Weekly trading reporting is one of those tasks that sounds simple until you are the person pulling it together.
You need sales, orders, conversion rate, AOV, product movement, discount usage, stock issues, channel performance and a view on what needs action. You also need it quickly, usually before the rest of the business starts asking why something is up, down or oddly flat.
This is where Claude can be genuinely useful. Not because it replaces a good ecommerce team, but because it can save time on the first pass of analysis.
Yes, you can ask for last week’s sales, your top products, your latest orders, or low-stock items. That is handy. But the more interesting opportunity is using Claude to ask better questions about the customer journey: where people arrive, what they browse, what they add to basket, where they hesitate, what they buy, and which channels seem to support trading performance.
Shopify gives you strong commercial and operational data. Claude gives you a conversational way to interrogate it. GA4 and Microsoft Clarity are the wingmen that help explain the behaviour behind the numbers.
Why does that matter? Because Shopify alone will not always tell you the full story. Shopify’s marketing reports can help you understand referrers, conversion paths and attribution models, but sales attributed to marketing only include trackable marketing activity, such as campaigns and activity using UTM parameters. Read more details at Shopify Help: Marketing reports.
GA4 adds event-level shopping behaviour, including product list views, product views, add-to-cart, cart views, checkout starts, purchases and promotion interactions. Developer documentation at Google Analytics: Measure ecommerce gives more details. Microsoft Clarity adds session recordings and heatmaps, helping you see where users click, scroll, pause and get stuck.
In other words, Shopify tells you what happened commercially. GA4 helps show how the journey behaved. Clarity helps you see what the customer actually experienced.
One warning before the prompts: if your tracking is weak, Claude’s output will be weak too. Poor CMP configuration, broken consent mode, missing GA4 ecommerce events and inconsistent UTMs all reduce the quality of the answer. Claude can help you move faster, but it still needs a decent view of the truth.
Before using the prompts below, tell Claude what data it can use:
- Shopify connector for products, orders, inventory, customers, discounts and Shopify analytics.
- GA4 exports or reports for events, traffic, landing pages, devices, funnels and channel behaviour.
- Microsoft Clarity screenshots, heatmap summaries, or session recording notes for visual friction.
- Google Search Console exports for queries, landing pages, impressions and organic CTR.
- Klaviyo or email campaign exports for campaign-level retention and revenue.
- Paid media exports for spend, ROAS, CPA and campaign intent.
Then ask Claude to separate what it knows from Shopify, what it knows from GA4, and what needs validation in another tool.
A weekly prompt to reuse.
Before you start asking individual questions, set the context:
“You are helping prepare a weekly ecommerce trading review. Use Shopify for sales, orders, products, customers, discounts, fulfilment and inventory. Use GA4 for traffic, acquisition, ecommerce events, landing pages, devices and funnels. Use Microsoft Clarity notes for visible user friction. Give me a practical summary for an ecommerce director, trading manager and merchandiser. Separate facts from assumptions. Tell me where the data looks incomplete. Focus on what changed, why it might matter and what we should check next.”
That prompt gives Claude a job. It also makes the output more useful for the people who need to make trading decisions, not just read analytics.
Weekly trading questions.
1. Which products are making the biggest contribution to revenue?
Prompt:
Using Shopify, show my top 20 products by revenue for the last 30 days. Include units sold, average order value contribution, discount usage and whether each product is currently in stock.
This is the basic trading view, but Claude can make it quicker to spot which products deserve more homepage, collection, email or paid support.
2. Which products sell often but pull down average order value?
Prompt:
Using Shopify, identify products with high order volume but low average order value contribution. Group them by product type and suggest where bundles, cross-sells, or threshold offers could help.
This is useful when revenue looks healthy, but the basket is not stretching far enough. Shopify gives the order data; the growth question is whether the product mix is helping or limiting the basket.
3. Which products are good basket builders?
Prompt:
Using Shopify order data, show products that are frequently bought with two or more other products. Suggest bundles, collection placements or PDP cross-sells based on those patterns.
This is where a product stops being a single SKU and becomes part of the journey. It can help inform bundles, recommended products, basket prompts and post-purchase email.
4. Which products are discount-dependent?
Prompt:
Using Shopify, show products where a high percentage of orders used a discount code. Compare revenue, units sold and average discount value against products that sell without discounting.
This helps separate genuine demand from promotion-driven demand. It also gives the team a better view of whether discounting is building value or training customers to wait.
5. What changed since the last trading meeting?
Prompt:
Compare this week against last week using Shopify. Show changes in revenue, orders, AOV, conversion rate, top products, discount usage and fulfilment status. Highlight anything that moved materially even if revenue stayed broadly flat. Finish with five points for the weekly trading meeting.
Revenue can hide a lot. Orders can rise while AOV falls. Conversion can improve while traffic quality drops. Discount usage can mask weaker buying intent. Ask Claude to pull the movement apart before the meeting starts.
Customer journey and conversion questions.
6. Where does the purchase journey appear to be leaking?
Prompt:
Using GA4 ecommerce events, build a simple funnel from view_item to add_to_cart to view_cart to begin_checkout to purchase. Show the biggest drop-off point and suggest what to investigate first.
GA4 recommended ecommerce events include view_item, add_to_cart, view_cart, begin_checkout and purchase, which makes this kind of journey question possible when tracking is set up properly. Read more details at Google Analytics Help: Recommended events.
7. How many visitors show no shopping intent?
Prompt:
Using GA4, estimate the percentage of users who visit but do not view a product, search, add to cart or start checkout. Break this down by landing page and channel.
This is one of the most useful ecommerce questions because it moves the conversation away from “we need more traffic” and towards “what kind of traffic are we bringing in?”
8. Which landing pages attract traffic but fail to start a shopping journey?
Prompt:
Using GA4 landing page data, show pages with high sessions but low product views, low add-to-cart rate and low revenue. Group by channel and suggest whether the issue looks like targeting, page intent or merchandising.
This is ideal for homepage, blog, collection and campaign landing pages. A page can look strong on sessions and still fail to move people into consideration.
9. Which PDPs get attention but not add-to-cart actions?
Prompt:
Using GA4, list product detail pages with strong views but weak add-to-cart rate. Then use Clarity heatmap or session notes to suggest what might be stopping users.
This is where Clarity earns its place. The numbers can show that a PDP is underperforming; recordings and heatmaps help explain whether users are missing sizing, struggling with imagery, hesitating on delivery, or simply not seeing the call to action.
10. Are shoppers using the basket as a wishlist?
Prompt:
Using GA4 and Shopify, compare add-to-cart, view-cart, begin-checkout and purchase behaviour. Identify signs that users are adding products to hold or compare rather than buy immediately.
Fashion, gifting and considered purchases often have this behaviour. It should influence basket messaging, email capture, wishlist prompts, price-drop messaging and stock urgency.
11. Is checkout friction linked to shipping or payment?
Prompt:
Using GA4 checkout events, compare begin_checkout, add_shipping_info, add_payment_info and purchase. Show whether the largest drop-off appears before shipping, after shipping, or around payment.
If the drop-off is around shipping, the issue could be delivery cost, delivery clarity or expected delivery date. If it is around payment, check wallet options, failed payments, 3DS friction, and express checkout visibility.
12. Do search users convert better than browsers?
Prompt:
Using GA4 site search and ecommerce events, compare users who search against users who do not. Show product views, add-to-cart rate, conversion rate, AOV and revenue per session.
Search is often a signal of intent. If search users convert well, improve search visibility. If they do not, investigate zero-result terms, predictive search quality, product naming and collection structure.
13. What are users searching for that we do not merchandise well?
Prompt:
Using GA4 search terms and Shopify product data, identify search terms with high volume but weak product engagement or low conversion. Suggest collection, product naming or content improvements.
This can feed SEO, merchandising and onsite search improvements. It is also a good way to find language customers use that the brand does not use clearly enough.
Channel and customer quality questions.
14. Which channels drive sessions, and which channels drive revenue?
Prompt:
Using GA4 and Shopify marketing reports, compare sessions, orders, revenue, conversion rate and AOV by channel for the last 30 days. Highlight where traffic volume and commercial value do not match.
This is a classic attribution question, but Claude can make the comparison faster. Paid social may drive volume but low conversion. Email may drive fewer sessions but stronger revenue. Organic may support consideration before another channel closes.
15. Where might channel reporting be under-selling a useful traffic source?
Prompt:
Using GA4 channel data and Shopify marketing reports, identify channels with high engagement but low last-click revenue. Suggest where first-click, linear or assisted consideration may change the interpretation, but flag anything that needs better attribution data before we trust it.
Shopify supports several attribution models, including last non-direct click, last click, first click, any click and linear, depending on the report configuration. That can help a trading team look beyond the final click, but it still depends on trackable journeys and consistent campaign tagging. Find more details at Shopify Help: Marketing reports.
16. Which campaigns are suffering from poor UTM hygiene?
Prompt:
Using GA4 acquisition data, find campaigns, sources or mediums that appear as unassigned, not set, inconsistent or duplicated. Suggest a clean UTM naming structure for future campaigns.
This is deeply unglamorous and extremely useful. If UTMs are messy, channel reporting becomes harder to trust, and AI analysis will inherit the same confusion.
17. Which channels bring shoppers back rather than introducing them?
Prompt:
Using GA4 and Shopify, compare new versus returning customer behaviour by channel. Show sessions, conversion rate, revenue, AOV, and repeat purchase patterns.
This helps avoid judging every channel by the same job. Some channels introduce customers. Some bring them back. Some close the sale. Some support retention.
18. Is email really underperforming, or is tracking under-reporting it?
Prompt:
Using GA4, Shopify and email campaign exports, compare email sessions, orders, revenue, campaign tagging and returning customer rate. Highlight whether the issue looks like campaign performance or tracking quality.
Email often looks weaker when UTMs are inconsistent or when flows and campaigns are not separated clearly. Do not assume the channel is failing before checking the tracking.
19. What channel should we investigate with Clarity first?
Prompt:
Using GA4, identify the channel and landing page combination with high sessions, low engagement or weak conversion. Recommend which Microsoft Clarity recordings or heatmaps to review first.
This makes Clarity less random. Instead of watching recordings aimlessly, point the review at the journey segment with the clearest commercial question.
Product, merchandising and stock questions.
20. Which high-demand products are being held back by stock?
Prompt:
Using Shopify inventory and GA4 product view data, identify products with strong product views or sales velocity but low or zero stock. Estimate the lost opportunity and suggest priority restocks.
Shopify can show stock and sales. GA4 can help show demand that did not convert because the product was unavailable, hidden, or not merchandised well.
21. Which collections help customers move into product consideration?
Prompt:
Using GA4, compare collection landing pages by product views, add-to-cart rate, revenue and exit rate. Identify collections that attract traffic but do not move users into PDPs.
This is a useful CRO and merchandising prompt. Collection pages are often treated like product grids, but they are decision-making pages in their own right.
22. Which homepage or promotional blocks appear to influence revenue?
Prompt:
Using GA4 promotion events, compare homepage or campaign blocks by view_promotion, select_promotion, product views, add-to-cart actions and revenue.
GA4 ecommerce measurement can include promotion views and promotion selections, allowing teams to understand the influence of promoted content and product placement on revenue. Find more details at Google Analytics: Measure ecommerce.
23. Which products need better content for SEO and conversion?
Prompt:
Using Shopify product data, GA4 landing page data and Search Console queries, identify products with impressions or traffic but weak conversion. Suggest content improvements for titles, descriptions, FAQs and internal links.
This is where SEO and CRO should stop living in separate rooms. If a product gets impressions but does not convert, the answer may be intent, content, price, trust, merchandising or stock.
24. What should we feature next based on seasonality and demand?
Prompt:
Using Shopify sales, GA4 landing page performance and Search Console query trends, suggest which categories or products should be prioritised in next month’s content, email and merchandising plan.
This is helpful for brands with strong seasonal patterns: gifting, sleepwear, bedding, swimwear, partywear, outdoor, supplements, skincare or replenishment-led products.
Retention and customer value questions.
25. Which customers or segments are most likely to buy again?
Prompt:
Using Shopify customer and order data, identify customer segments with strong repeat purchase behaviour. Group by product type, first purchase category, discount usage and time between orders.
This can feed email flows, replenishment prompts, subscription ideas, loyalty, bundles and remarketing audiences.
26. Which first purchase products lead to better lifetime value?
Prompt:
Using Shopify order history, compare first purchase products or categories by repeat purchase rate, second order value and time to second purchase.
This is a better question than simply “what sells most?” A product that introduces valuable customers may deserve more support than a product that generates one-off orders.
27. What should we test next?
Prompt:
Using Shopify, GA4 and Clarity evidence, create a prioritised CRO backlog. For each idea, include the observation, hypothesis, page or journey stage, expected impact, effort and the metric we should monitor.
This is the question that turns analysis into work. The output should not be a long list of opinions. It should be a prioritised set of actions tied to a measurable part of the journey.
How to get better answers from Claude.
The quality of Claude’s answer depends on the quality of the question and the quality of the data.
Use this structure:
“You are reviewing weekly ecommerce trading performance. Use Shopify for orders, products, customers, inventory and revenue. Use GA4 for event-level behaviour, traffic, landing pages, devices and funnels. Use Clarity notes for visual friction. Separate facts from assumptions. Highlight tracking gaps. Recommend the next action, not just the observation.”
Then give Claude:
- A clear date range.
- The market or region.
- The channel or campaign if relevant.
- The page type or journey stage.
- The metric you care about.
- The data sources available.
- Permission to say “the data is not available”.
That last point really matters. We want Claude to be useful, but we also want it to admit when Shopify, GA4 or Clarity do not contain the evidence needed to answer properly.
How do ecommerce agencies use AI?
The most valuable use of Claude is not asking for generic ecommerce advice. It is using AI to get to a usable first view of trading performance faster.
For us, the stronger questions tend to sit around:
- Whether tracking is complete enough to trust the story.
- What changed week on week, beyond headline revenue.
- Which products, collections and campaigns need attention.
- Which users show genuine shopping intent.
- Which channels drive valuable customers rather than just sessions.
- How merchandising, content, email and paid activity support the same customer journey.
- What should be tested next, and why.
That is also where Shopify, GA4 and Clarity work best together. Shopify gives you the commercial outcome. GA4 gives you the behavioural pathway. Clarity gives you the visible friction. Claude helps turn that into questions your team can act on.
It still needs human judgement. But used properly, it can shorten the distance between data, insight and action.
Ready to make your store AI-ready?
Want to learn more about how you can use Claude for your Shopify store? Check out these two related posts: Claude, Attribution Tools & Dashboards Won’t Fix Bad Ecommerce Data and How to Connect Your Shopify Store to Claude for Faster Trading Insight.
If you want to use Claude to speed up your weekly trading reports but need to ensure your GA4 and Shopify tracking is actually capturing the full truth first, let us know! At blubolt, we build and optimise ecommerce stores. Our clients get the revenue. As a Shopify Platinum Partner agency, we ensure your data architecture and UTM hygiene are flawless, so your AI tools analyse real customer behaviour rather than broken tracking gaps. Explore our services and browse our work to see how we’ve built rock-solid foundations for other brands, or get in touch to chat with our team about turning your ecommerce data into actionable trading insights.
