Arrange a demo → Talk to us
Home/ Platform/ Recommendations
Platform · REC · Core module

Show the next product before they go looking for it.

Recommendation works when it is built on your data rather than a generic model — your catalogue structure, your customers, your purchase history. Surfaces run across the storefront, the cart and outbound campaigns.

On-site personalisation Cart cross-sell Campaign audiences Merchandiser controls
What it does

Personalisation you can still control.

Automatic surfacing with merchandiser overrides, because there are always products you must not recommend and products you very much want to.

On-site surfacing

Related, similar and complementary products surfaced on product and category pages from live catalogue and behaviour data.

Cart and checkout cross-sell

Relevant additions offered at the point of highest intent, bounded by rules so the basket does not fill with noise.

Campaign audiences

Segments and product affinities passed to CRM so outbound campaigns carry the same personalisation as the site.

Merchandiser overrides

Pin, boost, exclude. Automatic surfacing that a human can still correct when commercial priorities say otherwise.

Cold-start handling

New products and first-time visitors handled by category and popularity rules rather than by showing nothing.

Performance tracking

Recommendation surfaces measured on attributed revenue in the analytics engine, so their value is a number rather than a belief.

In detail

Where recommendations appear.

Similar and complementary products surfaced where the customer is already evaluating, using catalogue structure and behavioural signals together.

  • Similar products by attribute and behaviour
  • Complementary products by purchase affinity
  • Category-level popularity fallbacks
  • Merchandiser pinning and exclusion
Questions

The things buyers actually ask

Your catalogue structure and your customers' behaviour on your own properties. Nothing is pooled across other eqomOS tenants — tenant isolation is enforced at the database level.

Yes, and they should be able to. Pin, boost and exclude controls exist precisely because commercial priorities sometimes outrank statistical relevance.

Recommendation surfaces are tracked as attributed revenue in the analytics engine, so their contribution is measured on the same basis as any other merchandising decision.

See recommendations running on your catalogue.

A working walkthrough with your catalogue, your order flow and your questions. No slideware.

Arrange a demo