AI Assistant is here! - Make the leap to conversational eCommerce

Frequently Bought Together

An automated recommendation feature that displays products other customers have purchased together with the product the user is currently viewing, based on real co-purchase data.


Definition

Definition

The system analyzes order history to identify products that are frequently purchased together with statistically significant frequency. It uses collaborative filtering: recommendations are based not on product similarity, but on the actual purchasing behavior of thousands of users. If many buyers of Camera A also purchase Battery B and Case C, that association emerges automatically without anyone having to configure it.

The co-purchase signal is stronger than the co-view signal, but slower to accumulate because it requires transaction history. The system identifies product pairs or groups that occur together more often than chance would predict and generates a list ranked by purchase affinity.

From a business perspective, this type of recommendation has an advantage over traditional upselling: it does not require convincing the user to spend more on the same product, but rather to complete their purchase with items they are likely to need. As a result, it creates less perceived resistance and delivers more tangible value.

What it's used for

What it's used for

Increases average order value (AOV) organically without requiring manual product-by-product configuration. It is particularly effective in categories where products naturally complement one another, such as electronics and accessories, apparel and add-ons, or food and related items. Because the recommendations reflect actual purchasing patterns, they are generally perceived as helpful rather than intrusive.

Doofinder

Doofinder includes a recommendations module that allows merchants to display Frequently Bought Together blocks on product pages. The system learns from the store's own transaction data to identify the most relevant co-purchase relationships, eliminating the need for manual setup on a product-by-product basis.

Merchandising rules can also be used to prioritize or exclude specific items from recommendations, for example, to promote slower-moving inventory or maintain minimum margin requirements.

Example

Case study

A photography equipment store adds a Frequently Bought Together block to the page of an entry-level mirrorless camera.

The algorithm identifies that 34% of customers who purchased the camera also bought an extra battery, while 28% included a carrying strap in the same order. Both products are displayed as preselected recommendations within the block, and the combined price of all three items is calculated automatically.

References

Sources