Related product recommendations in Ziadah don't rely on manual categorization — AI analyzes browsing, search, and purchase behavior to suggest what the customer actually wants, even before they know it themselves.
What's the Difference Between "Related Products" and "Cross-Sell"?
The difference may seem small but is crucial:
Cross-Sell: Recommends complementary products based on purchase data (what's bought together)
Related Products: Recommends similar and comparable products based on browsing data (what's viewed together)
Practical example:
A customer browsing a Samsung 55" TV:
Cross-Sell recommends: TV wall mount + HDMI cable + soundbar (complementary items)
Related Products recommends: LG 55" TV + Sony 55" TV + Samsung 65" TV (similar alternatives)
Both are important, but they serve different customer needs.
How Does AI Analyze Browsing Patterns?
Browsing Data Sources
Ziadah's system tracks and analyzes:
Products viewed in the same session: If a customer views 5 products, the system understands their interest pattern
Products viewed and added to cart (vs. just viewed): Distinguishes between serious interest and casual browsing
Session duration on each product: Longer time = higher interest
Search terms: The words the customer types in the search bar
Multi-Factor Relevancy Score
For each product, the system calculates a "relevancy score" based on:
Shared categories and subcategories
Price similarity (products within a close price range)
Shared viewing history (customers who viewed this also viewed that)
Shared tags (material, size, color, brand, style)
Real-World Examples of Browsing-Based Recommendations
Fashion Store
Customer viewing: A light blue formal shirt at 180 SAR
AI recommends as related products:
Navy blue formal shirt by the same brand — 180 SAR
Patterned shirt in the same style — 195 SAR
Same shirt in other colors — 180 SAR
Why these choices? Customers shopping for formal shirts typically compare between brands and colors before making a decision. AI puts comparison options in front of the customer instead of letting them leave the store.
Electronics Store
Customer viewing: Sony over-ear headphones at 450 SAR
AI recommends as related products:
Bose headphones similar to Sony specs — 480 SAR
Sony on-ear headphones (lighter, lower price) — 320 SAR
JBL headphones (popular alternative) — 399 SAR
Beauty Store
Customer viewing: Anti-aging cream by La Mer — 1,200 SAR
AI recommends as related products:
Estée Lauder Anti-aging cream — 850 SAR (cheaper alternative)
La Mer day cream (same brand) — 950 SAR
Clinique Anti-aging serum — 600 SAR (budget alternative)
Setting Up Related Products in Ziadah
Automatic Mode (Recommended)
From the dashboard → "Related Products" → Activate
Select display pages: Product page (essential) + "You may also like" page
Set number of suggestions: 4 to 8 products
Activate automatic learning
Manual Mode
Link specific products to each other
Useful for new products without browsing history
The system automatically switches to auto mode once sufficient data accumulates
Hybrid Mode
Define a pool of allowed products for recommendations
AI selects the most relevant ones from this pool
Good for controlling brand and category display
Measuring Related Products Effectiveness
| Metric | Target (Month 1) | Target (Month 3) |
|---|---|---|
| Click rate on suggestions | 8% - 15% | 12% - 22% |
| Pages per session | +1.5 to +2.5 | +2 to +4 |
| Session duration | +30% to +50% | +40% to +70% |
| Bounce rate reduction | -10% to -20% | -15% to -30% |
Advanced Strategies for Related Products
"Visual Similarity" Strategy
AI compares product images to find visually similar products. This is especially useful in fashion and décor — the customer may not know the exact product name but knows the "look" they want. Recommendations based on visual similarity dramatically increase browsing time.
"Viewing Trends" Strategy
Show a "Trending Now" section within related product recommendations. Products experiencing a viewing increase in the last 24 to 48 hours get higher priority. This creates an urgency effect and makes the customer feel they're following the latest trends.
"Price Range" Strategy
If the customer is viewing products in the 200 to 300 SAR range, all recommendations should be within the same range. Showing a 50 SAR product or an 800 SAR product will be irrelevant and reduce click-through rate.
How Does the Related Products Feature Boost Customer Retention?
Related product recommendations reduce the "exit to compare" problem. Without these recommendations, a customer comparing products will open a new tab and search on Google or go to a competitor store. With smart related product recommendations, comparison options are available within your store, reducing the probability of exit by 30% to 50%.
Additionally, related recommendations increase the number of pages per session, which means the customer discovers more products in your store and develops a better understanding of your catalog. This raises the probability of returning to the store by 25% to 40%.
Tips for Improving Performance
Exclude out-of-stock products — nothing frustrates a customer more than clicking a product and finding it unavailable
Regularly update recommendations — especially after adding new products
Analyze CTR by position — sometimes the 4th recommendation gets more clicks than the 1st
Test different positions — below the description, in the sidebar, or after reviews, and choose the position with the highest engagement
