TL;DR: Once you connect Shopify to Claude, the hard part isn’t the setup — it’s knowing what to ask. This is a list of 12 plain-English questions Shopify owners actually use, grouped by what you’re trying to figure out: revenue health, what to restock, who your best customers are, and which marketing actually paid off. Each one replaces a multi-step dashboard hunt with a sentence.
Most Shopify owners who connect Shopify to Claude (or to Datavessel’s chat, which uses the same engine) make the same mistake in week one: they ask “show me my sales.” Claude returns a number. They don’t know what to do with it. They close the chat and go back to the Shopify admin.
The problem isn’t the integration. The problem is that “show me my sales” is a dashboard question, not a chat question. Chat is good at answering the questions that dashboards make hard — the ones that need cross-referencing, comparisons, or “and what about” follow-ups. The 12 questions below are the ones store owners come back to once they get the rhythm of asking a chat instead of clicking a dashboard.
This post assumes you’ve already wired things up. If you haven’t, our step-by-step guide to connect Shopify to Claude covers the five-minute setup. Come back here when you’re staring at an empty chat and wondering what to type.
Revenue Questions That Beat the Sales Dashboard
Shopify’s revenue dashboard shows you a number and a chart. It does not tell you whether that number is good. These four questions do.
1. “How does this week compare to the same week last month, and why?”
The dashboard shows week-over-week. That’s not useful — last week was probably similar. Comparing to a month ago surfaces real trend changes, and asking why forces the agent to look at order count, AOV, and traffic source breakdowns to explain what moved. You get a sentence like: “Revenue is up 22% vs four weeks ago — order count is flat, but AOV jumped from $58 to $71, driven by more multi-item carts.”
2. “What’s my AOV trend over the last 90 days, and which collections moved it?”
Average order value is the single most ignored Shopify metric. Most owners check it once and forget. Asking the trend plus the cause (which collections) tells you whether your bundling, upsells, or new product mix is actually working.
3. “Did discount code SAVE10 actually grow revenue or just cannibalize full-price sales?”
This is the classic discount-code question every store owner avoids because the answer requires cross-referencing pre-discount and post-discount baselines. Claude does it in one query: compares customer cohorts and tells you whether the code pulled in new buyers or just gave existing ones a discount they’d have paid full price for.
4. “Show me revenue by traffic source for the last 30 days, ranked.”
Shopify’s reports separate this poorly. Asking in chat returns a clean ranked list (organic, paid social, email, direct, referral) with revenue and order count side by side — the breakdown that tells you which channel deserves more budget.
Inventory and Restock Questions
The inventory tab is the part of Shopify admin everyone underuses. These three questions surface what it would normally take three filtered views to find.
5. “Which products are about to sell out at current sales velocity?”
“Low stock” filters by quantity, not by velocity. A product with 50 units that sells 80 a week is more urgent than a product with 10 units that sells one a month. Asking for velocity-based restock urgency is the question Shopify can’t answer natively — and the one that prevents stockouts.
6. “Show me products that haven’t sold a unit in 60 days but still have inventory.”
Dead stock. Capital tied up in your warehouse. Most owners don’t audit this until they do a year-end clean-up. Asking once a month catches it early enough to discount, bundle, or pull from the catalog.
7. “Which product variants have the highest return rate, and what’s the reason?”
If return reasons are tagged, Claude pulls them. If they’re free-text, Claude summarizes the themes. Either way, you get “size XL of product X has a 14% return rate, mostly ‘runs small’” instead of scrolling through individual returns.
Customer Questions That Reveal Who Actually Matters
“Customer reports” in Shopify is a list of names with order totals. These three questions turn that list into actionable segments.
8. “Who are my top 10 repeat customers and what did they buy?”
Knowing the names matters less than knowing the pattern. If eight of your top ten bought from the same collection, that collection deserves a dedicated email or landing page. Repeat-customer purchase patterns are often the cheapest growth insight on the store.
9. “What’s the average time between first and second purchase, and which products drive it?”
This is the metric that defines whether your store has retention. If second purchases come within 30 days, you have a flywheel. If they come at 180 days or never, your acquisition costs are working against you. The follow-up — which products drive the second purchase — tells you what to promote in post-purchase emails.
10. “Show me customers who spent over $500 last year but nothing in the last 90 days.”
Lapsed VIPs. The most valuable email segment most stores never build. One question identifies them; a follow-up like “draft a win-back email subject line for this segment” turns the answer into action.
Marketing and Content Questions That Cross GA4 + Shopify
These two are the questions that require both your analytics and your store to be connected — the kind that justified wiring up Shopify with Claude in the first place.
11. “Which blog post drove the most visitors who actually bought something this month?”
This is the question every content marketer wants the answer to and almost nobody has, because it requires joining GA4 attribution data with Shopify order data. With Datavessel, Claude does the join. The answer is a sentence: “Your ‘How to choose a CRM’ post drove 47 purchases — 3× your next-best post.” Now you know which content topic is actually paying for itself.
12. “Compare this month’s Instagram-sourced revenue to last month, and tell me if it’s worth increasing ad spend.”
Channel ROI question. The agent looks at revenue, order count, conversion rate, and AOV from Instagram traffic across both periods, then gives you a recommendation — not just numbers. The “should I spend more” framing forces the answer into a decision instead of a data dump.
Why You Connect Shopify to Claude in the First Place
Look at the list again and notice what the dashboard-friendly questions have in common. They all need at least one of: a comparison, a calculation, a cross-source join, or an interpretation. Shopify’s admin is built to show you raw data. Chat is built to answer the question on top of the data. That’s the entire reason to connect Shopify to Claude.
If you’re using Datavessel, every one of these questions also works against Google Analytics and Search Console in the same chat — so you can chain “which blog post drove the sales?” into “and what search terms brought visitors to that post?” without changing tools. The Shopify integration is one source; the conversation crosses all of them.
What Not to Ask
Two patterns that waste the integration:
- “Show me sales.” — Too vague. The agent picks a default time range and you get a number. Ask for a comparison, a breakdown, or a “why” instead.
- “Export all orders from 2025 to a spreadsheet.” — That’s a Shopify admin job. Chat is for questions, not data dumps. Use Shopify’s bulk export when you need rows.
Everything in the middle — anything that needs thinking, comparing, or joining — is where the integration earns its keep.
Try It
Already connected Shopify to Claude? Pick three questions from above, paste them into the chat, see what you get back. If you haven’t connected yet, start free at datavessel.io — the Shopify integration takes under five minutes and the AI account stays on your own Claude, ChatGPT, or Gemini bill (no markup).
Sources
- Connect Shopify to Claude: Setup Guide — Five-minute step-by-step setup if you haven’t wired the integration yet.
- Shopify Integration Announcement — Original announcement covering the underlying MCP-based connection.
- Talk to Your Google Analytics in Plain English — The companion piece for the GA4 side of the same conversation.
Leave a Reply