It would probably be a fair to say that site search functionality on a vast number of websites can be a laborious and generally bad experience for customers. So what are the problems that many of them face?

  • Language – in the same way that many people say words in different ways and even use different words to mean the same thing, site searches require the terminology of the user to match the words used in the product index to get good results
  • Boolean dilemma – when two or more words are used in a search query, the in-site search engines will connect the two words together to run the search using Boolean connectors such as “AND” or “OR”. This can cause a problem if a person is searching for a specific product and will often lead to poor results
  • Spelling errors and typos – it would probably be a fair estimation that more than half of all search queries are spelt wrong. Missing one single character can produce a zero search result
  • Naïve search results – the traditional approach to verify search results is using word frequency and word proximity. The problem with this is that there is no guarantee that either word frequency or word proximity equate to relevancy because it does not take into account the context in which the word is being used
  • Field Weightings – many search engines apply the same level of importance to product description and the product name or the product category, when product category is more often than not the most important element. As a result many in-site search engines deliver results that appear out of context, even though technically they are correct
  • Optimisation and Control – once implemented, many search engines can be hard to adjust and optimise the way in which it works, particularly if you wish to create a certain order of results when trying to promote certain products.

Unfortunately we live in an environment where we have come to expect instant results particularly in the online world. When it comes to search, having a website that cannot deliver a simple search query correctly is only going to drive customers to look for alternatives.

The views expressed in this blog are the views of the author, Andre Brown and do not represent the view of Locayta, its employees or its shareholders. For more information about Locayta, visit