You’re the analytics lead at a multi-specialty outpatient healthcare provider operating 35 clinics across 6 states, with ~2.5M patient visits/year. The business has grown quickly through acquisitions and now offers multiple service lines (Primary Care, Orthopedics, Imaging, Physical Therapy, Behavioral Health, and Telehealth). Revenue is a mix of fee-for-service and value-based contracts, and margins are under pressure due to rising labor costs and payer mix shifts.
At the last board meeting, the CFO reported that company-wide EBITDA margin fell from 14% to 9% YoY, despite total revenue growing +18%. The CEO believes “some departments are subsidizing others,” but the current reporting is inconsistent: some costs are booked centrally, some at clinics, and some at service lines. Department heads are disputing the numbers, claiming shared costs (call center, scheduling, IT, rent, clinical supervisors) are being allocated unfairly.
You are asked to design a repeatable profitability analysis framework by department (e.g., clinic operations, call center, billing, corporate) and by service line (e.g., Imaging, PT, Ortho) that can be used for monthly business reviews and pricing decisions.
Stakeholders want answers to:
You have 10 business days to deliver a recommendation that will influence next quarter’s operating plan and potentially a $20M cost reduction program.
| Source | Description | Granularity |
|---|---|---|
| encounters | Patient visits with service line, CPT codes, rendering provider, clinic, timestamps | Encounter / line item |
| claims | Submitted claims, allowed amount, paid amount, adjustments, denial codes, payer | Claim / line item |
| general_ledger | Expenses and revenue postings with cost center, department, vendor, account | Journal entry |
| payroll | Hours, wages, role, department, clinic, provider productivity measures | Employee-day / pay period |
| scheduling | Appointment slots, utilization, no-shows, cancellations, lead time | Appointment |
| patient_acquisition | Marketing spend, channel, attributed new patients, campaign | Campaign / week |
Constraints: