Smart Recommender elevates your website’s product discovery by delivering personalized recommendations that adapt to each user’s behavior, interests, and purchase history. Instead of showing the same generic products to everyone, personalization ensures that each customer sees the most relevant suggestions at the right time—driving higher engagement, stronger conversions, and a more seamless shopping experience.
What is personalization?
Personalization means tailoring product recommendations to each individual based on their data, activity, and preferences. With effective personalization:
Customers are introduced to products they’re likely to love.
Irrelevant suggestions—such as items they already purchased or recently viewed—are filtered out.
Product discovery becomes effortless, leading to increased customer satisfaction and higher retention rates.
By applying advanced personalization criteria and filtering rules, Smart Recommender transforms recommendations into meaningful experiences that directly impact revenue growth.
Segments
Segments let you tailor recommendations for specific audience groups, making it possible to scale personalization without losing relevance. By applying segments in Smart Recommender campaigns, you can deliver context-aware experiences that resonate with each user type.
Examples of segmentation in action:
New vs. Returning Users: Use Visiting Behavior segments to design different recommendation strategies—for instance, introduce new users to bestsellers while offering returning users more personalized or complementary products.
Low AOV Users: Target Purchase Behavior segments with strategies aimed at increasing average order value, such as cross-sell or bundle recommendations.
Segmented campaigns ensure the right products and messages reach the right audience at the right time—boosting engagement, relevance, and conversions.
For deeper guidance on creating and applying segments in your personalization efforts, visit Web Segments to explore more strategies.
Rules
Rules determine when and where product recommendations appear. They ensure your recommendations are context-aware and aligned with user actions, page types, and your merchandising goals.
Example Use Cases
Upsell Opportunities: Apply the Product Added to Cart rule (from Page Rules) to recommend higher-value or complementary items when a user adds a product to their cart.
Intent Matching: Recommend items from the user’s last visited category to reflect their most recent browsing behavior and increase conversion likelihood.
By defining clear rules, you can:
Deliver timely, relevant recommendations at key points in the customer journey.
Align recommendation strategies with broader campaign objectives (e.g., revenue growth, cross-selling, retention).
Ensure consistent execution of merchandising logic across your site.
Explore Campaign Rules to learn more about available options and advanced setups.
Personalized Algorithms
Smart Recommender provides a rich library of algorithms to help you deliver the right product to the right user at the right time. Each algorithm is optimized for different goals—whether that’s boosting engagement, driving upsells, or increasing order value.
Key algorithms and their use cases
User-Based: Recommend products liked by users with similar tastes. It is best for increasing discovery through collaborative filtering.
Real-Time User Engagement: Recommend products based on live, in-session behavior. It is best to extend the session duration and keep users browsing.
Recently Viewed: Surface items a user previously explored. It is best for re-engaging returning users and reminding them of prior interests.
Purchased with Last Purchased: Suggest complementary items to the last purchased product. It is best for cross-sell and upsell strategies that boost AOV.
Location-Based Top Sellers: Show top-selling products in the user’s region. It is best for driving relevance with local popularity trends.
By matching each algorithm to your campaign objectives, you can:
Optimize product discovery.
Personalize recommendations at scale.
Deliver measurable lifts in engagement, conversion, and revenue.
Explore Recommendation Algorithms for deeper technical details and examples.
Exclusions
Exclusions ensure your recommendations remain relevant and clutter-free by automatically filtering out items that users have already interacted with. This prevents wasted impressions and enhances the overall customer experience.
Exclude Cart Items: Hide products that are already in the user’s cart to avoid redundancy.
Exclude Recently Viewed: Remove items the user has recently browsed, using a configurable time frame or count.
Exclude Recently Purchased: Prevent showing items the user has already bought, using a configurable time frame or count.
Exclusions matter because they:
Keep recommendations fresh and valuable.
Prevent user fatigue by avoiding the same products being displayed repeatedly.
Encourage discovery of new or complementary items.
You can configure exclusions directly in your Recommendation Strategy Settings.
Attribute Affinity
Attribute Affinity enhances personalization by recommending products that share similar characteristics with items a user has previously viewed, clicked, or purchased. This ensures recommendations feel more contextually relevant and aligned with individual preferences.
You can select up to 5 attributes from your product catalog (e.g., brand, color, material, size, category).
Smart Recommender assigns higher weights to products with matching attributes.
The system then surfaces products with the closest match to the user’s demonstrated interests.
Example: If a user frequently browses red dresses from a particular brand, Attribute Affinity will prioritize showing them similar items (e.g., other dresses from the same brand or red dresses from different brands).
The benefits of Attribute Affinity are:
Relevance: Users see items that align more closely with their preferences.
Discovery: Encourages exploration of related products within the same attribute set.
Conversion: Increases the likelihood of purchase by surfacing products with proven appeal.
Enable Attribute Affinity in your Recommendation Strategy Settings to deepen personalization for your campaigns.
Dynamic Filters
Dynamic Filters ensure that recommendations remain relevant by adapting to the user’s real-time browsing context. Instead of static rules, they automatically adjust to what the user is currently viewing, providing a seamless and personalized shopping experience.
You can configure dynamic filters to personalize recommendations based on attributes from the item being browsed (e.g., brand, category).
Operators such as “matches the item they’re currently viewing” allow the system to surface products that align directly with the active context. This enables hyper-personalization without the need for manual rule creation.
Example: If a user is viewing sneakers from Brand A, Dynamic Filters can recommend:
Other sneakers from Brand A
Related items that match the same attributes (e.g., sports shoes within the same category)
The benefits of Dynamic Filters are:
Contextual Relevance: Recommendations adapt to the exact product a user is browsing.
Scalability: Reduce reliance on manually created rules.
Hyper-personalization: Deliver real-time suggestions aligned with user intent.
Enable Dynamic Filters in your Recommendation Strategy Settings to maximize contextual personalization for your campaigns.
Best Practices
Following best practices ensures that your recommendation strategies deliver personalization while still encouraging product exploration.
Balance Personalization with Discovery: Avoid overfitting by ensuring recommendations don’t become too narrow. Encourage users to explore new products that align with their interests beyond their immediate focus.
Leverage Attribute Affinity: Utilize Attribute Affinity to boost contextual relevance and align recommendations with users' demonstrated preferences.
Experiment with Exclusion Windows: Apply exclusion windows to prevent repetitive recommendations and optimize the freshness of results.
Track Performance Metrics: Regularly monitor conversion rates, click-throughs, and engagement metrics to evaluate and refine recommendation strategies.
Effective recommendation strategies require a balance between personalization, discovery, and ongoing optimization.