Book a call

Reducing search refinement to improve search ROI

Learn how the BCIT web team managed to reduce search refinement on their website to drive up revenue.

Stéphane Recouvreur

Stéphane Recouvreur 05 September 2022


British Columbia Institute of Technology

Founded in 1964, the British Columbia Institute of Technology (BCIT) inspires students beyond classroom studies and provides job-ready graduates to industries across Canada and the world. BCIT has established itself as one of Canada’s leading educational institutions in business, technology and trade programs.


Higher Education



Unsatisfactory search results

After the discontinuation of the Google search appliance, BCIT replaced Google site search with another search tool. The primary goal for the new site search at BCIT was to help site visitors find the content they need, quickly, without having to make multiple searches.

Google Search Appliance

The team also wanted the ability to promote results with the same styling as standard results and the ability to customise the look and feel of their search results pages.

However, the results were unsatisfactory, to say the least.

Alan Etkin, Web Analyst at BCIT, recounts, “Unfortunately, the results proved to be nothing short of frustrating for our users - even though we made extensive customizations to the interface, and configured it to search the specifics of our content.“

Within the configuration of this other search solution, the BCIT team prioritized searches for courses and programs, which had the unintended and adverse effect of burying results for common non-academic searches like parking, registration deadlines and service departments.

After receiving negative feedback from the community, and after exploring the limited options for improving the configuration, they started to look at other products.

Comparison test

The team started its comprehensive research process by evaluating several different search solutions.

The test covered both academic and general information queries. The team focused on their top search terms and highest search term refinements. Course programs were of course at the top of the list, but some less obvious keywords - such as “parking” for visiting students - were also included.

“We ran a detailed comparison looking at results across 30 search terms, and Funnelback – with basically out-of-the-box indexing – provided better results for 26/30 of the terms,” explained Alan Etkin, BCIT web analyst reports.

The team at BCIT and Squiz worked together to implement Funnelback Search.

Faceted Navigation

Faceted navigation, people search and related searches were important factors in creating a positive search experience for the team and website visitors.

Funnelback’s best bet feature and curator tool empowered the team to customize results in the style that matched BCIT’s existing results. The web team at BCIT promoted results to the top of search results pages that answered the most searched terms over the previous year.

Monitoring results with Google Analytics

The team had long been measuring the effectiveness of search with Google Analytics. In addition to the number of people using search, the team carefully considered search refinements - searchers who conduct a second search immediately after the first, likely because the first page did not contain the desired result.

The web team at BCIT had carefully calculated monetized goal values in Google Analytics for key behaviors on their site.

Search, it turned out, was a high-value activity on the website.

With Funnelback, they wanted to maximize these monetized behaviors and increase the value of each website visitor. Their new site search solution would need to increase the number of search users and reduce the number of search refinements.

Decreasing search refinements

“Our goal for internal search is to help site visitors find the content they need, quickly, without having to make multiple refinements to their search terms,” noted Alan Etkin at BCIT.

Using Google Analytics data to segment new and returning visitors to the BCIT site, the team has seen an overall 6.7% drop in search refinements. For mobile users, a growing percentage of site traffic, and search refinements have decreased by as high as 22%.

“Looking at our most searched terms over the previous year, we checked their search rankings and promoted approximately 50 to the top of the results. The search refinements for many of these terms dropped by up to 60%,” Etkin reported.

Distinct from the earlier solutions, Funnelback allowed BCIT to promote results and appear with the same styling as other organic results.

The knowledge that visitors who use site search are more valuable to the institution has already significantly impacted their business goals and results.

ROI of search results

Continuing to monitor key metrics in Google Analytics, the team saw that over the first six months with Funnelback search, site search users have a per-user goal value of approximately $114, compared to the site average of $23.

Key results

Site visitors using Funnelback search are meeting more of BCIT’s business goals. The team saw a 10.6% increase in the use of search and a 7.6% increase in per-user goal value from search users since implementing Funnelback. By multiplying the increased goal value by the total number of search users, BCIT has seen a solid return on investment.

Now the team is committed to driving more traffic to the BCIT site and encouraging visitors to use internal site search as much as possible. With Funnelback, BCIT has an intelligent search engine to power the site experience and drive its business goals.