In any electronic catalog or web store, it’s essential to make it easy for customers to find products, and to achieve this, it’s crucial to have the most effective product search engine possible.
We can classify search engines according to where they perform their searches:
Internal search, focused on internal knowledge (own domain). This type of search engine is designed to search only within the system’s own database. That is, it only shows you results for products, categories, or information already stored. This is the most traditional approach and focuses on offering accurate and relevant results within the system’s existing knowledge.
External search, performs searches outside the system’s own knowledge base. The search engine accesses one or more external sources, such as product databases, supplier catalogs, specialized search engines, or even information on the web, allowing it to access knowledge not yet in its databases.
Hybrid or extended search, which combines internal and external search. These search engines combine the best of both worlds: precision, speed, and extended knowledge, making them the best solution of all.
There are various technologies that allow for the implementation and management of hybrid search engines, the most relevant of which are:
Algolia: Offers hybrid search and can integrate with multiple sources, enabling expanded search across websites and applications.
Coveo: Specializes in intelligent and personalized search, combining internal and external data to deliver relevant results.
Elasticsearch: Highly flexible and powerful, it can be configured to search multiple sources and combine results.
Swiftype: Platform that enables hybrid and personalized search, ideal for B2B sites that need to expand their search capabilities.
The problem with all these technologies is that they require a knowledge base to feed them.
At Factory Data, we have the NEO system, which implements search technology for vehicle spare parts products. This allows us not only to improve internal search engines but also to analyze all search data to detect new SKUs with sales potential, potential obsolete products, missing equivalents in the internal database, etc.
