AnalysisHigh riskComparison recommended

AI Inventory Overstatement Detection Playbook

A $180M wholesale distributor is being acquired. Due diligence reveals that gross margin has been 3.2 points above industry average for 18 consecutive months. You suspect inventory valuation manipulation. The company uses FIFO and has 4 warehouse locations.

When to use this playbook

  • Use this playbook when the decision looks like the situation above: A $180M wholesale distributor is being acquired.
  • It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "Inventory Overstatement Detection".
  • Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

  • Inventory subledger for 18 months (all SKUs, all locations)
  • Purchase order and receiving log
  • Gross margin by product category
  • Industry margin benchmarks (attached)
  • Prior-year physical count results

Attachments: Spreadsheets (Spreadsheets)

The Prompt

You are a forensic accountant investigating potential inventory overstatement at a $180M wholesale distributor. I am attaching:

Work only from the attached source files. If a conclusion is not supported, say so.

Produce:
1. Identify SKUs where the recorded cost basis is inconsistent with purchase order pricing over the same period—a sign of cost capitalization manipulation.
2. Flag product categories where the distributor's gross margin exceeds the industry benchmark by more than 2 standard deviations.
3. Assess whether the variance is explainable by volume purchasing, private label mix, or geographic pricing—or whether it requires an inventory write-down.
4. Estimate the inventory overstatement range and its impact on the acquisition EBITDA multiple.
5. Tell me what questions to put to the CFO and which warehouse location to physically count first.

Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.

What to expect

  • SKU-level cost variance report
  • Margin anomaly by category with statistical thresholds
  • Estimated overstatement range
  • Physical count priority list with rationale

Review before you act

  • Validate this output against source files before relying on it: Identify SKUs where the recorded cost basis is inconsistent with purchase order pricing over the same period—a sign of cost capitalization manipulation.
  • Validate this output against source files before relying on it: Flag product categories where the distributor's gross margin exceeds the industry benchmark by more than 2 standard deviations.
  • Validate this output against source files before relying on it: Assess whether the variance is explainable by volume purchasing, private label mix, or geographic pricing—or whether it requires an inventory write-down.
  • Validate this output against source files before relying on it: Estimate the inventory overstatement range and its impact on the acquisition EBITDA multiple.
  • Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
  • Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
  • Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.

Why compare models on this

For Inventory Overstatement Detection, running the same attachments across independent models is useful because the hard part is classification and completeness, not fluency. The workflow is already designed to surface sku-level cost variance report; margin anomaly by category with statistical thresholds; estimated overstatement range; physical count priority list with rationale. Those are comparison artifacts — they only exist if more than one model runs. Models often split on qualitative materiality, intent versus error, and whether a newly formed counterparty is a red flag or a legitimate intermediary. Those splits are the review queue — not noise.

Forensic AccountingInventory and Cash SchemesAnalysisHighSpreadsheets

See governed multi-model AI on your own prompt

Compare GPT-5, Claude, and Gemini side by side, with human review and a decision record built in.