AI personalisation One shop. A different one for every customer.
In B2B you know your customers: industry, assortment, order history. We use that for recommendations, search order and the home page, following rules your team understands and can override at any time.
Electrician
- Last ordered: NYM-J 3×1.5 mm² cable
- Goes with it: junction boxes
- New in the range: smoke alarms
Plumbing company
- Last ordered: 15 mm press fittings
- Goes with it: pipe clamps
- At your contract price: radiator valves
What you already know
You know more about your customers than any consumer shop.
-
Account and industry
Customer group, industry, price tier and permitted assortment are already in Magento or the ERP.
-
Order history
What gets reordered regularly, in which quantities and at what interval.
-
Behaviour in the session
Searches, viewed products, cart. Only with consent.
-
Relationships in the catalog
Accessories, spare parts, successor items and what is often bought together.
Where it works
Not more recommendations. Better ones.
-
Reordering
The home page shows what this customer regularly needs, instead of the deal of the week.
-
Accessories and spare parts
On the product page, what technically fits, not just what others clicked.
-
Search order
The customer’s own assortment and familiar brands rank higher.
-
Content
Banners, category texts and notes matching the industry of the logged-in customer.
Control
Your team keeps the final say.
-
Rules first
Pin must-have items, exclude products, favour brands. What sales decides applies.
-
Data second
Similar products, frequently bought together, next likely order. Each with a traceable reason.
-
Always test against the default
Every change runs as an A/B test against the previous view. It only stays if it performs better.
Honestly
Twenty regular customers do not need an algorithm.
Worth it if
- many customers from different industries order in the same shop,
- your catalog is large and customers need only a fraction of it,
- a large share of revenue comes from reorders.
Probably not if
- you have few customers who are looked after personally anyway,
- the catalog is small and the same for everyone.
Recommendations like socks from the laundry basket? We find the matching pair.
We look at your customer groups and order data and show where personalisation pays off first.
