Skip to content
Fresh Margin Systems

How it works

A structured diagnostic for food purchasing leakage.

Fresh Margin Systems turns purchasing noise into a ranked review surface. Every step is designed to find what moved, where it moved, and what is worth checking next.

This public site explains the method and captures intake. The core diagnostic workflow runs through internal proprietary systems and operator-led review outside this repo.

Human reviewedFictional sample data in demosNo live customer data shown

What we need from you

  • Vendor list with primary contacts
  • Recent invoices (60-90 days preferred)
  • Current price sheets or contract rates
  • Purchase history by SKU and category
  • Contracts and amendments if available
  • Rebate agreements and tracking notes
  • Freight terms, delivery schedules, and surcharge details
  • SKU/category mapping file
  • Manual notes on known pain points or recent changes

What you receive

  • Margin Leak Brief
  • Vendor Drift Summary
  • Category Risk Console
  • Price Exception Queue
  • Rebate and Freight Review
  • Data Quality Notes
  • Pilot Decision Memo
  • Ranked Review Actions

Timeline

Four weeks from intake to decision memo. Faster when data is accessible and an executive sponsor joins Week 1.

01

Intake

Operator completes the pilot intake form. We review business type, purchasing volume, vendor complexity, and data availability to confirm fit before scoping.

Input

Intake form, brief discovery call

Output

Fit confirmation and scoped review plan

02

Data request

We request the exports and documents the operator already has: vendor lists, invoices, price sheets, purchase history, contracts, rebate notes, freight terms, and SKU/category files.

Input

Existing purchasing documents and exports

Output

Structured data inventory and quality snapshot

03

File review

We review the files for coverage, consistency, and gaps. Messy exports are expected. We note what is usable, what is missing, and what assumptions we need to flag.

Input

Raw exports and documents

Output

Data quality notes and coverage assessment

04

Normalization

We normalize pack sizes, unit costs, vendor names, and category mappings so comparisons are apples-to-apples. This is where hidden pack-size changes surface.

Input

Raw SKU, invoice, and price-sheet data

Output

Normalized purchasing data set

05

Vendor drift detection

We compare current invoices against price sheets and contracts to find unit-cost movement, substitution patterns, and terms changes that may not have been flagged internally.

Input

Normalized data + contract references

Output

Vendor drift signals and substitution list

06

Category risk mapping

We map margin pressure across protein, produce, dairy, dry goods, frozen, disposables, beverage, and cleaning supplies to show where concentration and movement overlap.

Input

Normalized data by category

Output

Category Risk Console with severity labels

07

Exception queue creation

We build a ranked queue of price exceptions, invoice variances, freight anomalies, and rebate gaps. Each item gets a severity label and a recommended review action.

Input

Drift signals and category risks

Output

Ranked Price Exception Queue

08

Human operator review

A human reviewer pressure-tests the findings, validates assumptions, and confirms context. No output is published without this step.

Input

Preliminary findings and operator context

Output

Reviewed and validated leak brief

09

Decision memo

We deliver the Margin Leak Brief, Vendor Drift Summary, Category Risk Console, Price Exception Queue, Rebate/Freight Review, Data Quality Notes, and Pilot Decision Memo.

Input

Reviewed findings

Output

Written diagnostic deliverables

10

Follow-up action plan

We provide a ranked action plan with effort estimates, business relevance, and next steps. For continuing pilots, we schedule recurring review cadence and track actions over time.

Input

Operator capacity and priorities

Output

30 / 60 / 90-day action plan

Data quality caveat: A diagnostic is only as good as the data provided. Findings require human review and are not guaranteed savings. Sample reports and demos use fictional data. No live customer data is shown on this site.

Start

Begin with a pilot intake.

A 30-minute fit review confirms whether the diagnostic makes sense for your operation. If yes, the next step is a scoped 30-day review.