Enhanced Search Functionality in Optimizely Experimentation
Problem:
Currently, the search function in Optimizely Experimentation only filters by a single word, even if multiple words are entered. For example, if a user searches for “Headers for PDP”, the results only return matches for “Headers”. This limits discoverability and forces users to rely on entering very specific single keywords.
Proposed Solution:
Update the search functionality to apply all words entered in the query, rather than restricting results to the first word only. This way, users can perform both broad and specific searches as needed.
Reasoning / Business Impact:
For customers running numerous tests, Efficiently searching and reviewing experiments is critical for both ongoing analysis and future reference. A fully functional multi-word search would save significant time and improve overall usability.
In short, this is a fundamental search capability that users expect, but it currently does not work as intended.
chabu chabu
shared this idea
-
Simon Born
commented
Performance (ie. speed) of the search is a related topic that I currently find sub-par and should be improved.
-
Kristian Reich
commented
Also need to have search corpus include variable datasets and values. When you have json for most variables, understanding what values exist in which variables is critical - especially when you have a large number of variables.
Alternate: allow user to include/exclude data sets to the search corpus so I could add variables to specific queries without impacting every other base data set and default queries don't have to include all to limit performance impact. By variable datasets I mean default it the smallest dataset, but, critically, allow me to expand my search radius when I need to. I don't always need to search history, but when I need it, I need it. I don't always need to search variant json, but sometimes I absolutely need to. In those cases, I select from drop-down to access/include those search corpus. Much like segmenting in reports.
Target data sets: flag names, flag descriptions, rules, audience, events, variables, variations, user actions (history), etc... ALL the things.