Two S-Curves, Not One

Practices sit on two lifecycles simultaneously. The Revenue curve tracks top-line growth health. The EBITDA curve tracks margin / profitability health. A practice can be strong on one and weak on the other — these are different levers.

Stages (both curves)

Stage assignment: score + gates

A practice is placed on score band (S1:0–20, S2:20–40, S3:40–60, S4:60–80, S5:80+). Gate failures can demote a stage when discipline is weak despite a high score — this is how we resolve "why S3 not S2?" explicitly.

Metric weights (v1)

EBITDA v1: margin proxy

Without direct cost data (labor %, supply %, lab %, overhead %) in the current source workbooks, EBITDA is scored as a margin-driver proxy — weighted toward the metrics we can observe that correlate with profitability: collection capture, PPV quality, scale absorption.

v2 upgrade path: swap in labor ratio, supply cost %, lab cost %, overhead %, provider comp %, AR >90d — sourced from Power BI. Schema and weights table already support this swap.

Lever engine — how $ EBITDA is estimated

Each practice is evaluated against 5 lever types. For each applicable lever, theoretical gap × flow-through × realization factor produces a 12-month EBITDA estimate, capped at 8% of annual gross.

• Close Collection Gap — lift GC rate to 62% target
• Lift Production-per-Visit — throughput quality to 60th pctile
• Drive New Patient Volume — demand engine to 50th pctile
• Restore Growth Momentum — YoY back to +10%
• Add Capacity — S5 scaling when chair-time is the ceiling

Assumptions: 40% marginal EBITDA flow-through, 30% realization within 12 months, 8% of annual gross cap per lever. Conservative by design — meant to survive board scrutiny.

Right-metrics hypothesis (open)

This v1 uses percentile-based scoring on 5 dimensions. The richer hypothesis — not yet wired in because the data feeds aren't available in the source workbooks — expands to:

These require Power BI labor / margin feeds and Dental Intel case acceptance + reappointment pulls.