Web search in particular is one of the most utilitarian instances of software these days. Having spent 2.5 years doing search quality / UX assessment at MSFT, I'm a firm believer that every change in search should be tested vigorously and that a design team that isn't enthusiastic about testing isn't worth having.
That said, my opinion for other software products is less strict. The critical flaw, in any situation, is putting design modifications to test without attempting to learn from the test to improve future design. Playing roulette with a testing protocol every time you do something new is a horrible way to do business, even if you have the user traffic to detect miniscule differences. In many, if not all, test protocols, the design of the test should help the team better understand the users, tasks, etc. so that future design decisions can be made more effectively. AndyEd... http://surfmind.com On Sat, Mar 21, 2009 at 3:08 AM, Peter Van Dijck <[email protected]> wrote: > If the 2 versus 4 pixels thing is on a crucial page like the Google search > results or list of adsense ads, surely it's a MUST to test it and let the > data speak? No? or would you redesign the ads, see revenue go down and not > change your mind? > Peter <snip> ________________________________________________________________ Welcome to the Interaction Design Association (IxDA)! To post to this list ....... [email protected] Unsubscribe ................ http://www.ixda.org/unsubscribe List Guidelines ............ http://www.ixda.org/guidelines List Help .................. http://www.ixda.org/help
