Updated Search Functionality of Galvanizeit.org
Whether you are a frequent visitor or newbie to the wealth of knowledge living on galvanizeit.org, it doesn’t take long to learn the value of the search box. In the past the AGA has used Google Search to improve the results returned on any search of the site. Overall this system worked quite well and we had few complaints. However, when the website was upgraded earlier this year, the search functionality was not operating as desired.
During the upgrade of the website, AGA staff and users noticed that the search function seemed less “smart” than the old site. After investigation, we discovered that the search function had defaulted to the content management system (CMS) search rather than back to Google – and was not producing the best results. After a discussion with our web developers weighing the positives and negatives of many possible options, we opted to upgrade to a new Software as a Service (SaaS) platform, Algolia.
In October, the new search platform went live. If you haven’t utilized its power yet, I highly encourage you to check it out. There are a number of key, noticeable improvements provided by the system. The first, and probably most obvious improvement, is the speed of the search. As soon as you click inside the search box, Algolia is working to find results for you. The results populate immediately in an overlay on the page where you started the search and will continue to refine as you keep typing keywords allowing for a much more seamless experience. Furthermore, the results show your input terms highlighted in blue. The previous search required you to input your search term and hit enter or click the search button with you mouse to take you to a results page where all content was in one scrollable list. This may not seem like a huge difference, but once you use the new system, you realize just how much more effective and efficient it is.
Another improvement that is immediately apparent is the federated search results. Federated search basically means the results are categorized into separate “types” or indices. For the AGA, those results are federated into either main site content, Dr Galv/Knowledgebase or Project Gallery. This helps the user easily see what “type” of content the result is before clicking on the link.
So far, the improvements we have discussed have been “front-end” user experience items. However, the long-term benefits of Algolia truly lie in the back-end capabilities. The system allows AGA to control the results returned in a number of ways. First, we are able to develop general rules to weight the content on the site and within the pages. For example, we can tell the system that a keyword match in the title of a page/article is more important than a keyword match in the body of a page. We can also set our own synonyms, teaching the system that words such as “cold galvanizing” and “zinc rich paint” should be considered equal. We can also pull analytics from the platform that show us what people are searching, how frequently, and what results they are receiving. Therefore, overtime, we can continue to improve the user’s search experience by; making synonyms, adjusting typo tolerances for misspellings and common input errors, and resolve any ‘No Results’ searches – allowing us to create better content relevant to what the user is looking for.
I believe the AGA and its users can easily see the numerous benefits that the new system provides. We continuously strive to make a user’s experience on our site better, faster and easier, and the improved search functionality is another great leap forward in accomplishing this goal.