As of 13 August 2026, AI can only partly manage stock reorders for your business.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, power-user skill, and roughly 1 hour until you can act on the result.

   How to actually do it

   1. Open your inventory or stock spreadsheet and export the current SKU, available stock, reserved stock, stock on order, unit and location data.
   2. Gather the relevant sales or usage export and record the period covered, removing cancelled transactions and identifying any unusual one-off demand.
   3. Open your supplier records and add the current supplier, pack size, unit cost, minimum order quantity, lead time, order cut-off and delivery terms for each SKU.
   4. Paste the business context, stock data, sales data and supplier data into the prompt, marking every unavailable field as unknown rather than estimating it.
   5. Ask the chatbot to produce the reorder table, missing-data list, assumptions, anomalies and draft purchase orders exactly in the requested format.
   6. Compare every recommended quantity and supplier term against the live inventory system, current price list, open purchase orders and supplier account before approving anything.
   7. Ask a colleague with responsibility for stock or purchasing to approve the final quantities, then enter or send the approved orders through your normal purchasing system.

   Prompt

   You are helping me prepare a stock reorder review for a UK business. Do not place orders, contact suppliers or invent missing figures. Use only the data I provide.
   
   Business context:
   - Business type: [BUSINESS TYPE]
   - Review date: [REVIEW DATE]
   - Ordering rules or target stock levels: [RULES, OR WRITE UNKNOWN]
   
   Stock data:
   [PASTE A TABLE OR CSV WITH SKU, ITEM NAME, CURRENT STOCK, STOCK ON ORDER, UNIT, LOCATION AND ANY RESERVED STOCK]
   
   Sales or usage data:
   [PASTE RECENT SALES OR USAGE DATA, INCLUDING THE PERIOD COVERED]
   
   Supplier data:
   [PASTE SUPPLIER, UNIT COST, MINIMUM ORDER QUANTITY, PACK SIZE, LEAD TIME, ORDER CUT-OFF, DELIVERY TERMS AND CONTACT DETAILS]
   
   Produce:
   1. A reorder table with SKU, item, current available stock, recent usage, assumed or provided lead time, reorder point, recommended quantity, supplier, estimated cost if the unit cost is provided, and the reason for the recommendation.
   2. A separate list of items that should not be ordered because data is missing, contradictory or stale.
   3. A list of every assumption and every figure you calculated, showing the formula used.
   4. A short list of anomalies, such as negative stock, duplicate SKUs, unusual usage or a quantity below a supplier minimum.
   5. A purchase-order draft grouped by supplier, clearly labelled DRAFT FOR HUMAN APPROVAL.
   
   Use the same units and pack sizes as the source data. If a reorder point or target level is not provided, do not make one up. Instead, show what information is missing and give a conditional calculation using the available figures. Do not present a recommendation as safe or final. End with a checklist for a colleague to verify against the live inventory system, current sales, supplier terms and cash budget before sending any order.

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.