Margin Review System
Diagnostic architecture that turns purchasing data into operator decisions.
The Margin Review System is the operating framework behind every Fresh Margin Systems diagnostic. Data ingestion, normalization, signal detection, human review, deliverables, and operating cadence. Some layers are delivered today by the founder. Others are in internal prototype. Software surfaces are planned and shaped by early operator pilots.
Six layers. From raw data to operator decision.
Every diagnostic follows the same pipeline. The difference between engagements is the data, not the process.
Available now
Data ingestion
- Invoices and purchase orders
- Vendor lists and contacts
- Price sheets and amendments
- SKU/category mapping files
- Freight records and surcharge schedules
- Rebate notes and tracking spreadsheets
- Contracts and amendments
- Substitution logs and approval records
Available now
Normalization
- Vendor name normalization
- SKU mapping and deduplication
- Unit-of-measure normalization
- Pack-size normalization
- Category mapping and consolidation
- Contract-to-invoice matching
- Date and period alignment
Available now
Signal detection
- Invoice variance detection
- Vendor price drift scanning
- Freight creep identification
- Rebate slippage flagging
- Pack-size movement tracking
- Substitution pattern detection
- Contract staleness alerts
- Category pressure mapping
Available now
Human review
- Human validation of every signal
- Confidence scoring and labeling
- Assumption logging
- Data quality flagging
- Operator context integration
- Decision boundary documentation
- False positive filtering
Available now
Deliverables
- Margin Leak Brief
- Vendor Drift Summary
- Category Risk Console
- Price Exception Queue
- Freight/Rebate Review
- Data Quality Snapshot
- Pilot Decision Memo
Internal prototype
Operating cadence
- 30-day diagnostic kickoff
- 90-day pilot review
- Monthly exception review
- Quarterly vendor review
- Contract renewal prep
- Annual scope reassessment
How data moves through the system.
1. Ingest
Collect vendor lists, invoices, price sheets, purchase history, contracts, rebate notes, freight terms, and SKU/category files. Accept messy exports. Flag missing data.
2. Normalize
Standardize vendor names, SKUs, units of measure, pack sizes, and category mappings. Match contracts to invoices. Align dates and periods.
3. Detect
Scan for invoice variance, vendor price drift, freight creep, rebate slippage, pack-size movement, substitution patterns, contract staleness, and category pressure.
4. Review
Every signal is validated by a human before delivery. Confidence is scored. Assumptions are logged. False positives are filtered. Operator context is integrated.
5. Deliver
Produce the Margin Leak Brief, Vendor Drift Summary, Category Risk Console, Price Exception Queue, Freight/Rebate Review, Data Quality Snapshot, and Pilot Decision Memo.
6. Decide
The operator reviews the findings, prioritizes actions, and decides what to do. The diagnostic provides visibility and recommendations, not mandates.
Truth posture
We do not claim finished software that does not exist.
- The first commercial product is a service-first diagnostic, not a SaaS platform.
- Software modules are in development and shaped by early pilots. No fixed launch date.
- We will not announce a feature before it is tested with real operator data.
- No autonomous procurement claims. No AI-replaces-buyers claims. No guaranteed savings.
See it in action
The sample audit shows the current deliverable format.
Every layer described above appears in the sample audit. Review it to understand what a real diagnostic looks like for your operation.