Use case B23 · Business
Ask AI about your sales history to decide what to make, stock and promote: a candle maker and a craft store
- Finance for owners
- Beginner
- Not named (AI), Shopify, Trello, AI tools (not named individually)
- Source checked · news article
The short version
What they did
Two shop owners ask AI questions about their own sales history instead of building reports, and use the answers to decide what to make, what to buy and what to promote.
A candle maker deciding what to pour and what to buy
Albert Feldman, whose Boston company hand-pours its candles, uses AI on historical sales data to decide which products to make more or less of and how many wax and scent vessels to buy. He also uses it for financial planning and marketing campaigns. The article does not name the tool. CBS News, 2026-06-17.
A craft store pulling sales answers in seconds
Brittany Naylor uses AI inside tools like Shopify and Trello to analyze sales data, send personalized email campaigns and organize workshops, so staff stay on the shop floor. Denver7, published 2025-11-26 at 21:46 Mountain time (2025-11-27 UTC), confirmed in article metadata.
Result (as claimed)
The candle maker: “It helps us understand what products we should make more of and which we should move away from.” And: “By leveraging historical sales data, it helps us know how many wax and scent vessels we need to buy, and helps us make heads and tails of everything.” No figures are given.
The craft store: “Other businesses would have a whole data analyst dedicated to telling me that information,” Naylor said after asking AI for sales data for a product and receiving it seconds later. “Or, I could run a report and try to figure that out myself, it would probably take me, personally, like days.” Denver7: “Naylor estimates the roughly 15 tools combined cost around $500 a month, but save the staff countless hours of work and save the business thousands of dollars in payroll.”
Prompt
The candle maker:
I make handmade candles and want help planning production and buying materials. Two years of sales by product and month are [attach]. Materials used per product, their costs and current stock are [attach], and any known unusual orders or stockouts are [describe].
1. Rank products by profit where costs support it, and show sales trends and last year’s seasonal peaks.
2. Suggest which products to make more or less of, explaining the evidence and uncertainties.
3. Estimate wax, fragrance and vessel purchases for the next three months. Show planned units, material calculations and stock deducted.Return clear tables and a summary under 200 words. Don’t treat past sales as guaranteed demand or invent missing costs and stock. Ask for needed figures, and label assumptions so I can check them before ordering.
The craft store:
I run a craft store and want to understand sales of [product] before planning a promotion. My last 90 days of sales data are [attach dated transactions with product, quantity and revenue]. Our workshop date and details are [paste], and our email tone is [describe].
1. Show weekly units and revenue for the product. Identify items bought with it only if order-level data allows that comparison.
2. Compare sales before and after the workshop, using clearly stated periods. Don’t assume the workshop caused a change.
3. Suggest one email campaign for customers who bought it, with a subject line and draft under 120 words.Return a short table and the draft. Mark missing data and ask questions rather than inventing purchases, discounts or customer interests. Don’t send the campaign.
Do it yourself
These steps are my suggested way in, not any one person’s setup.
What you need
- A chat assistant that can analyze spreadsheets, such as ChatGPT or Claude.
- A sales export by product and month from your store or point-of-sale system.
- For makers: what each product uses in materials and what those cost.
Start simple
- Export one to two years of sales by product and month.
- Upload it and paste the prompt above that fits you.
- Ask for one chart of the top and bottom ten products by profit.
- Pick one product to make or stock more of and one to cut. Write down why.
- Check the result after a month and ask it what changed.
Take it further
- Use the AI tools already built into your store platform, if it has them, to ask questions without exporting.
- Ask for a buying plan for the next three months that allows for last year’s busy season.
- Pair the sales answers with an email campaign to the customers who bought a slow product.
Check your work
- Check the totals against your store’s own sales report.
- Watch out for one-time spikes, like a big wholesale order, that skew the trend.
- Order materials in steps. Don’t place a large order on a forecast you haven’t checked.
Before you rely on it
- This is general information, not financial, tax or accounting advice for your situation.
- AI tools make arithmetic mistakes and misread statements, and they sound just as sure when they do. Check every number against your own records before you act on it.
- For decisions with tax, loan, insurance or retirement consequences, talk to a qualified professional who knows your situation.
Source checked. The link, author, date and quote were confirmed against the source. The result itself has not been reproduced. Prompt adapted for this library, inspired by the case. Business size: Not stated; single store with staff. Results are quoted as the source claimed them and never improved. Spot an error or a broken link? Email charlie@cbarmorecpa.com.