Understanding Personalisation Rules
The logic layer that defines what Liva 7 is allowed to recommend
Overview
Personalisation rules are one of the core control layers inside Liva 7. They sit above algorithms, above widgets and above page-level strategies. Their purpose is simple: rules define what the engine is allowed to show, regardless of shopper behaviour or algorithmic ranking.
This article explains the logic behind rules, how they influence the engine and how they shape recommendations across your storefront.
What personalisation rules represent
Rules act as global constraints that shape the engine’s decision space. They do not generate recommendations themselves. Instead, they define the boundaries within which the engine must operate.
Rules allow you to:
- force visibility for specific products
- block visibility for products or collections
- control fallback behaviour when relevance is low
- ensure consistency across widgets, Smart Content and Smart Bundles
Rules do not replace the algorithm. They restrict and guide it.
Why rules exist in Liva 7
Liva 7 is designed to balance two forces:
- Adaptive intelligence
The engine reacts to signals, behaviour, context and product relationships. - Merchant control
You decide what is allowed, what is not allowed and what must always be shown.
Rules are the mechanism that ensures your merchandising intent is always respected, even when the engine is adapting in real time.
Rules protect your brand and merchandising priorities.
The four rule types and what they mean
Liva 7 includes four global rule types. Each one influences the engine in a specific way.
Pin Products
Forces selected products into recommendation output whenever space allows.
- Guarantees visibility
- Overrides ranking
- Useful for hero items, seasonal pushes or brand priorities
Excluded Products
Prevents specific products from ever appearing.
- Removes items from the engine’s consideration
- Useful for discontinued items, low-margin items or brand-sensitive products
Excluded Collections
Blocks entire collections from being recommended.
- Removes all products within the collection
- Useful for clearance, restricted categories or controlled merchandising
Fallback Behaviour
Controls whether the engine is allowed to fill empty slots when relevance is low.
- When enabled, the engine fills gaps with safe defaults
- When disabled, widgets may show fewer items
- Useful when you want strict control over output
Rules apply globally across your storefront.
How rules influence recommendation output
Rules override algorithmic ranking when necessary. They do not change how the engine thinks, but they change what the engine is allowed to choose from.
Examples of rule impact:
- A pinned product appears even if it is not the top ranked item.
- An excluded product never appears, even if it is highly relevant.
- Excluding a collection removes all its products from consideration.
- Disabling fallback prevents the engine from filling empty slots.
The algorithm ranks products. Rules decide which products are eligible.
How rules fit into the wider Liva 7 system
Rules are one of several layers that shape personalisation. Understanding how they interact with other layers helps merchants predict behaviour.
Interaction with algorithms
- Algorithms determine relevance
- Rules determine eligibility
- The engine blends both to produce final output
Interaction with widgets
Widgets control:
- layout
- display type
- strategy
- page placement
Rules control:
- which products the widget is allowed to show
Interaction with Smart Content
Smart Content respects all rules when generating product lists.
- Excluded items never appear
- Pinned items are prioritised
- Fallback behaviour applies when content includes product lists
Interaction with Smart Bundles
Bundles avoid excluded products and collections automatically.
- Pinned items may appear when relevant
- Fallback behaviour influences bundle completeness
Rules cascade across all systems that generate product recommendations.
When merchants typically use rules
Rules are most valuable when you want predictable behaviour layered on top of adaptive personalisation. Common use cases include:
- highlighting seasonal or promotional products
- hiding discontinued or low-priority items
- preventing certain categories from appearing
- enforcing brand guidelines
- controlling how strict or flexible the engine should be
Rules give you a stable foundation while still allowing the engine to adapt to shopper behaviour.
How rules support long-term personalisation strategy
Rules help you maintain consistency across your storefront by ensuring:
- your brand priorities are always respected
- the engine never recommends products you do not want promoted
- key products receive consistent visibility
- fallback behaviour aligns with your merchandising approach
They allow you to combine the flexibility of AI with the control of manual merchandising.
