Par/Ex Analytics • Solutions
OPERATIONAL OPTIMIZATION + SOFTWARE & DATA ENGINEERING

Precision Par Benchmarks & Margin Engineering.

Most data engineers don’t understand kitchen par levels, prime cost, or RevPAR. Most hospitality consultants live inside broken spreadsheets. Par/Ex Analytics bridges both worlds—deploying production SQL, Python, and machine learning pipelines engineered specifically for F&B groups, boutique hotels, and hospitality tech platforms.

Native Data Ingestion & Modeling Across:

Toast & Square POS Oracle Micros / Opera Resy & OpenTable 7shifts & HotSchedules Python / SQL / dbt
Live Operational Signal

Prime Cost & Variance Telemetry

ILLUSTRATIVE MODEL
Predictive Prep Variance Reduction -3.4% COGS

Inputs: 18mo Toast POS + NOAA Weather + Resy Covers

K-Means Guest Cohort Retention Lift +14.2% Mid-Week

Cluster: "High-Margin Local Regulars" (n = 142,800 checks)

Engagement Model 2–4 Week Sprints
Deliverable Production Code + ROI
Fixed-Scope Engagements

Productized Analytical Sprints

Operators don’t buy open-ended data science retainers—they buy measurable fixes to margin compression. Each sprint runs 2 to 4 weeks with clear raw data inputs, production code handoff, and direct P&L impact.

F&B OPERATIONS 3 WEEKS

Predictive Prep & Waste Forecasting

Replaces gut-feel par sheets with a lightweight Python time-series forecasting pipeline trained on your historical item-level sales, local weather forecasts, day-of-week seasonality, and reservation pacing.

Required Data Inputs

12+ mos POS item-level exports (Toast/Square), prep/waste logs, reservation covers.

Production Output

Automated daily prep sheet generator + kitchen variance dashboard.

Target Impact 2.0% – 4.0% COGS Cut
Scope Sprint →
F&B & BOUTIQUE HOTELS 3 WEEKS

Guest Behavioral Clustering & Retention Engine

Applies unsupervised ML clustering (K-Means / RFM segmentation) across merged POS and reservation logs to separate high-LTV regulars from one-and-done visitors—revealing exact menu affinities and churn triggers.

Required Data Inputs

Anonymized POS check logs, Resy/OpenTable guest history, PMS folio summaries.

Production Output

4–6 actionable guest archetypes, menu pairing matrix, & CRM segment exports.

Target Impact +10% – 15% Repeat Visits
Scope Sprint →
TECH & MULTI-UNIT OPS 2–4 WEEKS

Automated Multi-Unit Data Warehouse & ETL

Eliminates 10+ hours of weekly GM spreadsheet copy-pasting. Builds automated Python/SQL ingestion pipelines uniting POS, labor scheduling (7shifts), and accounting into one clean cloud source of truth.

Required Data Inputs

API keys or scheduled exports from POS, labor platform, and GL/accounting.

Production Output

Automated SQL warehouse + live executive prime-cost & store-comp dashboard.

Target Impact 100% Automated Reporting
Scope Sprint →
Applied Research & Architectures

Case Studies & Technical White Papers

Representative architectures and simulated operational benchmarks demonstrating how SQL, Python, and unsupervised ML solve real hospitality unit-economics problems. Click any brief to inspect the methodology and data model.

Interactive P&L Sensitivity Tool

Estimate Your Prime Cost Recovery

In multi-unit F&B and hotel dining operations, small unmeasured variances in prep waste and dead-hour labor quietly erode net operating income. Model the annualized bottom-line impact of a 2.5% prime cost compression.

6 Units
1 Unit 12 Units 25 Units
$2,400,000
$800K $4.0M $8.0M
64.0%
52% (Lean) 64% (Industry Avg) 75% (Compressed)
Projected Annualized EBITDA Lift
$360,000

Based on a conservative 2.5% reduction in food prep waste & scheduling variance across $14.4M portfolio revenue.

New Prime Cost 61.5%
Monthly Recovery $30,000/mo
Why Work With Me

The Rare Intersection of Production Data Engineering & Hospitality P&L Mastery

</>

Deep Technical Execution

Grounded in a background building production software and data architecture inside the hospitality technology ecosystem. I write clean, documented, maintainable code that your team actually owns after the sprint ends.

  • ✓ Data Engineering & Warehousing: Automated Python/SQL ETL pipelines, schema normalization across fragmented POS/PMS APIs, and cloud database modeling.
  • ✓ Machine Learning & Forecasting: Unsupervised behavioral clustering (K-Means, hierarchical), RFM customer scoring, and time-series demand forecasting.
  • ✓ Zero Black-Box Lock-In: Full Git repository handoff, clear documentation, and seamless handoff to internal ops or engineering teams.
🖧

Hospitality Operations Fluency

Trained at the Cornell Nolan School of Hotel Administration (Master of Management in Hospitality). I translate raw transaction logs directly into the metrics owners, GMs, and asset managers care about.

  • ✓ F&B Unit Economics: Theoretical vs. actual food cost variance, menu contribution margin engineering, and prime cost optimization.
  • ✓ Hotel & Revenue Dynamics: Net RevPAR after OTA commission leakage, F&B outlet capture rates, and guest lifetime value modeling.
  • ✓ Operator-Ready Communication: Executive briefings designed for culinary directors, VPs of Ops, and private equity operating partners—not academic data dumps.
Start With a Low-Friction Diagnostic

Book a 20-Minute Data & Margin Diagnostic

Tell me a bit about your current tech stack (POS, PMS, labor tools) and where margin or reporting friction is hurting most. We’ll identify whether a 2-to-4-week sprint makes sense.