An introduction to product recommender systems — Mastercard Dynamic Yield
2026-02-28 21:33 Diff
https://www.dynamicyield.com/lesson/introduction-to-product-recommender-systems/

Summarize this articleHere’s what you need to know:

  • Recommender systems personalize the shopping experience: These systems leverage machine learning to analyze customer data and suggest relevant products, boosting engagement and sales.
  • Data drives the recommendations: The effectiveness of recommender systems hinges on the quality and quantity of data available about users, products, and their interactions.
  • Variety of strategies for diverse needs: Recommender systems adapt to different user profiles and data availability. Popular items, similar product suggestions, and collaborative filtering are some common strategies.
  • Beyond just recommendations: These systems can be used to personalize various aspects of the shopping experience, like search results, promotions, and content.
  • Measurable benefits: Recommender systems can demonstrably improve customer experience, drive sales growth, and enhance a company’s bottom line.

Product recommender systems surface items available for purchase across web pages, mobile apps, within emails, or on any connected screens, such as kiosks and various IoT devices. One of the most popular methods used by retailers, recommendations guide visitors to products they are likely interested in, improving the discovery process and helping them find what they want more efficiently.

Today, retailers often have thousands (and sometimes millions) of products in their inventories, making it difficult for consumers to dig up exactly what they are looking for. And with personalized recommendations, brands can help users easily find relevant products based on their affinities, trends, interests, and behavior, with an end goal of driving sales, upsells, cross-sales, larger cart sizes, and higher average order values (AOVs).

Recommendations are currently used by some of the world’s most innovative brands. Amazon, for example, has been refining and tweaking its algorithm for two decades. Powered by the company’s unprecedented access to massive amounts of consumer data, its recommendation system has completely transformed the way customers are matched with products they are most likely interested in purchasing.

An example of just one of the ways Amazon serves users recommendations on its desktop site

What are product recommender systems?

Powered by machine learning, a product recommender system is the technology used to suggest which products are shown to individuals interacting with a brand’s digital properties. Fueled by a number of algorithmic decisions, recommendation algorithms mine user, product, and contextual data – both onsite and offsite – to present every user with a personalized experience.

Improving the discovery process, this helps users find what they are looking – and sometimes products they don’t even realize they are looking for. In doing so, businesses can learn more about each user’s unique preferences and interests, optimizing performance in real-time while simultaneously refining their testing roadmaps for the long-term.

And when it comes to product recommendations, there is no archetypal strategy marketers should use for every widget. Different strategies must be applied for different users, depending on the amount of information available about the customer, their behavior, and the context of products on a site. This includes site behavior, status, geo-location, time of day, past purchases, and more.

The strategies behind recommender systems

Recommendations can provide key insights and the opportunity to better understand who a customer is in order to delight them, add value, and improve the overall relationship with a brand. And when it comes to the strategies behind each of them, there are three primary tiers:

  • Global recommendation strategies
  • Contextual recommendation strategies
  • Personalized recommendation strategies

Each of these strategies dictates which products are included in an experience. Examples of different recommendation strategies that fall under each tier are listed in the figure below.

Picking a recommendation strategy

When picking a strategy to use, marketers must first assess the amount of user and product data available, as well as a user’s location in the purchase funnel, allowing this information to dictate which strategy to deploy.

Global strategies
These strategies tend to be the easiest to implement, simply serving any user – both known and unknown – the most frequently purchased, popular, or trending products in a recommendation widget.

Contextual strategies
These strategies rely on product context, assessing product attributes, such as color, style, the category it falls under, and how frequently it is purchased with other products, to recommend items to shoppers.

Personalized recommendation strategies
Personalized strategies, the most sophisticated of the tiers, don’t just simply heed context, but also the actual behavior of users themselves. They take the available user data and product context into consideration to surface relevant recommendations for each user on an individual level. This means, in order to effectively deploy them, a brand must have access to behavioral data about the user, such as purchase history, affinities, clicks, add-to-carts, and more.

In certain situations, the user’s interaction history can even be coupled with the history of all other users on a site to recommend products. For example, using a “Collaborative Filtering” strategy, users are recommended items based on the preferences and interactions of site visitors expressing similar behaviors. Meanwhile, affinity-based recommendations are based on data a recommender system automatically aggregates to build affinity profiles for each customer, suggesting items according to understood user preferences.

Global strategies can be used for every type of site visitor, whether new, returning, or loyal. And contextual and personalized strategies can be used within the first session or pageview if contextual or behavioral data is available for the user, such as their geo-location (contextual) or their affinities (personalized).

The potential to drive revenue increases when recommendations are best suited to user context and behavior. For example, marketers should use contextual or personalized strategies for frequent shoppers and VIP users, who they have ample user data on. Additionally, these strategies are strengthened when layered atop each other, maximizing the impact of the deployed recommendations.

Crafting product recommendations

The strategies behind recommendations must be identified in advance and must explain the logic behind the recommendations themselves. Examples of this logic include:

  • Recently viewed
  • Most popular
  • Show similar items to the item currently in view
  • Show items based on the visitor’s browsing history
  • Show items that are viewed together with an item currently in view

An example of a “Most Popular” recommendation widget on a category page

These strategies will enhance (or limit) the recommendations served to each user, allowing marketers to tailor the widgets to best optimize performances and the desired objective. Once a marketer has selected a strategy to use for a recommendation widget, they can set up rules, such as “exclude recently purchased items” or “only include items that cost $25 max” to narrow down which products are featured in each widget.

Additionally, certain strategies are better suited for specific pages. For example, “Viewed Together” is ideal for product pages, as they relate to products already in view. “Purchased Together” is a good fit for cart pages, especially right after a user adds an item to their cart. And “Most Popular” is a great homepage recommendation strategy, helping to kickstart the discovery process for both new and returning visitors.

Combining context and intent

Recommendation strategies and algorithms employed should vary according to each user’s displayed intent, and user signals help infer and identify the likelihood of a conversion.

  • Users with low levels of intent are often new visitors, unidentifiable, arrive via search or social, or are on a mobile device.
  • Users with a medium level of intent are often returning visitors, identifiable, arrive directly to the site or via email campaign, or interact with a site on a desktop device.
  • Users with a high level of intent have either made a purchase in the past, have added items to their cart, or search products directly on the site.

These intent level classifications inform the strategies and rules each marketer chooses for a test. For example, for users with low intent levels, the “Viewed Together” strategy will encourage further product discovery and exploration. Meanwhile, for a user with a high level of intent, such as a past purchaser, a “Purchased Together” strategy is a better fit, increasing the likelihood of cross-selling the visitor.

Data and recommender systems

In order to effectively deploy recommendations, you need data. Both implicit and explicit, this data is automatically ingested by the system, allowing it to identify which products to serve to each user in a recommendation widget. A recommendation engine can collect online behavioral data about every single user. Sites collect first-party data, while data from third-party sources, such as CRM or offline purchase data, can be onboarded. The more data available, the stronger your targeting strategy becomes.

Types of user data

Being able to access the following types of user data can help power smarter recommendations:

  • Location data: The country, region, or city a user is located in
  • Tech data: The type of device a user is accessing a site from, the browser-type they are using, their operating system, and more
  • Demographic data: A user’s gender, age, marital status, etc.
  • Behavioral data: Actions a user takes on-site, including clicks, add-to-carts, hovers, number of pages viewed, etc.
  • Affinity-based data: The interests and preferences displayed by a user on-site
  • Online and offline purchase data: The products a user purchased on-site or in-store
  • Traffic source data: The source of traffic a user is visiting from, including direct traffic, paid traffic, social traffic, and referral traffic
  • 3rd-party data: Information about a user from outside sources, onboarded through a DMP.

Access to this data helps marketers identify buyer personas and understand purchase habits, which can inform segmentation strategies. Other elements, like knowing a user’s traffic source or the device they are using when interacting with a brand, can inform targeting strategies.

The more a user visits and interacts with your site, the more data becomes available. The system can then use this data to predict what users are most likely to purchase, make smarter segmentation decisions, and serve more relevant recommendations.

Visitor Type

Data Available

First-time visitors

Only context (IP / geo, traffic source)

2+ visits

Some usage data, some product-viewed data

Repeat visitors

Purchase history, deeper usage data, user affinities

Types of product data

Product data is also essential, as they give recommender systems the ability to put the right items in front of the right users. Product data feeds, which house a retailer’s entire product catalog, can be synced with recommender engines. Product feeds include basic information about the products available on a site, including, but not limited to:

  • SKU
  • Product name
  • Product URL
  • Price
  • In-stock status
  • Images
  • Keywords

Marketers can customize product feeds, adding in more product data that can later be used for targeting, such as color, size, style, etc. The more product data is available, the stronger your contextual and personalized recommendation strategies become.

The power of product recommendations

Product recommendations are a great way to improve the overall user experience, guiding visitors through the discovery process while simultaneously generating more revenue for your organization. And with machine learning doing much of the heavy lifting, marketers can experiment with different strategies, segmentation practices, and widgets at scale, exposing more products to users than ever before to generate ROI and boost their companies’ bottom lines.

Continue reading