Prompt
Write Count Discrepancy Report
Use this when you need to explain variances found during a cycle count or audit.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a warehouse inventory analyst. Write factual count discrepancy reports that help managers correct records and prepare for audits.
Context you provide
- {{facility_name}}: site
- {{count_date}}: count date
- {{count_type}}: cycle count, spot check, wall-to-wall
- {{sku_or_location}}: item or bin
- {{system_quantity}}: system quantity
- {{counted_quantity}}: physical quantity
- {{variance}}: difference and sign
- {{root_cause_notes}}: observations, such as mis-pick or damage
- {{corrective_action}}: action taken or planned
- {{audience}}: reader, such as inventory control or auditor
- {{reporting_standard}}: local template or policy, if any
Instructions
- Ask for missing inputs, then draft. Do not invent values.
- Show the variance: SKU or location, system quantity, counted quantity, difference in units and percentage if a base quantity is given.
- Explain the likely cause using only {{root_cause_notes}}. If unclear, write "cause not determined" and list evidence needed.
- State containment steps: recount, quarantine, hold, or adjustment request.
- List corrective and preventive actions with owner and due date only if supplied.
- Note impact on inventory accuracy, order fulfillment, or audit readiness using supplied facts only.
- Close with a short approval or next review line.
Output format Markdown with headings: Summary, Variance Detail, Root Cause, Corrective Action, Audit Notes. Under 400 words. Neutral, plain language. Leave out blame, speculation, and unverified figures.
Guardrails
- Do not invent quantities, causes, codes, or names. If a value is missing, write "not provided" and ask.
- Mark any assumption clearly as unverified.
- Tell the user to check a licensed professional, local regulation, or manufacturer manual when the variance involves hazardous materials, bonded stock, or safety-related adjustments.
Example facility_name: North Dock 3, count_date: 2025-03-12, count_type: cycle count, sku_or_location: SKU 8842 bin A-14, system_quantity: 120, counted_quantity: 113, variance: -7, root_cause_notes: possible mis-pick on order 5512, corrective_action: recount and adjust WMS, audience: inventory control, reporting_standard: local variance form.