Smart search
My role: UX, UI & Interaction Designer
The challenge:
Treat search as a conversation where answers are being provided based on context, user intent and data!
I believe that being able to understand and to support your users when they are searching for something on your website, can play a big role in having a happy and loyal customer.
Shortcomings of search engines
1. Most searches are designed for people who know what they are looking for
2. Search is based on each individual word instead of semantic search
3. Optimized for one-off transactions instead of being optimized for conversation
Ideal search behavior
The ideal search should meet the following criteria:
1. It will focus on customer intent
2. It will focus on conversation rather than on search
3. It will support the user through all the stages of the customer journey
4. It will give a sense of control to the user
Visual mockups
In order to identify the user intent, the search will use natural language processing and the history of the user with the website and with other channels.
Auto suggestions will be shown based on the same principles.
If the user is stuck for more than 5 - 10 seconds (this time window needs to be tested) before typing, then a smart assistant will appear in order to guide the users in finding their solution.
The users will have control over the smart assistant: if they want to continue exploring, then they can close it and continue their usual journey.
Another situation when the smart assistant can be used in search is when there are too many results (as a general rule, more than 10 results).