As of 13 August 2026, AI can only partly create a stock purchasing budget.

What to actually do

1. Use a tool built for this

   The route this page recommends

2. Do it yourself

   Second choice

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

   How to actually do it

   1. Open a spreadsheet and create tabs for products, sales, current stock, purchase orders, suppliers, cash limits and the final budget.
   2. Gather the latest stock count, open purchase orders, supplier price lists, minimum order quantities, order multiples, lead times, payment terms and delivery charges.
   3. Export sales by product and period from your till, ecommerce or accounting system, then remove cancelled orders and label promotions, seasonal periods and unusual one-off sales.
   4. Paste the cleaned tables and your cash limit into the prompt, replacing every bracketed slot and stating any information you do not have.
   5. Copy the AI's recommended order quantities, costs, arrival dates, assumptions and scenarios into the final budget tab, keeping the source data beside each calculation.
   6. Compare every product cost, stock balance, open order, supplier term and lead time in the draft against the original records and current supplier quotations.
   7. Ask a colleague responsible for purchasing or finance to challenge the demand assumptions, cash commitment, shortage risks and overstock risks before any order is placed.
   8. Amend the approved quantities in the spreadsheet, record who approved them and send only the final purchase instructions to suppliers.

   Prompt

   Create a stock purchasing budget for [business name] covering [budget period]. Use only the data I provide and do not invent figures or assumptions. If a required input is missing, list it clearly and show how it affects the result.
   
   Business constraints:
   - Available cash for stock: [amount and currency]
   - Target gross margin or pricing rule: [details]
   - Desired service level or acceptable stockout risk: [details]
   - Storage, shelf-life or product constraints: [details]
   - Existing purchase commitments: [details]
   
   Product data:
   [Paste a table with SKU, product name, unit cost, selling price, current stock, stock already on order, supplier, minimum order quantity, order multiple, lead time, shelf life and any discontinued or seasonal status.]
   
   Demand data:
   [Paste historical sales by SKU and period, current orders, promotions, known seasonal changes and any approved sales forecast.]
   
   Supplier data:
   [Paste supplier prices, price breaks, delivery charges, payment terms, lead times and order cut-off dates.]
   
   Produce:
   1. A purchasing budget by SKU and period with recommended order quantity, unit cost, total cost, expected arrival and supplier.
   2. The assumptions and source data used for every recommendation.
   3. A cash commitment summary, including stock already on order and planned purchases.
   4. A list of items at risk of shortage, overstock or unreliable demand.
   5. Three scenarios labelled conservative, base and higher demand, using only ranges or assumptions I explicitly provide.
   6. A list of missing data and questions that must be answered before placing an order.
   
   Keep calculations transparent and show the formulas or logic for reorder quantities. Do not place orders, claim that demand is certain, or recommend borrowing. Format the result so it can be copied into a spreadsheet. End with a short checklist for a colleague to verify against the source records and supplier quotations.

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

3. Hand it to a person

   The distant third

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