Webshop KPIs Worth Checking Every Week
By CodexierPublished 5 min read
Most webshop dashboards show everything and help with nothing. Sessions, page views, bounce rate and follower counts move every day without telling you what to do. A useful weekly review is short: a handful of numbers that each point to a decision. This guide sets out those numbers, how to calculate them honestly, and which popular metrics you can stop watching.
The five numbers to check weekly
| KPI | How to calculate | Decision it supports |
|---|---|---|
| Orders and revenue | Count and sum, excluding cancelled orders, vs same week last year | Are we growing, or just in season? |
| Conversion rate | Orders divided by sessions from real channels | Is the shop or the traffic the problem? |
| Average order value | Revenue divided by orders | Pricing, bundles, free-shipping threshold |
| Contribution margin | Revenue minus product cost, shipping, fees and returns | Which products and channels earn money |
| Returning customers | Orders from customers with an earlier order | Retention and email efforts |
Compare with the same week last year where you have the data. Swedish e-commerce is strongly seasonal around salary day, Black Week, Christmas and summer, and a week-on-week comparison often only shows the calendar.
Traffic quality over volume
More visitors is not the goal; more buyers is. A campaign that doubles sessions and halves the conversion rate has not helped. Look at conversion rate and revenue per channel, not in total.
- Split by channel: organic search, paid search, social, email, direct
- Filter out your own visits, bots and test orders
- Watch revenue per session per channel, which combines conversion and order value
- Check the mobile conversion rate separately; a gap to desktop usually points to checkout problems
If numbers look strange after a consent-banner change, remember that analytics only sees visitors who accept tracking. Compare with order data from the shop platform, which is complete.
Margin after fees and returns
Revenue is the number people report; contribution margin is the number that pays salaries. For each order, subtract the product cost, the shipping you pay, the payment fee from Klarna, Swish or the card provider, the packaging, and an estimate of returns for that product category.
| Cost item | Where to find it |
|---|---|
| Product cost | Your inventory or accounting system |
| Shipping | Carrier invoices or shipping integration |
| Payment fees | Payment provider reports |
| Returns | Return rate per category times the cost of a return |
| Discounts and free shipping | Order data from the platform |
Doing this once per product category is enough to start. It often shows that a best-seller is barely profitable once returns are included. Our guide on returns handling and margin goes deeper.
Repeat customers
Returning customers cost almost nothing to acquire again, so their share of orders says a lot about the health of the business. Track the share of weekly orders from customers who have ordered before, and, monthly, how many first-time buyers from a given month come back within ninety days.
If the share is low, the usual levers are post-purchase emails, a reason to return such as refills or new collections, and a delivery experience worth repeating. These are cheaper than buying the same customer twice through ads.
Metrics to stop watching
- Total sessions and page views without a channel split
- Bounce rate on its own, which says little about intent
- Social followers and likes
- Time on site, which rises when visitors are confused
- Daily revenue swings, which are mostly noise in a small shop
These are not useless, but they belong in a monthly deep dive, not in the fifteen minutes on Monday morning.
Setting up the weekly view
The simplest version is a spreadsheet updated from the shop platform's reports every Monday. The next step is a dashboard that pulls orders, costs and analytics automatically. Either way, the numbers must be trustworthy: consistent definitions, test orders excluded, and conversion tracking that matches real orders.
If you need that measurement in place, our tracking and analytics setup covers GA4, consent and e-commerce events. When you do not need us: if your platform's own reports already give you these five numbers and you trust them, a spreadsheet is enough. See pricing or book a call.
Frequently asked questions
Which conversion rate is good for a webshop?
It varies too much by product, price and traffic mix to have a universal target. Track your own trend by channel and device instead; our guide on what a good conversion rate is explains why.
Why do analytics and the shop platform show different revenue?
Analytics misses visitors who decline tracking, and ad blockers and refunds also create differences. Use the shop platform for revenue and orders, and analytics for channels and behaviour.
How much time should the weekly review take?
Around fifteen minutes once the numbers are set up. If it takes longer, the data is not automated enough or the dashboard shows too much.
Should I track customer lifetime value weekly?
No, lifetime value changes slowly. Review it monthly or quarterly and use the weekly share of returning customers as the early signal.
Not sure your numbers can be trusted?
On a short call we look at how your webshop is measured today and what it would take to get a weekly view you can act on.
Book a free 15-minute call