Merchandising Rules

Merchandising rules are the logic that controls which products get recommended to which shoppers in a guided selling experience. They translate between what a shopper says they need and what the product catalog actually contains, using product attributes, business constraints, and merchant priorities to produce controlled recommendations.

Also known as: recommendation rules, product matching rules, merchandising logic.

Why they matter

Product feeds are full of technical attributes: engine displacement, blade width, thread count, active ingredient concentration. These attributes hold real value for matching shoppers to the right product, but unless a shopper is a power user in the category, the raw attributes mean nothing to them.

Merchandising rules bridge that gap. They let the merchandising team define how shopper-friendly questions (“how big is your yard?”) map to technical product attributes (engine size, cutting width), so the shopper gets a recommendation they trust, and the brand gets a match based on real product differences.

Without merchandising rules, a guided selling experience is just a form. The questions don’t connect to the catalog, the recommendations aren’t controlled, and the results drift as the catalog changes.

Two types of merchandising rules

Evergreen rules (attribute-based)

These rules tie shopper answers to product attributes from the feed. They are the backbone of any guided selling experience.

Example: Troy-Bilt lawn mower finder

A lawn mower product feed includes technical specs like engine size and cutting blade width. A shopper buying a lawn mower probably doesn’t know what engine displacement they need, but they do know how big their yard is and whether they care about cutting speed.

Merchandising rules translate between these:

• “How big is your yard?” maps to engine size (bigger yards need more power)
• “How much do you value cutting it quickly?” maps to blade width (wider blades cover more ground per pass)

Shopper answer	Product attribute
Yard size: Small / Medium / Large	Engine power bands
Priority: Fastest cut / Balanced / Maneuverable	Cutting width bands

The merchandising team creates the rules that connect shopper answers to the product attributes that actually determine which mower is the right fit.

The key property of evergreen rules: they scale with the catalog. When a new lawn mower is added to the feed, the system automatically knows when to recommend it, as long as the product has the right attributes. The merchandising team doesn’t need to manually decide where each new product fits.

Targeted rules (boost and bury)

Targeted rules give the merchandising team fine-grained, time-sensitive control over what gets recommended:

• Boost: increase the prominence of a product in results (e.g., a new launch, a high-margin item, or a seasonal priority)
• Bury: suppress a product from results (e.g., being discontinued, quality issues, or overstock you don’t want to push)

The critical distinction: a well-built system applies targeted rules on top of the evergreen rules, not instead of them. A boosted product should still only appear when it matches the shopper’s needs. Going back to the lawn mower example: you would never recommend a lawn mower just because a merchant boosted it if it doesn’t also meet the shopper’s stated needs.

This is the difference between “merchandising control” and “merchandising override.” Control means the merchant can influence results within the guardrails of shopper intent. Override means the merchant can break the experience.

What the brand gets

• Recommendations that stay consistent as the catalog changes
• Merchandising control with guardrails (not blind overrides)
• Clear, explainable logic that teams can iterate without engineering bottlenecks

What breaks without merchandising rules

• Wrong product recommended. Without rules connecting shopper answers to product attributes, recommendations fall back on defaults like popularity, manual picks, or incomplete rules, not fit.
• Recommendations go stale. If rules are tied to specific products instead of attributes, every catalog change requires manual updates. New products don’t get recommended until someone remembers to add them.
• Shopper trust erodes. If a quiz recommends a product that obviously doesn’t match what the shopper said they need, they lose confidence in the experience (and the brand).
• Out-of-stock and fallback failures. Without guardrails and fallback logic, the experience recommends products that can’t be purchased, or shows nothing at all.
• Returns can increase. A shopper who buys the wrong product because the recommendation wasn’t controlled is a return waiting to happen.

Cartful context

Merchandising control is one of the core reasons enterprise brands choose Cartful. The platform gives merchandising teams direct ownership of the rules that govern recommendations:

• rules, guardrails, and fallbacks that connect shopper answers to product attributes from the feed
• scoring and weighting to control how strongly different answers influence the recommendation
• boost and bury controls that respect shopper intent (targeted adjustments, not blind overrides)
• a no-code visual editor (Studio) so merchandising teams can update rules without filing an engineering ticket
• evergreen rule structures that scale with the catalog: new products are automatically matched when they have the right attributes

Common pitfalls

• Being too literal with product feed attributes, showing shoppers technical specs instead of translating them into meaningful questions
• Tying rules to specific products instead of attributes (rules break when the catalog changes)
• Boosting products without respecting shopper intent (erodes trust in the experience)
• Not defining fallback behavior for out-of-stock or edge cases
• Requiring engineering involvement for every rule change (slows the merchandising team down and creates a bottleneck)