For You
For You recommendations
POST /indexes/{index_name}/recommend
Purpose
Surface personalized product selections based on a user's interaction history, preferences, and behavioral patterns. Leverages past purchases, clicks, cart additions, and browsing behavior to deliver highly relevant recommendations that match individual customer preferences and increase engagement.
Background
Personalized recommendations use machine learning to analyze user behavior patterns including purchase history, product views, cart additions, and engagement signals. The system builds individual preference profiles to identify products that align with each user's demonstrated interests, shopping patterns, and brand affinities.
When to use
- Homepage personalization: Show tailored content for returning visitors and logged-in users
- Email marketing: Create personalized product recommendations in newsletters and campaigns
- Account dashboards: Display relevant suggestions in user profiles and account pages
- Re-engagement campaigns: Target users with products based on their historical preferences
Example uses
| Use Case | Description | Interaction Source | Business Impact |
|---|---|---|---|
| Homepage "For You" Section | Display personalized recommendations on homepage for logged-in users | Purchase + browse + cart history | Increases engagement, improves homepage relevance |
| Email Recommendations | Include "Based on Your Purchases" or "You Might Like" in email campaigns | Purchase history + recent views | Drives return visits, increases email CTR |
| Account Dashboard | Show "Recommended for You" in user account or profile pages | Complete interaction history | Extends session time, improves loyalty |
| Retargeting Campaigns | Target users with ads featuring products aligned with their preferences | Behavioral data + preferences | Improves ad relevance, increases conversion rates |
| Mobile App Personalization | Create personalized product feeds within mobile applications | App usage + purchase patterns | Enhances app engagement, drives mobile sales |
Input sources (interaction types)
- Purchase History: Products previously bought by the user - strongest personalization signal
- Cart Activity: Items added to cart (even if not purchased) - indicates strong interest
- Product Views: Recently viewed or frequently browsed items - shows browsing preferences
- Engagement Signals: Likes, saves, shares, or other interaction data - reveals preferences
- Search Patterns: Historical search queries and clicked results - indicates intent and interests
Example (cURL)
curl -X POST "https://api.marqo.ai/indexes/product-catalog/recommend" \
-H "Authorization: Bearer $MARQO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"method": "for_you",
"user_id": "abc123",
"interaction_types": ["purchase", "cart", "view"],
"limit": 12
}'
Parameters
| Name | Type | Required | Description | Example |
|---|---|---|---|---|
| method | string | yes | Recommendation strategy identifier. | "for_you" |
| user_id | string | yes | User identifier for personalization. | "abc123" |
| interaction_types | array[string] | no | Types of interactions to consider (purchase, cart, view, engagement). Default: all types. | ["purchase", "cart"] |
| limit | integer | no | Max number of results to return. Defaults to 10. | 12 |
| offset | integer | no | Offset for pagination. | 0 |
| filters | object | no | Server-side constraints (e.g., in_stock, price ranges, brand, category). See Marqo Filter DSL. |
{ "in_stock": true } |
| session_id | string | no | Optional session identifier. | "xyz789" |