Most ecommerce reporting stops at what happened. This free prompt reads your Shopify, ad, email and account-change data and writes a weekly brief that explains why the numbers moved, what is coming and what to do about it, with an owner for every action.
On this page
Most ecommerce reporting tells you what happened last week and stops there. Sales were up 12%, Meta spend was down, email revenue dipped, and you're left to work out the why, the so-what and the next move yourself, usually on a Monday morning with three dashboards open and a call in twenty minutes.
The Ultimate weekly business brief is a free AI prompt that does that work for you. You paste it into Claude or ChatGPT, name your store, and it pulls your sales, margin, ad, email and account-change data, then writes a brief that says what happened, why it happened, what's coming and what to do about it, in that order. It started life as an internal test, but the brands using it found it so useful that we've published it in full.
This post covers what ecommerce reporting should actually do, what the brief includes section by section, the rules that stop it making things up, and how to set it up in about ten minutes.
What is ecommerce reporting?
Ecommerce reporting is the process of turning raw store data into summaries a team can act on, built around specific business questions rather than every metric you can collect. Shopify's guide to ecommerce reporting puts it the same way: the point is to answer questions like "why did conversion drop last week?", not to fill a dashboard.
Most brands have plenty of reports. Shopify has its own, every ad platform has its own, and the email tool has another. What they rarely have is one weekly sales report that joins them up and explains the movement.
Why most weekly sales reports don't change anything
A typical weekly ecommerce performance report is a list of numbers with arrows next to them. Revenue up, ROAS down, conversion rate flat. Three problems come with that format.
- It reports, it doesn't explain. An arrow tells you a number moved, but not whether it was a product mix shift, one bulk order, a discount ending or someone in the ad account changing a budget on Thursday.
- It treats every number as equally important. Twenty metrics with equal weight means the one that matters gets the same attention as the nineteen that don't.
- It looks backwards only. Last week is useful, but the more valuable question is what happened at this point last year and what you need to prepare for in the next three weeks.
The result is a report that gets skimmed and filed. The team meeting then spends its time rebuilding the story the report should have told.
What the weekly business brief does differently
The brief is written by an AI acting as a senior ecommerce analyst, with one instruction at its core: don't report numbers, tell the operator what happened, why it happened, what's coming and what to do about it. The front of the brief carries the conclusions, and everything after it is evidence.
| Typical weekly report | The weekly business brief |
|---|---|
| Lists every metric with an arrow | Leads with a 60-second summary: headline, biggest win, biggest risk, what's coming |
| Compares to last week only | Compares to last week, the same week last year, and each campaign's own 3-week baseline |
| Shows ROAS | Shows contribution margin, Marketing Efficiency Ratio and your break-even point, with the cushion above or below it |
| Silent on who changed what | Lists every human change in Meta, Google Ads and Klaviyo, and whether it lined up with performance |
| No actions, or generic ones | 3 to 5 actions, each tied to a finding in the brief and given a named owner |
| Ignores the calendar | Checks what spiked at this point last year, verifies why, and tells you how to prep |
The brief also runs monthly. Ask for "the month of September" and every comparison switches from week-on-week to month-on-month.
What's in the brief, section by section
The full brief is about two printed pages, built so that someone who stops after the first screen still has the complete picture.
A 60-second executive summary
One bold lead sentence, then up to five short lines: the headline number, any one-off distorting it, the biggest win, the biggest risk and what's coming up. The prompt includes an example of the shape it's aiming for:
August grew 36% month-on-month, but a third of that is one order. A single $32,800 bulk order drives much of the year-on-year margin story; without it, contribution margin is still up 25%. Biggest risk: Klaviyo sends nearly doubled while revenue per recipient fell 57%, which is consistent with list fatigue and needs a check before the next volume spike.
That caveat line matters more than it looks. One wholesale order can make a flat month look like a breakout, and the brief separates it out before anyone celebrates.
Scorecard and break-even bar
Four cards: net sales, marketing spend, Marketing Efficiency Ratio and contribution margin, each with the week-on-week and year-on-year change. Under them sits a break-even bar that runs from gross margin to your break-even Marketing Efficiency Ratio to where you actually are, so you can see your cushion at a glance. Contribution margin leads in currency, not just as a percentage, for the reasons in our contribution margin ratio vs dollars breakdown.
This week's actions, with owners
Three to five numbered actions, highest impact first. Each one opens with a one-line instruction, gives two sentences of why that point back to a specific finding, and names an owner. Where the account history shows who runs a channel, it uses that person or agency; where it doesn't (Klaviyo has no actor field), it assigns the role, such as lifecycle or email.
Who changed what
A change log for Meta, Google Ads and Klaviyo: which people made which changes, to which campaigns, and whether any of them lined up with a performance shift. Automated changes, recommendation auto-applies and scripts are kept apart from people. It builds on Google's own change history and Meta's account history, and it works the same way as our standalone ad account change history audit.
Channel statuses and this week's standouts
One module per connected channel (Meta, Google, Klaviyo, TikTok) with a green, amber or red status, three key numbers and one insight. Then a short standouts section: at most one product, one campaign, one discount code, one email and one unusual event, each marked repeatable or one-off. If nothing stood out in a category, it says so rather than padding.
What happened last year at this point, and what's coming
The brief looks at the next three weeks of last year's data, finds any spike, identifies what drove it and checks the date is from the right year before it names a cause. A bank holiday that moved, or a tournament held in a different year, is exactly the sort of thing a quick read gets wrong. Weather only appears if your category is genuinely weather-sensitive and conditions were unusual.
The rules that stop it guessing
AI reports have a reputation for confident nonsense, so most of the prompt is rules about what the brief is not allowed to claim.
- Every cause gets checked against a second data source before it's stated as fact. A sales jump is only "driven by Meta" if Meta's own numbers and the order data agree.
- "Coincided with", not "caused". A budget change made on the last day of the week can't take credit for the whole week's result, and the brief is told to say what the timing actually supports.
- One-offs are isolated, never pitched as opportunities. A single bulk order gets explained and removed from the trend line, not turned into a new strategy.
- It never invents a person. Changes are only attributed to a name that appears in the platform's own account history.
- Gaps are disclosed. If a platform's history only covers part of the week, the brief says so instead of presenting a partial picture as complete.
- No abbreviations. It writes contribution margin, Marketing Efficiency Ratio and new customer acquisition cost in full, so a founder, a finance lead and an agency all read the same thing.
These rules are why the brief reads like a sharp analyst wrote it rather than a dashboard export. If you want the background on why MER is a better guardrail than ROAS, or which metrics belong on an ecommerce dashboard in the first place, those posts go deeper.
How to run it, step by step
Step 1: Connect your data to Claude or ChatGPT
The brief needs live data from your store and ad accounts. StoreHero's MCP connects Shopify, Meta, Google Ads, TikTok and Klaviyo to Claude, ChatGPT or any MCP-compatible AI tool in a few minutes; our Claude and Shopify setup guide walks through it.
Step 2: Copy the prompt
Get the full Ultimate weekly business brief prompt from the AI Skills Library. It's free, and you can download it as a file.
Step 3: Paste it into a new chat and name your store
Start a fresh chat, paste the prompt, and say which store and period you want, for example "run this for my store for last week" or "run it for the month of September". It pulls everything in parallel, then writes the brief.
Step 4: Read the first screen, then decide how deep to go
The summary, scorecard and actions are the brief. Everything below is there for when you want to check a claim or brief your agency. The brief arrives as an HTML file you can forward as it is, or save as a PDF.
If your AI tool supports scheduled tasks, it will offer to run the brief every Monday morning so it's waiting for you before the week starts. Running several stores? The same idea works across a portfolio with our multi-store report.
Using it in your weekly business review
A weekly business review works best when the meeting starts from the same written brief rather than everyone's own dashboard. Send the brief an hour before, open the meeting on the actions list, and spend the time on the two or three decisions that matter. Owners are already named, so nobody leaves wondering who's on it.
For agencies, it's a ready-made client update: the change log shows the work you did, the standouts show what's working, and the honesty rules mean a bad week reads as a bad week, with a plan. If you're still comparing tools for the rest of your stack, our ranking of the best ecommerce analytics tools for profit is a good next read.
Most ecommerce reporting leaves the thinking to you. This brief does that first pass every week and shows its working, so you can spend Monday on decisions instead of rebuilding the story. StoreHero is what feeds it the joined-up sales, margin and ad data, and if you'd like help setting it up for your brand or your clients, book a call with the team.
What happened, why, what is coming and what to do about it, across sales, margin, ads, email and account changes. Works in Claude, ChatGPT or any MCP-compatible AI tool.
Key takeaways
- Good ecommerce reporting answers business questions (why did sales move, what do we do now), not just which numbers went up or down.
- The Ultimate weekly business brief is a free prompt that writes that report for you from your live store, ad, email and account-change data.
- It leads with a 60-second summary, a scorecard with your break-even point, and 3 to 5 actions that each have an owner.
- It shows who changed what in Meta, Google Ads and Klaviyo, and only says a change coincided with a result unless the timing proves more.
- One-off orders are separated out so they don't distort the trend, and it checks what happened at this point last year so you can prep.
- It runs weekly or monthly in Claude, ChatGPT or any MCP-compatible AI tool once StoreHero is connected.
Frequently asked questions
What is ecommerce reporting?
Ecommerce reporting is turning raw store data (sales, orders, margin, ad spend, email and traffic) into summaries a team can act on. Good reporting is built around business questions, such as why sales moved last week and what to do about it, rather than listing every available metric.
What is the best reporting tool for e-commerce?
It depends on the question you need answered. Shopify's built-in reports cover sales and orders, and each ad platform reports on its own spend. For a profit view that joins sales, margin, ads and email in one place, StoreHero connects them, and the free Ultimate weekly business brief prompt turns that data into a written weekly report in Claude or ChatGPT.
How do I automate ecommerce reporting with Claude?
Connect your store and ad accounts to Claude through an MCP connector such as StoreHero's, paste a reporting prompt like the Ultimate weekly business brief into a new chat and name your store and period. If your Claude plan supports scheduled tasks, you can set it to run every Monday morning.
What do ecommerce reporting tools track?
Most track net sales, orders, average order value, conversion rate, ad spend and return on ad spend. Profit-focused tools also track gross margin, contribution margin, Marketing Efficiency Ratio, new versus repeat customers and new customer acquisition cost, which tell you whether growth is actually making money.
Does the weekly business brief work in ChatGPT as well as Claude?
Yes. The prompt works in Claude, ChatGPT or any AI tool that supports MCP connectors. Where a tool can't create files, it gives you the brief as HTML you can save and open in a browser.
