Build commerce search pages

This is for:

In this article

How to build the search experience
Machine learning and personalization

Search pages are feature-rich user interfaces that let visitors search for products in your storefront. In Coveo for Commerce, search pages are the interfaces that power the Search product discovery solution.

Coveo for Commerce search pages provide the experience your commerce visitors expect, with features such as:

An intelligent search box that suggests personalized queries and products as visitors type.
Intelligent and personalized search results that are tailored to each visitor.
Dynamic faceted search that lets visitors filter search results by product attributes.
Pagination that lets visitors navigate through search results.

How to build the search experience

To implement a Coveo-powered search page:

Build your search page user interface.
Set up facets for the search page using the Facet manager in the Coveo Merchandising Hub (CMH).

Note

If the Facet manager isn’t enabled for your organization, you’ll have set up your facets in the next step using the Query Configurations API.
Create query configurations using the Query Configurations API. These configurations set certain parameters for your search pages, such as sort criteria and additional fields.
(Optional) Create or modify rules for the search page using the Search manager in the Coveo Merchandising Hub (CMH).

Machine learning and personalization

To leverage Coveo-powered commerce search pages to their full potential, configure Coveo Machine Learning (Coveo ML) models that will:

Personalize search results for each visitor.
Increase your product inventory coverage.
Improve the search relevance.
Provide relevant filter options.
Increase important business metrics such as conversion rate and average order value.

For optimal results, you should configure the following Coveo ML models to power your commerce search pages:

Model type	Purpose
Predictive Query Suggestion (PQS) or query suggestions	These models suggest personalized queries to visitors as they type in the search box. This feature helps visitors find what they’re looking for faster, and it increases the chances of them finding products they’re interested in. PQS brings personalization to another level by leveraging Coveo Personalization-as-you-go capabilities. To learn more about PQS models and how to configure them, see About Predictive Query Suggestions (PQS).
Automatic Relevance Tuning (ART)	This model improves the ranking of products presented in search results. It learns from user interactions like clicks, add-to-cart, and purchases to automatically adjust the ranking of products in search results and listing pages. ART uses machine learning to predict the relevance of each product for a given query and user context, and then adjusts the ranking of products accordingly. To learn more about ART models in the context of commerce and how to configure them, see About Automatic Relevance Tuning (ART) for Coveo for Commerce.
Intent-Aware Product Ranking (IAPR)	This model personalizes and ranks search results for each visitor based on their intent. It detects visitors' intent in real time and can quickly adapt to changes in behavior. This model also leverages Coveo Personalization-as-you-go capabilities and can increase your product inventory coverage by providing products that visitors are likely to be interested in based on their intent, not just on popularity. To learn more about IAPR models and how to configure them, see About Intent-Aware Product Ranking (IAPR).

Model type

Purpose

Predictive Query Suggestion (PQS) or query suggestions

These models suggest personalized queries to visitors as they type in the search box. This feature helps visitors find what they’re looking for faster, and it increases the chances of them finding products they’re interested in. PQS brings personalization to another level by leveraging Coveo Personalization-as-you-go capabilities.
To learn more about PQS models and how to configure them, see About Predictive Query Suggestions (PQS).

Automatic Relevance Tuning (ART)

This model improves the ranking of products presented in search results. It learns from user interactions like clicks, add-to-cart, and purchases to automatically adjust the ranking of products in search results and listing pages. ART uses machine learning to predict the relevance of each product for a given query and user context, and then adjusts the ranking of products accordingly. To learn more about ART models in the context of commerce and how to configure them, see About Automatic Relevance Tuning (ART) for Coveo for Commerce.

Intent-Aware Product Ranking (IAPR)

This model personalizes and ranks search results for each visitor based on their intent. It detects visitors' intent in real time and can quickly adapt to changes in behavior. This model also leverages Coveo Personalization-as-you-go capabilities and can increase your product inventory coverage by providing products that visitors are likely to be interested in based on their intent, not just on popularity.
To learn more about IAPR models and how to configure them, see About Intent-Aware Product Ranking (IAPR).