Enterprise KPI Governance & Multi-Department Forecasting
Rebuilding the operating cadence, forecasting precision, and performance governance across 8 departments, 12 divisions, and a $1B+ annual expenditure portfolio.
Eight departments, zero common metric baseline
The Allegheny County Department of Human Services manages an annual budget exceeding $1B across human services, housing, mental health, family preservation, and community safety. Despite immense operational responsibilities, performance tracking was deeply fragmented:
Each of the 8 constituent departments tracked operational progress using custom, self-defined metrics with no shared definitions. Every cross-functional leadership review turned into an exercise in data reconciliation rather than strategic decision-making. Analysts spent 12 business days per quarterly reporting cycle manually pulling disparate spreadsheets, verifying conflicting numbers, and answering 5 to 6 ad-hoc data questions per week.
Critically, forecasting inaccuracy was generating 50 to 60 mid-year fund reallocations annually, leaving approximately $100M in public funds misallocated at any given time due to delayed spend tracking.
Designing the 0→1 Governance Layer
The core issue was not analytical capacity; it was the lack of an institutional operating contract between data, finance, and department heads. I engineered a three-tier intervention:
1. Standardized KPI Taxonomy (8 Core Metrics): Rather than attempting to track hundreds of departmental signals, I collaborated with 8 department heads and 4 major contracted vendors to define 8 standardized, non-negotiable performance indicators spanning allocation velocity, contract burn rate, client service utilization, and compliance variance.
2. SQL-Driven Forecasting & Variance Engine: Queried 5 years of historical expenditure data across all departmental cost centers. Built multi-scenario forecasting models incorporating seasonality, legislative changes, and macro labor rates, creating an early-warning threshold triggered directly when departmental spend diverged by ±5% from planned curves.
3. Compressed Executive QBR Cadence: Replaced static 60-page PDF slide decks with interactive Power BI dashboards refreshed automatically. Executive reviews shifted from retrospectives on past numbers to forward-looking resource redeployment.
- 50–60 emergency budget reallocations per year
- ~$100M trapped in misallocated or stagnant accounts
- 12 days required to assemble executive quarterly review reports
- 5–6 ad-hoc data crisis inquiries received per week
- No automated integration with Workday Financials
- 10–15 planned, strategic reallocations per year (80% drop)
- Zero idle misallocation; capital actively deployed on schedule
- 3 days cycle time from month-end close to executive brief
- 55% reduction in ad-hoc requests via self-serve BI
- Proactive early-warning alerts directly tied to financial ERP
Driving adoption across 1,200 staff without direct authority
Public sector organizations present unique organizational dynamics: department heads protect budgetary autonomy, and changes to reporting structures are often met with skepticism. Success required active change management:
I conducted structured 1-on-1 alignment sessions with all 16 division leaders and finance partners. By demonstrating how the new forecasting model protected their funding from clawbacks and eliminated hours of manual audit preparation, we converted initial hesitation into collaborative ownership. The cross-department system rollout across 1,200 staff was delivered 3 weeks ahead of schedule with a lean 2-person analytical core.
"Governance is not about generating more reports. It is about creating the operating discipline where data is clean enough, fast enough, and trusted enough that leadership spends 100% of their review meeting deciding on actions rather than questioning the spreadsheet."