eCommerce Search: An Analysis of Query Types
Understanding the types of query used on an eCommerce website is an important first step to providing great relevancy.
eCommerce search excellence directly and measurably impacts sales.
Search is the most popular product location strategy for most shoppers, especially where the product catalogue is large. At Search Technologies, we work with all of the leading search products in the eCommerce space, and we help customers to tune relevancy to suit their product set.
Categorizing eCommerce Query Types
In seeking to improve eCommerce search relevancy, a great way to start is to analyse and categorize queries. To do this, you’ll need access to recent search logs, plus appropriate experience or tools to process and order them.
Each of the different query types requires a specific strategy for improvement. Here is a brief overview of five of the more common query types:
Exact product search: The customer knows exactly what they want. Search results should take the user directly to the appropriate product. Precision is important. The worst case scenario is that you stock the product, but an exact search fails to find it. Abandonment (on the assumption that the product is not in stock) is the usual consequence. Examples: Gibson Les Paul Standard, SONY Bravia KD49X8505BBU.
Generic product search: The customer has a broad idea about their intended purchase, but is looking for guidance. In addition to showing products that fully comply with the query, the search results should provide an appropriate range of navigation options, enabling the shopper to explore the catalog, understand what is available, and refine their way down to the product that is right for them. Examples: Laptop computer, red dress.
Qualified product search: Queries, whether exact or generic, are often qualified by additional terms, and depending on the nature of the product, a wide range of qualifiers can be used including color, size, and price range. The inclusion of these additional words in the search clue poses challenges for the query parser. Ideally, the search engine should be able to interpret qualifiers accurately, and map terms to appropriate fields within the search index. Example: blue dress size 10.
Search qualifiers are not always concerned with the physical attributes of a product. A qualifier could be the name of another product which is related to the need, for example, ink cartridge for HP Inkjet 2540. The qualifier can also be subjective, such as cheap flights to New York, or high quality men’s boots. Again, correct interpretation through query parsing is the key to getting this right. Some search products have good capabilities in this area. Alternatively, specialized tools such as Search Technologies QPL (Query Processing Language) can be used. QPL is search engine independent.
Symptomatic search: Depending on the context, these queries can be tricky to deal with. The customer’s query describes a problem that they are hoping to solve, rather than giving a description of the product that they wish to purchase. For example, sore throat treatments.
Non-product search: Sometimes visitors are looking for information about your company or services rather than a product, but they use your main search capability anyway. Perhaps they are looking for information about returning a product, or they have a query about a shipment.
With all of the above, it is best to be guided by your search logs, in terms of how much effort you need to put into addressing each query type. The mix varies, depending on the product catalogue.
Search Quality Analysis Services
At Search Technologies, we provide a search quality analysis service. It enables one of our experienced professionals to join your team, help define the query categories that you need to focus on, and then build and implement an action plan to improve your search relevancy, and your revenues.
Contact us for more information about this service.