For Retail Store Managers ·
What you'll accomplish
By the end of this guide, you'll use Claude to take your weekly inventory discrepancy list, categorize the likely causes (theft, receiving error, admin error), and produce an investigation priority list — so you spend your 2 hours on the highest-risk items instead of working through a random list.
What you'll need
From your POS or inventory management system, export your cycle count or physical count discrepancy report. You need: item name/SKU, location in store, quantity variance (system says X, you counted Y), and unit cost. Save as a spreadsheet or copy the key columns.
What you should see: A list with at least item name, variance quantity, and item value.
Troubleshooting: If your system doesn't have an export, manually type the top 20–30 discrepancies (most significant by value).
Go to claude.ai, start a new chat. Type:
I'm a retail store manager analyzing inventory discrepancies from our weekly cycle count. I need help categorizing them by likely cause and prioritizing my investigation. Here's the context about our store: [brief description — store type, size, typical high-theft items, any recent staffing or receiving issues]. I'll share the discrepancy list in the next message.
Copy your discrepancy data and paste it. Even a rough format works — Claude can handle messy tables. Include:
Type:
Please:
1. Calculate total inventory value at risk (sum of variance × unit cost for shortages)
2. Group these discrepancies by most likely cause: (a) potential theft — high-value, portable items; (b) receiving error — items that came in last Tuesday's delivery; (c) admin/scanning error — overages or small random variances; (d) unclear — needs investigation
3. Rank the top 10 items by investigation priority (highest value + most clear theft pattern first)
4. Flag any patterns that suggest a systematic problem (e.g., all from same location, all same brand, all from same receiving date)
Claude will produce a categorized list with investigation priorities. Use this to:
What you should see: A categorized list with a total dollar-at-risk figure, specific items grouped by cause, and a clear top-10 investigation priority list with rationale.
Troubleshooting: If Claude's categorization seems off, add more context about your store: "We have a known issue with the electronics wall" or "Our last receiving was a new vendor."
Quick pattern check: "Look at these discrepancies. Is there any pattern suggesting a systematic problem vs. random variance? [paste list]"
LP update for DM: "Based on this discrepancy analysis, write a 2-paragraph LP update for my district manager: total exposure, top issue areas, and my 3 investigation actions."
Receiving verification: "Here are items showing shortages that all arrived in Tuesday's delivery. Write a formal discrepancy notice to send to the vendor requesting credit for [X units of items listed]."
Year-over-year comparison: "Here's my shrink data for Q1 this year vs. last year. What's changed, and what should I focus on to reduce shrink this quarter? [paste data]"
Markdown vs. write-off decision: "These items have been on clearance for 90 days and still haven't sold. [list items + current price + qty on hand]. What markdown strategy would you recommend to clear this inventory by end of quarter?"