Commerce intelligence

What is commerce intelligence?

Every system knows a piece of your store. Your ad tool says ROAS 3, your store shows revenue, the marketplace keeps its commission, and your accountant tells you a month later what was left. Three stories about the same month, and they never quite match.

Commerce intelligence brings those pieces together in one model of your store: webstore, marketplaces, shops, ads, inventory, payments, shipping and accounting. With agreed definitions, so everyone means the same thing by an order, a return or a new customer.

The difference with ordinary reporting in one line: analytics shows what happened. Intelligence tells you why, and what to do about it.

Why it matters now

AI makes it easy to ask a question about your store. But on loose data, AI gives a fast, convincing answer that is often wrong, and customer data ends up in a chat.

The more AI agents steer budgets, prices and purchasing, the more a wrong target number costs. What counts as profit is an agreement between finance, marketing and purchasing. Commerce intelligence records that agreement, with the history, the checks and the access. The AI uses them.

From data to steering, in five steps

1. Connect. All your systems go in: webstore, marketplaces, ads, inventory, payments, shipping and accounting. You decide who sees what.

2. One reality. The same product, order and customer look slightly different in every system. Commerce intelligence makes them one, and checks that the data is complete, correct, up to date and traceable.

3. One definition. Every number gets a definition: what it means, how it is calculated, which source it comes from and who approved it. That happens in the model, from profit and inventory to customers, channels and advertising and products.

4. Understand and steer. Dashboards and alerts show how things are going. The Steering Model tells you where your attention is needed, and why.

5. Improve. From a signal to a plan, an action in your own systems that you approve, and then measuring what it delivered.

The order matters. Without one reality you count twice. Without one definition your AI says something different every day. Without steering you only have charts.

The Steering Model: where your attention pays off most

A dashboard tells you how a number is doing. The Steering Model tells you what that means for a product, channel, campaign or customer, and where to look first. Each gets a score, built from fixed steering points: a number, a norm, how much it weighs and who owns it. Every score comes with its cause.

Three things always stay separate: how healthy something is, how much of it you measure, and how reliable the data is. A high score on a third of the numbers is not a healthy store.

A score is only worth something if it ends in money and an action. Think of a product that sells well but loses money through returns and ads, advertising a product that is almost sold out, or a marketplace that earns less than your own webstore after commission and returns.

How it differs from BI, attribution and profit tools

A BI report shows what you had built. An attribution tool divides revenue across your ads. A profit app calculates profit per order, mostly for your own store. All useful, and they can stay.

Commerce intelligence looks at the whole. Every cost an order really carries is in it: purchasing, shipping, payment fees, commission, returns and refunds. Marketplaces, platforms, wholesale and shops sit next to your webstore, and inventory and purchasing count in the same profit. And it does not stop at seeing: the Steering Model points to where you step in.

Side by side: Ecommy compared.

An example: the same coat, three channels

The same coat sells through your own webstore in the Netherlands, a marketplace in Germany and a fashion platform with a long return period.

If you only add up the data, the fashion platform seems to earn the most. Calculate with the model, and you spread the warehouse costs from the monthly invoice over the orders and count the late returns. Then that channel turns out to cost money.

The data was the same. Only the agreements differ. That is the work of commerce intelligence.

Who it is for

For brands and retailers that sell through several channels: their own webstore plus marketplaces, fashion platforms, wholesale or shops. The more channels and costs, the more it pays off. It fits companies without their own data team best, in the Netherlands and across Europe.

The question differs per type of store: a D2C brand wants to know which campaign makes a profit after returns, a marketplace seller which product loses money after commission, a retailer with shops whether the stock is in the right place. See per type of store.

How Ecommy does it

Ecommy builds the model of your store, based on our Commerce Intelligence Model, and sets it up for your business: connecting, recording the definitions, putting the numbers next to your accounting and explaining every difference, and going live only after your approval. The profit you see per order and channel is the profit to steer by; the final profit stays with your accounting. How we set it up.

We keep it accurate: when data is missing or off, we flag it instead of guessing. You use it in dashboards with alerts, or in your own AI, such as Claude, ChatGPT or Copilot, via MCP. The Steering Model with scores per product, channel and campaign, and actions we prepare and you approve, are coming soon. See the model.