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.
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.
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
The highest-intent moment on the site. Additions are bounded by rules on count, price band and category so the basket stays coherent.
- Cross-sell bounded by rules
- Price-band constraints
- Category exclusions enforced
- Measured on attributed uplift
Product affinities passed into CRM so email and messaging carry the same personalisation the site does, rather than a generic bestseller list.
- Per-customer product affinities
- Feeds CRM segments and campaigns
- Consistent with on-site surfacing
- Attribution back to the analytics engine
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.
What this connects to
Every module runs standalone and every module talks to the kernel. These are the ones most often deployed alongside it.
Analytics Engine
Retail KPIs reported precisely, so decisions come from evidence.
Open page → Platform · CRMCustomer Relationships
Every interaction, order and ticket against one customer record.
Open page → Platform · PIMCatalogue Management
One source of truth for product data, published to every channel.
Open page →See recommendations running on your catalogue.
A working walkthrough with your catalogue, your order flow and your questions. No slideware.