Difference between revisions of "A/B Testing"

From MgmtWiki
Jump to: navigation, search
(Created page with "==Full Title or Meme== A stratagem for comparing two User Experience options by running both in parallel and comparing results. ==Context== ==References==")
 
(Context)
Line 2: Line 2:
 
A stratagem for comparing two [[User Experience]] options by running both in parallel and comparing results.
 
A stratagem for comparing two [[User Experience]] options by running both in parallel and comparing results.
 
==Context==
 
==Context==
 +
The difficulty in determining the impact of changes on the [[User Experience]] at a [[Web Site]] has lead to a range of techniques, the easiest to implement being the simple expedient of just enabling any proposed, and pretested [[User Experience]] as a live version of an already deployed [[Web Site]], or other rendering of a [[User Experience]]. Under the assumption that pre-testing has shown no problems, the best approach is to introduce it to a small of [[User]]s and ask for feedback from the user or just to monitor their reactions to the site. The larger the change, the more likely to take a while before the user response can be fairly evaluated.
 +
==Problem==
 +
==Solutions==
 +
 
==References==
 
==References==

Revision as of 15:05, 17 August 2018

Full Title or Meme

A stratagem for comparing two User Experience options by running both in parallel and comparing results.

Context

The difficulty in determining the impact of changes on the User Experience at a Web Site has lead to a range of techniques, the easiest to implement being the simple expedient of just enabling any proposed, and pretested User Experience as a live version of an already deployed Web Site, or other rendering of a User Experience. Under the assumption that pre-testing has shown no problems, the best approach is to introduce it to a small of Users and ask for feedback from the user or just to monitor their reactions to the site. The larger the change, the more likely to take a while before the user response can be fairly evaluated.

Problem

Solutions

References