For several years, usability guru Jakob Nielson has had “bad Search” as his number one “Top 10 mistakes in web design.” The reasons for this are obvious. If users who come to your site cannot quickly find the information or solution they are looking for they will leave and go to a competitor.
So website owners have tried to refine “user journeys.” However, typical users do not want to take your pre-defined “user journey” and read all your carefully crafted marketing messages. Instead, they want to define their own journey in different ways.
This is where conflict between the website owner and the website user can happen. The first wants to prescribe the journey and the second wants to do their own thing. It is therefore best to work with the user, to help facilitate their journey, but not necessarily prescribe it.
Search and search driven functionality puts users at the center of the web experience. With the right tools, web owners can facilitate what site users find, see and read, watch, play with or listen to, to enhance customer satisfaction and conversion.
Search with accurate and clear result landing pages are still one of the most important routes to content. Additions such as predictive search query suggestions can speed up the initial search. Results pages can then be clearly presented with the relevant search words displayed, pagination and faceting choices available.
Website owners can assist here by carefully “boosting” specific pages in search results (although purists would argue that search results should not be altered).
Brittany Ferries website showing predictive search query suggestions and search results landing page with pagination and faceting options.
Search does not need to be based on only one data source such as the content in your web content management system. Federated search allows users to search multiple data sources at one time. This allows you to expose data from various CRM systems or other applications if you choose. The results can be displayed in a single presentation or grouped to help users understand the content.
For example, an organization with numerous data sources like a council can use federated search. A “find my nearest” search on a post code or street name can return results for local councilors, nearest schools, leisure facilities, bin collection days, libraries, bus stops, etc. For additional engagement, this can be grouped under headings and the data can all be displayed on a map.
The number of websites that are adopting a “search first” approach is also growing, with councils like East Renfrewshire taking the lead.
East Renfrewshire Council website search first homepage
For more complex offerings, a faceted search can speed up user’s browsing. Facets, which are pre-defined groupings, allow users to quickly refine large numbers of results to a manageable list that meets the user’s specific criteria.
For example, a holiday search might have facets on number of bedrooms, availability of childcare and proximity to golf courses. These types of facets might be more important than which resort the holiday is in.
An extension to this, for searching over geographic locations, is to have the results highlighted on a map. This is particularly useful when users are unfamiliar with the geographic location they are looking at. When viewed on a mobile device, the additional benefit of content specific to the users location greatly enhances user experience.
Berkeley Group website map based search results
Using analytics software, websites can track all user journeys and present back augmented results that are related to a specific search. So if a user searches for “Cafés in Leicester Square,” as well as displaying the results for “Cafés in Leicester Square,” if other users have performed a similar search and gone on to look at other bars in central London, then these can be displayed in a “you may also like this” area.
Consideration also needs to be given to searches performed on mobile devices. 90% of smartphone searches result in an action such as making a purchase or visiting a business. Whilst out shopping, 79% of smartphone users report that they have used their phone to help with a purchasing decision. All of the above search techniques can be added to mobile websites, but some may work better than others. Using a device’s native geo-location functionality is a particularly strong way to display search results.
There are many tools (search engines) available to implement the search experiences described. The open source arena is dominated by the highly capable Apache Solr engine, itself based on the Apache Lucene library. These are used by many large sites like Instagram and The Guardian newspaper.
The advantages of a search engine like Apache Solr, is that in addition to providing a reliable and robust service, it is also highly expandable and configurable. This is vital to be able to quickly meet new search requirements as new needs and concepts emerge.
These examples show the tip of what is possible to enhance search functionality to deliver better results on websites and mobile devices. Better search results mean that web users can find the information they want quickly, in their own way, which in turn enhances customer satisfaction and conversion.
Title image courtesy of Edward Westmacott (Shutterstock).
Editor's Note: You may also be interested in reading: