Search is one of the most important parts of the ecommerce experience. Visitors who use search often have a clear purchase intent, making search a key driver of both product discovery and conversion.
In Voyado Product Discovery, the product discovery engine combines product intelligence, query analysis, and ranking logic to help shoppers find the right products. Instead of relying only on keyword matching, the engine works to understand what the customer is looking for and then matches that intent with the most relevant products in your catalog.
This article provides an overview of how search works in Product Discovery and introduces the key concepts that influence search results.
When a customer enters a search query, the engine performs three main tasks:
- Understand the search intent
- Find relevant products
- Rank the results
All three steps are required to deliver relevant search results.
For example, if a customer searches for:
black running jacket
The engine needs to:
- Understand that jacket is the product type
- Recognize that black is a product attribute
- Understand that running describes the intended use
- Find products matching those concepts
- Determine which of those products should appear first
Understanding search intent
The first step is understanding what the customer means. The engine analyzes each query and identifies meaningful concepts such as:
- Product types
- Brands
- Materials
- Colors
- Usage and occasions
This process is called query analysis. By understanding concepts instead of relying only on exact keywords, the engine can return relevant results even when the wording differs from the product data.
Matching products
Once the search intent has been understood, the engine matches the query against your product catalog.
Because products have been analyzed and enriched when they were imported, the engine can match products based on meaning rather than exact wording.
This allows shoppers to find products even when:
- Different terminology is used
- The search contains spelling mistakes
- The search is very specific
If no exact matches are found, the engine can automatically relax less important parts of the query to avoid empty result pages. Read more about search criteria relaxation.
Ranking results
Once relevant products have been identified, the engine determines the order in which they should be shown.
Product ranking takes into account factors such as:
- Customer interactions
- Purchases
- Product lifecycle signals
- Stock availability
- Business priorities
The goal is not only to show relevant products, but also to prioritize the products most likely to satisfy the customer and perform well commercially.
Search experiences across the customer journey
The same intelligence that powers search is used throughout Product Discovery. Understanding products, customer intent, and product relationships is valuable wherever customers discover and evaluate products, not just when they type something into a search box.
Search intelligence supports several parts of the customer journey, including:
Search and autocomplete
Understanding what a customer is looking for and matching that intent to relevant products.
Category and landing pages
Helping rank and expose products based on relevance, customer behavior, and business priorities.
Product recommendations
Using product understanding and customer behavior to suggest relevant alternatives, complementary products, and personalized recommendations.
Facets and filters
Grouping products into meaningful attributes and concepts, making it easier for customers to refine large assortments.
Product discovery
Helping customers browse and discover products through categories, recommendations, and merchandising experiences, not just through direct searches.
Personalization
Using customer behavior and product understanding to create more relevant experiences for individual shoppers.
Merchandising and product exposure
Supporting automated ranking and exposure strategies that balance customer relevance with business goals.