Ecommerce site search query classification
A customer searches for a 'brown leather jacket' on an ecommerce site. The search returns relevant results – a jacket made of leather that is brown in color. However, if the seller does not have that particular type of jacket, the search still shows jackets or different variants, for instance, in other fabrics or colors. How does this happen?
This answer lies in 'context awareness,' which forms the backbone of an intelligent site search that can understand user intent to show accurate results.
Before we look at some of these query types, let's look at each classification.
- Query Spectrum
These types of queries are used to specify search ranges. Here, users are either looking for a product or a particular type of product, searching for a symptom/problem, or even a piece of non-product information. - Query Qualifiers
Users sometimes use certain qualifiers to specify conditions. For instance, they may look for specific features or themes, whether a product is compatible with other products, etc. - Query Structure
This means using the actual syntax of the query, i.e., how customers are searching and the language they are using.
Types of queries
There are 12 types of queries in site search, but even the top 50 retailers in the US only support some of them. Let's take a look at five types of queries.
- Exact search
This search allows the user to find exactly what they are looking for. They may search using the exact product name or model number. The key here is to solve typos and misspellings without making users type the query again. - Product type search
Users here are not looking for a specific product but a type/category. They may also use various synonyms or different variations of it, for instance, sofa/couch, etc. - Feature Search
This is one of the most common query types users use while they search. Features may encompass color, price, performance specs, etc. - Thematic search
These search queries are often vague with fuzzy boundaries. For example, searches may include searches based on occasions, seasons, etc. - Slang, abbreviations, and symbol search
Users don't always search the way the site wants them to. More often than not, they use synonyms, abbreviations, or symbols. Such as the USD '$' symbol or search for a 'blow dryer' instead of a hairdryer.
The site must understand intent and return results that match the user's expectations.
Learn more about delivering stellar search results for your shoppers. Book a demo with Unbxd, and discover the intent behind every query.