Before you use this logic
Keep in mind that this logic may not always have recommendations for every products. This means that if a product is new or doesn't have another frequently purchased together products, there won't be recommendations for that product. If you still want to display recommendations for the product, you can use a fallback recommender.
Also known as Customers Who Bought This Also Bought. Use this to cross-sell other products bought by customers. This algorithm looks at purchase history of customers who've purchased a given product. Then recommend other products based on their purchase history. Results will be similar to Frequently Bought Together if most of your customers are one-time purchasers. The algorithm used to implement this is association rules.
| Model Update Interval | 1 day |
| Can Manually Update Model | No |
| Can Preview Output | No |
| Recommendations Build Time | Expect to wait for a few minutes. |
Requirements
- Product combinations (A -> B) must be purchased for at least once.
- Lift must be at least 1.5. This means that users are at least 1.5X more likely to buy product B if they've purchased product A.
Where should you place this recommendation?
Product Page
Due to it's high relevance, we recommend placing this recommendation in product page above all other product recommendations.
Cart Page
If this logic is used on cart page, we will find product recommendations based on your cart items. If cart is empty, there won't be any product recommendations.