Days of stock at the last 28 days' sales rate, as of 30 Apr 2026
Suggested quantity covers supplier lead time plus 3 weeks, rounded up to the case pack
This ecommerce KPI dashboard is a working concept on sample data for Hollandse Tuin & Co, a fictional garden-supplies company that sells through its own webshop, two marketplaces and a wholesale channel. It puts sales, the webshop funnel and stock on one screen, because in a seasonal business a strong month is only good news if the stock can keep up.
The top row has six figures: net sales, orders, average order value, webshop conversion rate, stock value at cost and days of cover. The sales figures compare with the same period last year and carry a small trend line; the stock figures compare with the stock position at the start of the period. You can switch between week, month and season, and if you filter to one channel the sales figures and the sales chart follow.
Net sales by channel is a stacked column chart over the last 12 weeks, 12 months or 8 seasons, with a table view that adds the change against last year. Next to it, the webshop funnel runs from sessions to product views, add to cart, checkout and orders, with the conversion rate of each step, so you can see where visitors drop off.
The lower half is about stock. Stock cover by category shows how many days each of six categories would last at the last 28 days' sales rate, measured against a reorder point (supplier lead time plus safety stock), and flags the categories below it. The low-stock table goes down to the product: stock on hand, days of cover, lead time, a suggested reorder quantity and a status of Reorder now, Order this week or OK. In the sample data, spring demand has pushed several categories under their reorder point, soil and fertiliser among them, which a garden retailer wants to know in April, not May.
It is a web app written in HTML, CSS and JavaScript, with the charts drawn as SVG. The page holds one sample dataset (daily sales and orders per channel from March 2024 to 30 April 2026, plus stock by category and by product), and every number on screen is calculated from it, so the KPI row, the charts and the tables always agree. The suggested reorder quantity is arithmetic you can check yourself: expected sales over the lead time plus three weeks, minus stock on hand, rounded up to the supplier's case pack.
The AI note at the top uses those same calculated numbers. It counts the products that will sell out before a new delivery can arrive, estimates the sales at risk and names the product with the biggest gap. In this demo the note is a template filled from the computed figures; for a client, a language model can write it from the same figures, so it never has to work out a number on its own.
For a real shop I replace the sample data with your own sources, such as your webshop platform, marketplace seller reports and the stock levels in your inventory or ERP system. Those refresh on a schedule, and sign-in keeps the dashboard to your own team. If your reporting already lives in Power BI, the same design can be built there.
If you are choosing KPIs for your own shop, I would start where this dashboard does: net sales, orders, average order value and conversion for demand, then days of cover against lead time for supply. Most shops already have both numbers, just in different systems.
I can build an ecommerce dashboard like this on your own webshop, marketplace and stock data. Live dashboards · All examples