Site Redesigns

Has anyone tried to measure the results of a redesign?

[mrsadm]

We've had a good client for several years who makes a good living selling products from her site. She has a business selling these products outside of the web so I think she knows her customer audience.

We did the original site design and over the years have received many emails about how much users like it. Now the client wants a redesign. The problem is her idea of design is (in our opinion) HORRIBLE. For instance, she wants very strong verticle stripes in the background of the pages, and at least 5 different fonts on the main page. Some of these fonts are really "goofy" and I do not think project a serious business. We proposed a modification that simplified and softened the look but she did not like it.

I'm sure many of you have had this problem, but my idea as to how to handle it is to compare the product sales levels before and after we put her design online to see if it has a negative effect. If the sales are about the same, then maybe we are wrong about what her customers like. If not, we would have some evidence to back up our opinion of her design.

Has anyone tried anything like this?
Thanks
Linda

[korkus2000]

Go with what she wants but have a fall back position. The customer is always right even when they are contradicting themselves. I have had to put a few horrendious sites on the web. The customer loved them. You are trying to make them happy nothing else.

[jackofalltrades]

Im sure it would be easy enough to guage the impact of a new design from stats and sales figures.

Also perhaps including a feedback form for users to comment on the new design may gave you (or her!) more ammo to use for your respective cases.

You should make it clear to her that any major site redesign will have a long term impact even if you change it back a week or a month later.

All your visitors in that time wont come back if the site sucks!

If shes adamant that the users will love her design and you are 100% sure they wont, then use a technical excuse as to why it cant be done.

"The search engines really dont like stripes in the background..." ;)

JOAT

(ps thats a bad excuse, make up a better one - use design concerns, usability or something like that!)