Nov 13, 2026 · 15 min read
30 ChatGPT & Claude Prompts for Shopify Data Analysis
The hard part of analyzing your Shopify data with ChatGPT or Claude is not the connection. It is knowing what to ask. These are 30 paste-ready prompts for Shopify data analysis, grouped by profit, ads, retention, inventory, and customers. Each one pulls a decision out of your store data instead of another surface-level chart.
Every prompt comes with a one-line reason to run it and a ready-to-paste block you can copy straight into ChatGPT or Claude. Fill the bracketed parts, like [your gross margin], with your own numbers before you send it, because the model cannot infer them. For how to connect your store and make these answers hold up, see the ChatGPT Shopify guide. For where they fit in the bigger picture, see how D2C brands use AI for analytics.
How to Use These Prompts
Copy the prompts that match a decision you are about to make, not all 30 at once. A prompt is only worth sending if the answer would change something you do this week.
The fastest way to run them is against a live connection. Shopify's connector for AI assistants, or a ChatGPT or Claude connector, lets a model read your current orders instead of a stale CSV, and the Shopify to Claude setup covers that read-only connection if you use Claude. A file export still works as a fallback for a one-off, but it goes stale the moment you download it. Either way, two habits make every answer better. First, give the tool the business context it cannot infer, like your true margin, your return window, or what you count as a new customer. Second, ask it to show which fields and rows produced each number, so you can catch a wrong join before you act.
Profit Prompts
Revenue prompts are easy and mostly useless on their own. These push past revenue to what you keep.
1. Your real profit leaders, not your revenue leaders. Top sellers by revenue are often not your top earners after cost of goods. See gross margin.
Using my connected Shopify store, list the 10 products with the highest gross profit over the last 90 days, not the highest revenue. My average gross margin is [e.g. 45%]; use it unless I paste per-product costs. Show units sold, revenue, estimated COGS, and gross profit, and return a table sorted by gross profit from highest to lowest.
2. High-volume products that lose money. Strong unit sales can hide a negative contribution margin once ads and returns come out.
From my connected Shopify orders over the last 90 days, find products with high units sold but low or negative contribution margin after COGS, returns, and allocated ad spend. My gross margin is [e.g. 45%] and my average return rate is [e.g. 8%]. List the 15 highest-volume products and flag any whose contribution per unit is near zero or negative. Show units sold, return rate, and contribution per unit in a table.
3. True net profit per order. The number most stores never compute, and the one that decides if growth is worth it. Work it through with the net profit per order guide.
Using my last 90 days of Shopify orders, calculate average net profit per order after COGS, shipping, payment fees, and ad spend. Use these assumptions: gross margin [e.g. 45%], average shipping cost per order [$6], payment fees [2.9% + $0.30 per order], and total ad spend for the period [$X]. Show revenue, each cost line, and net profit per order, and state which fields you used.
4. Discounted revenue and its real margin. Shows whether promotions are buying growth or buying orders you would have gotten anyway.
Group my last 90 days of Shopify orders into those that used a discount code and those that did not. For each group show order count, total revenue, average order value, and average gross margin using my [e.g. 45%] baseline margin minus the discount. Return a two-row table and tell me how much margin the discounts cost in total.
5. Discount codes that lose money. A single over-generous code can run at a loss for months. Check break-even with the margin calculator.
Group my last 90 days of Shopify orders by discount code. For each code show orders, total discounted revenue, and average order value after the discount. My blended gross margin before discounts is [e.g. 45%]. Flag any code whose average discounted order falls below break-even at that margin, and sort the table worst first.
6. What moved your blended margin. A drift of two or three margin points across the catalog is easy to miss and expensive to ignore.
Compare my blended gross margin this month to last month using my connected Shopify store and a baseline margin of [e.g. 45%] adjusted for discounts. Tell me the margin for each month, the change in points, and the three products or collections that moved it most. Return a short table of those drivers.
Advertising Prompts
Platform-reported numbers flatter every channel. These ask your Shopify data to check the ad platforms' homework.
7. Real blended CAC. Platforms double-count buyers they both touched, so their summed conversions run ahead of your real new-customer count. See blended CAC.
Using my connected Google Ads, Meta Ads, and Shopify data for last month, calculate blended CAC. Sum total ad spend across Google and Meta, then divide by the count of distinct first-time Shopify customers in the same period. Do not use platform-reported conversions. Show total spend, the new-customer count, and the CAC, and list the fields you used.
8. Channels that bring repeat buyers. A cheap first order from a channel that never repeats is more expensive than it looks.
Using my Shopify orders, group customers by the channel of their first order. For each channel show how many placed a second order within 90 days and the repeat rate as a percent. Return a table sorted by repeat rate, and note which channel looks cheap on first order but rarely repeats.
9. Real ROAS after returns and discounts. A campaign showing 3x on gross revenue can land near break-even once returns and a launch discount come out.
For each ad channel last month, calculate ROAS two ways using my connected Shopify and ad data: gross revenue ROAS, and net ROAS after returns and discount codes are removed. Show spend, gross revenue, net revenue, and both ROAS figures in a table, and tell me which channel drops the most between the two.
10. Paying to reacquire loyal customers. Paying for buyers who would have returned anyway is a common, invisible waste.
Using my connected ad and Shopify data for last quarter, split each campaign's attributed orders into first-time customers and returning customers. Show spend, first-time orders, and returning orders per campaign, and flag campaigns where most orders came from customers who had already bought before.
11. Platform-claimed vs real new customers. When platform math exceeds reality, you are over-crediting ads. The revenue mismatch guide covers why.
For last month, compare the number of new customers Google and Meta each claim to the distinct first-time customers recorded in my connected Shopify store. Show each platform's claimed figure, the summed total, the actual Shopify count, and the gap as a percent. Do not treat platform conversions as new customers.
12. CAC payback window. Tells you how long your cash is tied up before a new customer turns profitable.
Using my connected Shopify data, estimate how many months it takes to recover my customer acquisition cost. Use CAC of [$X] and contribution margin per order of [$Y], plus the average number of orders a new customer places in their first 6 months. Show the payback in months and the assumptions you used.
Retention Prompts
Acquisition gets the attention. Retention math is usually where the profit hides.
13. Repeat purchase rate trend. A falling repeat rate means you are refilling a leaking bucket. See repeat purchase rate and what a good one looks like.
Using my connected Shopify store, calculate my monthly repeat purchase rate for each of the last 6 months, defined as the share of customers with two or more orders. Show the rate per month in a table and tell me the direction of the trend.
14. First-to-second order gap. Sets the timing for your win-back and replenishment emails.
From my Shopify order history, calculate the median number of days between a customer's first and second order. Show the median, the 25th and 75th percentiles, and a short distribution, so I can time a win-back email around the typical gap.
15. Returning vs new revenue split. A healthy split signals a real base, not just paid growth. See the new versus returning revenue split.
For this month and last month, split my Shopify revenue into new customers and returning customers. Show revenue and order count for each group, the returning share as a percent, and whether the split moved between the two months.
16. Your win-back list. Your highest-value lapsed customers, sitting in the data unused.
Using my Shopify customer and order data, find customers who spent more than [$300] in total but have not ordered in the last 90 days. Return the top 50 by lifetime spend as a table with email, lifetime spend, order count, and days since last order.
17. LTV by acquisition channel. Some channels bring browsers, some bring buyers. See lifetime value.
Using my connected Shopify data, calculate average 12-month lifetime value by the channel of each customer's first order. Show customers acquired, average LTV, and average orders per customer per channel in a table sorted by LTV.
18. Best entry products. Points to the products worth pushing as a first purchase, not just the ones with the best first-order margin.
From my Shopify orders, group customers by the first product they bought and show the share who placed a second order within 90 days. Return the 15 first products with the highest repeat rate, with customers and repeat rate in a table.
Inventory Prompts
Cash is stuck in stock. These prompts find the stuck cash and the coming stockouts.
19. Slow movers and reorder risk. Your reorder-and-markdown list. See sell-through rate.
Using my connected Shopify store, list the 15 products with the lowest sell-through over the last 60 days. Include only products stocked for the full 60 days. Show units sold, units on hand, and sell-through as a percent of starting inventory, and sort ascending by sell-through.
20. Stockout risk on winners. A stockout on a fast seller is lost revenue you never see in a report.
From my connected Shopify inventory and sales, find products likely to run out in the next 30 days at their current 30-day sales velocity. Show units on hand, units sold in the last 30 days, implied days of cover, and projected stockout date, sorted by soonest stockout.
21. Cash tied up in dead stock. Dead stock is a balance-sheet problem hiding as a shelf. Work through clearing it in the dead stock guide.
Using my Shopify data, list products that have not sold a single unit in the last 90 days but still have stock on hand. Estimate the cash tied up using unit cost of [paste costs or an average], and return a table of the worst offenders by tied-up cash.
22. Inventory turnover by category. Shows where restock discipline is loosest. See inventory turnover.
Using my connected Shopify store, calculate inventory turnover for the last 90 days by product type or collection. Show units sold, average units on hand, and turnover per category in a table, and tell me which category turns slowest.
23. Products eaten by returns. A high return rate erases the margin on an otherwise strong product. See return rate.
From my last 90 days of Shopify orders, list the products with the highest return rate by units, excluding any product with fewer than [25] units sold. Show units sold, units returned, and return rate as a percent, sorted highest first.
24. Dead variants. Trimming variants that never sell frees cash without losing real sales.
Using my Shopify product and order data, for my top 20 products by revenue, show how each variant sells. List units sold and units on hand per variant, and flag variants with near-zero sales but stock on hand.
Customer Prompts
Averages hide your best and worst customers. These prompts split the base.
25. Your top 20 percent of customers. Your revenue concentration, and the profile to acquire more of. Walk through it in identify your top 20 percent.
Using my Shopify customer data, identify the top 20 percent of customers by lifetime spend. Show what share of total revenue they represent, their average order value, average order count, and the most common first product and acquisition channel among them.
26. AOV by segment, and where it is slipping. A falling AOV in your best segment is an early warning worth catching. See average order value.
Using my connected Shopify store, calculate average order value by customer segment, split by new vs returning and by first-order channel, for this month and last month. Return a table with AOV per segment per month and the change, and flag any segment where AOV is falling.
27. Discount-only customers. A segment that never pays full price is not the growth it looks like. See discount-dependent customers.
From my Shopify orders, find customers whose every order used a discount code. Show how many there are, their share of all customers, and their share of revenue, and compare their average margin to customers who sometimes pay full price.
28. Where your best customers come from. Tells you what to spend more on to get more of them.
For the top 20 percent of my customers by lifetime spend, show the most common acquisition channel and first product. Return a table of channels ranked by how many high-value customers they produced, so I can see which source to lean into.
29. Third-purchase rate. The second order can be luck. The third is a habit forming.
Using my Shopify order history, calculate the share of customers who go on to place a third order. Break it down by the quarter they were acquired, so I can see whether newer cohorts are forming the habit faster or slower than older ones.
30. Customers drifting toward churn. Catching a slipping customer before they leave is cheaper than winning them back.
Using my Shopify data, find customers whose time since last order is now longer than their own typical gap between orders. Return the top 50 by lifetime spend with email, average days between orders, days since last order, and how far past their normal gap they are.
Turn the List Into a Habit
A one-time answer is worth less than a standing one. The prompts worth running every week belong in a report that arrives on its own, not in a spreadsheet you rebuild each Monday.
Group them by cadence. Profit and ad-efficiency prompts are weekly, because a bad code or a margin slip costs money fast. Retention and customer prompts are monthly, because the numbers move slowly and weekly noise misleads. Inventory sits in between, tied to your reorder cycle. For the exact set of numbers a weekly report should carry, see the AI weekly report spec.
Frequently Asked Questions
What are the most important prompts to run on my Shopify data?
Start with profit and retention, because they are most often missed. The two highest-value prompts are usually true net profit per order after all costs, and repeat purchase rate over time. Revenue and traffic are easy to see and rarely the problem. Profit tells you if growth is worth it, and repeat rate tells you if you are building a base or refilling a leaking bucket.
How do I ask my Shopify data these prompts without a data analyst?
Connect your store to an AI assistant that reads live data, then paste a prompt and fill in your own numbers. The barrier used to be joining and filtering across sources, not the math, and stating the conditions in words removes it. Give the tool the business context it cannot infer, like true margin, and ask it to show which rows produced each number so you can verify before acting.
Why does a profit prompt return the wrong products?
Because Shopify does not store your landed cost, so a tool reading raw orders ranks by revenue and labels it profit. Your best seller by units can be a mediocre earner after COGS, and a lower-volume item can be your real margin leader. Supply the cost, shipping, and fee assumptions in the prompt, or the answer will confidently point you at the wrong products.
How often should I run these prompts?
Group them by how fast the numbers move. Profit and ad-efficiency prompts are worth a weekly look, because a bad discount code or a margin slip costs money fast. Retention and customer prompts are better monthly, since weekly swings are mostly noise on small volume. Inventory prompts track your reorder cycle. Running a slow-moving metric too often just adds noise.
Can I trust the answers an AI gives to these prompts?
For direction yes, for the final number verify it. A model reading your store can rank slow movers or flag money-losing codes well, but it will produce a confident figure even when the data cannot support one, and it can join sources wrong. Ask it to show its fields and rows, and confirm any number you plan to spend or reorder against directly in Shopify before you act.
Conclusion
The gap between stores that grow and stores that stall is rarely the data. It is whether anyone asks it the right questions. Copy the six prompts that map to a decision you face this week, feed the tool your real costs, get grounded answers, and act.
For how to connect a store and run these reliably, see the ChatGPT Shopify guide. To make the recurring ones automatic, see the weekly report spec.
About the author
