Evaluating UX over time
Contents
Longitudinal Testing via panel
Exemplary Research Question
Exemplary Research Design
Long-term Testing via Web Analytics
Exemplary Research Question
- How do less search attempts within one session leads to more conversion?
- How many accounts could be converted from the old Ebay Kleinanzeigen to Kleinanzeigen?
Exemplary Research Design
- Timespan should be as long as possible, but be at least one or two years to see seasonal differences - The data should be queried weekly - The timing for collecting the data doesn't has to be special but should be in a small time area
Needed Data - data about how much time passes between search requests of one person inside one session - how many times does one person search a certain topic inside one session - how many sessions does one person start to search a similar topic -> the idea is to find out how many attempts does one person need to find what the persons looks for => the target is than to optimise the internal search engine
- Simple account history data (Date of the set up, number of sells, number of buys etc.) - This data has to be put in timely comparison -> pre-switch to post-switch -> the idea is to find "dead accounts" => the target is to find out if and how the switch effected usage and activeness of the app by users
