SGA Production — Data Trust Check

SGA Bridge vs Gen4 Pacing File · how well does our internal data agree with the source of truth?
… April + May 2026
April Data Trust
May Live Production
Doctor Days
Scheduled Production

Off-pace cohorts — where the problem is

Click any cohort to filter the investigation list below

Investigation queue

Clean (≤1%) Minor (1–2%) Watch (2–5%) Action (>5%) No data — closed/divested Bands chosen: ±1% = within rounding · >5% = likely pipeline defect
Practice · PMS · ROD SGA Pacing Drift Gap $ Likely cause Status Owner
▸ See all 93 practices — full reconciliation grid All metrics inline · SGA / Pacing / Drift per cell
Practice SGA Prod Pacing Δ % Context drift Budget %

Crosswalk gaps Pacing rows that don't map to a SGA practice

Per-location production · May 2026

Practice Legacy Completed (May 1 – today) Scheduled (after today) Booked Scheduled %

What this tab shows

  • Truth: data-inbox/Copy Doctor Days 2026.xlsx → Provider Report. April Budget = col V, Worked = col Y. Dentists only.
  • SGA Bridge: [Days Worked] per Provider · filtered to Dr or Hyg = "Dentist" · rolled up by Practice Code.
  • Gen4 PBI: DISTINCTCOUNT(Production[on_date]) where completed_production > 0 · filtered via Providers[User Type] = "DEN" · rolled up by location_id.
  • Grain: location-level — EEID in the truth file is a Paylocity-style ID that doesn't match either PBI's PMS-specific provider_id, so per-provider keys don't align. Per-provider truth is preserved in the row drill-down.

Per-location reconciliation · dentist days

Clean (≤5%) Minor (5–10%) Watch (10–25%) Action (>25%) Days-worked bands wider than $ bands — partial-day counting + threshold definitions vary by source
Location · Ops Truth
Budget		 Truth
Worked		 SGA
Budget		 SGA
Worked		 Gen4
Worked		 SGA Worked
Δ vs Truth		 Gen4 Worked
Δ vs Truth

What this tab shows

  • SGA Bridge: SUM(SAP_Cleaned_FullLoad_Output[estimated_amount]) where Deduced_Status = "Scheduled", date in target month.
  • Gen4 PBI: SUM(Production[scheduled_production]) for the target month from the Production fact.
  • April is fully in the past, so scheduled production should normally be near zero (any non-zero values represent appointments still flagged Scheduled that never got reconciled). Use this as a data-hygiene signal.
  • Once May concludes the same audit will rerun here — pacing-truth source becomes the Gen4 pacing file for May.

Per-location scheduled production · April 2026

Location SGA Scheduled Gen4 Scheduled Gap $ Notes