extraction & Decoding
PROBLEM / PURPOSE
During HRIS migration, there was no method to bulk download stored documents (PDFs and images) from the legacy system. Manual extraction was unscalable, and encoded data posed challenges for decoding and storage.
KEY ACHIEVEMENTS / IMPACT
Eliminated the need for manual document extraction across thousands of client records. Delivered a scalable, organized archive of HRIS documents that could be imported into the new system and referenced reliably. Improved traceability and operational efficiency during the HRIS migration.
Data Product Management / Strategy
Payroll Case Relevance Prioritization Prototype
PROBLEM / PURPOSE
The Payroll Operations Team was missing critical information buried in Salesforce cases due to lack of labeling, causing a high error rate in payroll processing. Teams were overwhelmed and unable to manually flag relevant cases.
KEY ACHIEVEMENTS / IMPACT
Delivered a working prototype to prioritize high-risk payroll cases with automated summaries and scoring logic. Though full implementation was paused due to shifting company priorities, the solution was positively received and seen as highly impactful by the Operations Team.
Sales Process Migration from ClientSpace to Salesforce
PROBLEM / PURPOSE
The company needed to transition its Sales process and Pricing workflows from ClientSpace to Salesforce to improve maintainability, reporting capabilities, and integration with broader systems.
KEY ACHIEVEMENTS / IMPACT
Enabled a seamless transition of the Sales process into Salesforce with zero errors at launch. Built a robust setup still used by the Pricing Team today, supporting both operational use and long-term analytics.
Pricing & Sales Proposal Export Automation
PROBLEM / PURPOSE
The existing pricing and sales proposal exports were outdated, complex, and required extensive manual edits. Sales Operations had to manually replicate Pricing's changes, leading to errors and inefficiencies. Switching between sales models required redoing work, and glitches and formula nuances caused frequent mistakes.
KEY ACHIEVEMENTS / IMPACT
Eliminated manual edits across departments, reducing errors and saving Sales Operations hours per deal. Enabled seamless switching between sales models without rework. Improved usability for non-technical users.
Sales Operations & Pricing Resource Hub
PROBLEM / PURPOSE
The Sales Operations Team frequently submitted incomplete or inaccurate health insurance quote requests, causing delays, compliance risks, and sales friction. Meanwhile, the Pricing Team lacked a centralized system for storing SOPs, reference files, and institutional knowledge, leading to inconsistent practices and onboarding challenges.
KEY ACHIEVEMENTS / IMPACT
Submission accuracy improved from 50–60% to 83% in the first month, and stabilized at 95–99%. Enabled faster, more compliant sales cycles. Greatly reduced Pricing Team’s onboarding time and improved consistency by consolidating scattered procedures and institutional knowledge into a single source of truth.
Insurance & Risk Training and Reference Hub
PROBLEM / PURPOSE
New employees had no centralized resource for understanding the complex insurance landscape, risk management practices, and unique product structures at the company. This created challenges in training, knowledge transfer, and consistency across Pricing and Risk teams.
KEY ACHIEVEMENTS / IMPACT
Accelerated onboarding and training for new team members and leadership. Improved consistency and understanding of complex insurance operations. Resource is still in active use across multiple departments.
Sales / Revenue Operations (BizOps)
Automated Sales Performance & True-Up Reporting
PROBLEM / PURPOSE
Sales Team manually created a complex report to track quotas, closed-won deals, true-ups, and non-materialized sales. The process was very labor-intensive.
KEY ACHIEVEMENTS / IMPACT
Reduced manual reporting workload by automating data retrieval and calculations. Provided a scalable, multi-year reporting structure with detailed true-ups and flexible first-invoice annualization.
Company-Wide Mass Repricing Implementation
PROBLEM / PURPOSE
A company-wide pricing change required repricing all clients—typically a high-risk, manual process with potential for client confusion and internal errors.
KEY ACHIEVEMENTS / IMPACT
Successfully repriced all clients in a single coordinated process. Enabled clean client communication and streamlined implementation across teams. Execution was error-free.
Client Profitability Report
PROBLEM / PURPOSE
The Pricing Team had no way to holistically assess a client’s profitability when Account Managers escalated repricing requests. Without visibility into billing, claims, or insurance-related costs, repricing decisions lacked data-driven support.
KEY ACHIEVEMENTS / IMPACT
Data-enabled the Pricing Team to assess clients holistically and make informed, targeted repricing decisions. Introduced benchmarking and per-stream profitability analysis, improving precision and defensibility of pricing strategies.
Unemployment Rate Forecasting & Risk Stratification
PROBLEM / PURPOSE
Accurate forecasting of state unemployment insurance (SUI) rate schedule shifts and client-level renewals was critical for financial planning, risk management, and client pricing—but was historically reactive and imprecise.
KEY ACHIEVEMENTS / IMPACT
Accurately predicted state SUI schedule changes, enabling proactive strategy planning. Improved risk segmentation and client renewal forecasting accuracy. Strengthened credibility of Pricing and Risk teams.
Mergers & Acquisitions Due Diligence – Competitor Acquisition
PROBLEM / PURPOSE
ProService Hawaii was acquiring a major competitor and needed rigorous due diligence to evaluate profitability, validate financial data, and assess risk exposure across billing, insurance, and tax.
KEY ACHIEVEMENTS / IMPACT
Played a key role in a successful acquisition by providing trusted financial projections and surfacing potential risk areas. Helped leadership make informed decisions during a high-impact transaction.
Metadata / Data Governance
Self-Service Data Object Suite
PROBLEM / PURPOSE
Reliance on SSRS and Tableau created a bottleneck for report creation, since only the Analytics Team had Snowflake access or technical skills. As a result, many valid data needs from other teams went unmet or were manually handled in Excel/Google Sheets.
KEY ACHIEVEMENTS / IMPACT
Delivered robust, scalable data architecture and self-service products that preemptively addressed cross-functional reporting needs. Greatly reduced reliance on Analytics Team and improved access to trusted data sources. Strong user adoption and positive feedback across departments.
Sales Operations & Pricing Resource Hub
PROBLEM / PURPOSE
The Sales Operations Team frequently submitted incomplete or inaccurate health insurance quote requests, causing delays, compliance risks, and sales friction. Meanwhile, the Pricing Team lacked a centralized system for storing SOPs, reference files, and institutional knowledge, leading to inconsistent practices and onboarding challenges.
KEY ACHIEVEMENTS / IMPACT
Submission accuracy improved from 50–60% to 83% in the first month, and stabilized at 95–99%. Enabled faster, more compliant sales cycles. Greatly reduced Pricing Team’s onboarding time and improved consistency by consolidating scattered procedures and institutional knowledge into a single source of truth.
Error Communication Flagging & Reporting
PROBLEM / PURPOSE
Leadership needed a way to identify client or internal emails referencing possible service errors (e.g., payroll issues), but scanning raw email threads was inefficient and unreliable.
KEY ACHIEVEMENTS / IMPACT
Delivered a dynamic reporting solution that enabled stakeholders to efficiently monitor communications for possible service breakdowns. Provided a flexible interface to refine keyword sets and reduce manual email review time.
Insurance & Risk Training and Reference Hub
PROBLEM / PURPOSE
New employees had no centralized resource for understanding the complex insurance landscape, risk management practices, and unique product structures at the company. This created challenges in training, knowledge transfer, and consistency across Pricing and Risk teams.
KEY ACHIEVEMENTS / IMPACT
Accelerated onboarding and training for new team members and leadership. Improved consistency and understanding of complex insurance operations. Resource is still in active use across multiple departments.