How do I know Fuse's search is performing well?
Fuse uses an approach called ‘Engine Scoring’ to measure the relevancy of our search engine. The basic premise behind this is, that if every time a user performs a search and the content they needed is ranked 1st in the search results, we would have a relevant search engine.
Fuse records every search query and the position of content clicked on by the user in a search. For example, you might search for ‘Spring Release 2022’ and click on the 2nd result in the list. Fuse measures this interaction at scale (a minimum of 10,000 queries and clicks) to generate an engine score.
The table below shows the number of clicks in each result position on the first page of the search results:
In 95% of cases, the user finds what they want on the first page of search results.
75% of the time, the user finds what they want within the first 3 results.
Result page | Result position | Click count | % of total |
|---|---|---|---|
PAGE1: 95% | 1 | 4963 | 47% |
2 | 1865 | 18% | |
3 | 1030 | 10% | |
4 | 655 | 6% | |
5 | 419 | 4% | |
6 | 337 | 3% | |
7 | 238 | 2% | |
8 | 221 | 2% | |
9 | 200 | 2% | |
10 | 165 | 2% |