Introduction

AIPersonalize360 delivers intelligent product recommendations that boost engagement and sales. The plugin uses two distinct approaches: WooCommerce data-based recommendations (no AI required) and AI-powered personalized recommendations through Recombee integration.

This flexibility allows you to start immediately with smart recommendations and optionally enhance them with machine learning later.

How Recommendations Work

Without AI (Default)

When no AI engine is selected, the plugin uses intelligent fallback strategies based on your WooCommerce store data:

1. Best Rated Products

  • Products with highest average customer ratings
  • Minimum reviews required for inclusion
  • Sorted by rating, then by number of reviews

2. Best Selling Products

  • Products with highest total sales
  • Based on WooCommerce order data
  • Updated with each order

3. On-Sale Products

  • Products currently discounted
  • Shows regular and sale prices
  • Prominent “SALE!” badges

4. Recent Products

  • Recently added to your store
  • Sorted by publication date
  • Great for highlighting new arrivals

The plugin intelligently combines these strategies to show the most relevant products to all users.

Smart Fallbacks

If no rated products exist, it falls back to best sellers. If no sales exist, it shows on-sale products, and so on. This ensures recommendations always display.

With AI (Recombee)

When Recombee is configured, recommendations become personalized for each user:

For Registered Users

  • Behavioral Learning: Tracks products viewed, added to cart, and purchased
  • Pattern Recognition: Identifies preferences and shopping patterns
  • Collaborative Filtering: “Users like you also bought…”
  • Content-Based: Products similar to previous views

For Guest Users (Premium)

  • Session-based tracking
  • MAC address identification
  • Visitor behavior analytics
  • Cross-session continuity

Cold Start Handling

For new users with no history:

  • Falls back to popular products
  • Shows trending items
  • Gradually builds personalization as user interacts

Types of Recommendations

1. General Product Recommendations

Broad recommendations for any context:

  • Use Case: Home page, landing pages, blog posts
  • Logic: Best products for this user or overall popular items
  • Shortcode: [aipers_product_recommendations]

Context-specific recommendations:

  • Use Case: Single product pages
  • Logic: Products similar to or frequently bought with current item
  • Shortcode: [aipers_related_product_recommendations]

How Recommendations Display

Recommendations appear as a product grid with:

Visual Elements

  • Block Title: Customizable heading (e.g., “Recommended Products”)
  • Description: Optional subtitle text
  • Product Images: Featured product photos
  • Product Names: Clickable titles linking to product pages
  • Prices: Regular price or sale price display
  • Sale Badges: “SALE!” indicator for discounted products
  • Add to Cart Buttons: Quick purchase actions

Layout

  • Responsive grid design
  • Adapts to screen size (mobile, tablet, desktop)
  • Number of columns based on available space
  • Consistent with WooCommerce product grids

Personalization Levels

Level 1: No Personalization (Default)

  • Same recommendations for all users
  • Based on overall store performance
  • No tracking required
  • Works immediately

Level 2: AI Personalization (Recombee Free)

  • Personalized for registered users
  • Tracks views, cart additions, purchases
  • Machine learning improves over time
  • Free tier available

Level 3: Advanced Personalization (Premium)

  • Guest user tracking
  • Cross-session continuity
  • Advanced analytics
  • Requires premium Recombee plan

Improving Recommendation Quality

For WooCommerce-Based Recommendations

  1. Enable Product Reviews:
    • Go to WooCommerce → Settings → Products
    • Enable ratings and reviews
    • Encourage customers to leave reviews
  2. Track Sales Accurately:
    • Ensure orders are marked as “Completed”
    • Sales data updates recommendation logic
  3. Use Sale Prices:
    • Set sale prices on products
    • Discounted items attract attention
  4. Keep Product Catalog Updated:
    • Add new products regularly
    • Remove discontinued items
    • Update product information

For AI-Based Recommendations

  1. Collect Sufficient Data:
    • Allow time for user interactions
    • AI improves with more data
    • Minimum 100+ interactions recommended
  2. Ensure Event Tracking:
    • Verify product views are tracked
    • Cart additions logged
    • Purchases recorded
  3. Regular Data Sync:
    • WordPress cron runs daily sync
    • Products and users stay updated
    • Check sync status in Recombee dashboard
  4. Monitor Recombee Analytics:
    • Review recommendation performance
    • Check click-through rates
    • Analyze conversion metrics

Best Practices

Placement Strategy

  • Home Page: General recommendations to engage visitors
  • Product Pages: Related products to increase order value
  • Cart Page: Last-minute suggestions before checkout
  • Thank You Page: Encourage repeat purchases
  • Blog Posts: Relevant products mentioned in content

Display Optimization

  • Number of Products: Show 3-6 items for best results
  • Descriptive Titles: Use action-oriented titles
  • Mobile Optimization: Test on all devices
  • Loading Speed: Monitor page performance

Content Strategy

  • Varied Recommendations: Mix popular, new, and sale items
  • Seasonal Updates: Adjust for holidays and seasons
  • Category Balance: Show diverse product categories
  • Price Range: Include various price points

Measuring Success

Key Metrics to Track

  • Click-Through Rate: % of users clicking recommendations
  • Conversion Rate: % of clicks resulting in purchases
  • Average Order Value: Impact on order size
  • Revenue Per Visit: Overall revenue increase

Using Recombee Analytics

  1. Log in to Recombee dashboard
  2. Navigate to KPI section
  3. View recommendation performance:
    • Recommendations shown
    • Click-through rates
    • Conversion rates
    • Interactions over time

A/B Testing

  • Test with vs. without AI
  • Try different placements
  • Vary number of products shown
  • Test different titles and descriptions

Common Scenarios

Scenario: New Store with Few Products

  • Use non-AI mode
  • Focus on showing all products
  • Highlight new arrivals
  • Add AI when catalog grows

Scenario: Established Store with Sales Data

  • Enable AI for personalization
  • Leverage existing sales history
  • Use best sellers as fallback
  • Monitor conversion improvements

Scenario: Seasonal Products

  • Update recommendations seasonally
  • Create dedicated landing pages
  • Use manual product selection if needed
  • Track seasonal performance