- A good weekly report fits on one page and answers three questions: how did we do, why, and what will we change.
- Use store data for revenue and orders, and treat ad platform numbers as supporting detail.
- Compare each week with the same week last year and with plan, not only with last week, and add a short written commentary.
Why do so many weekly reports get ignored?
Most weekly reports are data dumps: dozens of metrics from every platform, colored red or green, with no explanation. People skim them, nobody acts on them and eventually nobody opens them.
A useful report is short and opinionated. It shows the handful of numbers that drive the business, explains what moved them and lists the decisions for the coming week. It is built on a clear KPI framework, so everyone knows which metrics matter and why.
What should the weekly report include?
Group metrics into four sections. Keep each section to a few lines.
1. Business results
| Metric | Source | Compare with |
|---|---|---|
| Net revenue | Store platform | Same week last year, plan |
| Orders | Store platform | Same week last year, plan |
| Average order value | Store platform | Same week last year |
| Gross margin or contribution | Store and finance data | Plan |
| Returns and refunds | Store platform | Previous 4 weeks |
2. Marketing efficiency
| Metric | Source | Compare with |
|---|---|---|
| Total marketing spend | Ad platforms, invoices | Plan |
| Marketing efficiency ratio (MER) | Store revenue ÷ total spend | Target |
| New customer orders | Store platform | Same week last year |
| Cost per new customer | Spend ÷ new customers | Target |
| Channel highlights | Ad platforms, GA4 | Previous 4 weeks |
3. Customers and retention
| Metric | Source | Compare with |
|---|---|---|
| Returning customer orders | Store platform | Same week last year |
| Email and SMS revenue per recipient | Email platform | Previous 4 weeks |
| Subscriber growth | Email platform | Plan |
4. Site and operations
| Metric | Source | Compare with |
|---|---|---|
| Sessions and conversion rate | GA4 or store analytics | Same week last year |
| Stock-outs on top products | Inventory system | Previous week |
| Shipping delays or support spikes | Helpdesk, fulfillment | Previous 4 weeks |
Why use store data as the source of truth?
Ad platforms each claim credit for the same orders, so adding their reported revenue together often exceeds actual store revenue. GA4 undercounts because of consent and ad blockers. Your store platform records every real order.
Use store data for revenue, orders and new versus returning customers. Use platform data to explain movements and to optimize campaigns, not to report business results. See eCommerce KPIs for metric definitions and the GA4 versus Shopify revenue gap for why numbers differ.
How should you compare weeks?
Week-over-week comparisons are noisy. A promotion, a holiday or an email send can swing results by 30% or more. Better comparisons:
- Same week last year (aligned by weekday) to account for seasonality.
- Plan or forecast to show whether you are on track.
- Rolling 4-week average to smooth out one-off spikes.
Mark known events in the report (sales, launches, stock issues) so readers can interpret the numbers.
What should the commentary say?
The written part is the most valuable piece. Three short sections:
- What happened: “Revenue was 8% below plan, mainly because the spring collection launched three days late.”
- Why: “New customer orders fell 15% while returning orders held steady. Meta spend was reduced while creative was being refreshed.”
- What we will do: “Relaunch Meta prospecting with new creative on Monday. Move the launch email to Tuesday. Check stock for the top 10 products before the weekend.”
Keep it to five to eight sentences. If nothing important changed, say so.
Worked example: a one-page weekly report
(Illustrative numbers.)
Week 45 summary: Net revenue $142,000 (plan $150,000, last year $128,000). Orders 2,030, AOV $70. MER 4.1 (target 4.0). New customer orders 820 (last year 760). Cost per new customer $34 (target $36).
What happened: Revenue grew 11% year over year but missed plan by 5%. Email revenue was below average after a delayed campaign. New customer acquisition was on target.
Decisions: Send the delayed campaign Tuesday. Increase Google Shopping budget 10% for bestsellers with good stock. Investigate a rise in return requests for one product.
That fits on a single screen, and everyone knows what will happen next.
How do you automate the report?
Manual reports take hours and break easily. Options:
- Store platform reports for orders, revenue and customers. See Shopify analytics reports.
- A dashboard tool like Looker Studio that pulls store, GA4 and ad platform data into one view. See eCommerce dashboard in Looker Studio.
- A spreadsheet fed by connectors, for smaller teams.
Automate the numbers, but keep the commentary human. The value is in the interpretation.
Checklist: weekly eCommerce report
- [ ] Report fits on one page or one screen
- [ ] Revenue, orders and customers come from store data
- [ ] Each metric compares with same week last year and plan
- [ ] Rolling 4-week averages used for noisy metrics
- [ ] Known events (sales, launches, stock issues) noted
- [ ] MER and cost per new customer included
- [ ] Short commentary covers what happened, why and what changes
- [ ] Decisions have owners
- [ ] Numbers are automated, commentary is written by a person
- [ ] Report sent at the same time every week
FAQ
Which metrics belong in a weekly eCommerce report?
Start with net revenue, orders, average order value, margin, marketing spend, MER, new customer orders and cost per new customer. Add retention and operations metrics that your team can act on weekly.
Should I report ROAS from each ad platform?
Include it as supporting detail for channel managers, but do not add platform ROAS figures together or use them as the main business result. Use store revenue and total spend for overall efficiency.
How long should the commentary be?
Usually five to eight sentences: what happened, why it happened and what the team will do next week. If the report needs more explanation than that, the metrics may be too detailed.