g1smd
msg:3920516 | 7:01 pm on May 26, 2009 (gmt 0) |
Is the content much the same (or even identical)? Did you move to new URLs or keep the old ones? Are there any particular pages where the bounce rate has dramatically increased, or fallen?
|
shallow
msg:3920540 | 7:59 pm on May 26, 2009 (gmt 0) |
The content is the same and a lot of it is identical). To the extent possible, the old URLs we kept and the developer implemented hundreds of redirects. As to any particular pages where the bounce rate has dramatically increased or fallen, I haven't examined the stats. I really don't look at analytics a lot but I'll take a look at the page stats.
|
jkovar
msg:3922800 | 4:16 am on May 30, 2009 (gmt 0) |
The first thing that comes to mind for me, is perhaps the old layout took into consideration things like scripts loading last, giving images widths/heights and other HTML related things that can affect how fast a _readable_ section of the page is displayed. If the new layout doesn't take these things into consideration, new visitors who don't have elements such as CSS stylesheets or JS cached may be experiencing enough of a load delay to drive them away before the page becomes usable for them.
|
ken_b
msg:3922803 | 4:28 am on May 30, 2009 (gmt 0) |
I've been there. I made a change last fall that cost me 25% of my page views literally over night. I thought folks might get accustomed to the change after a wwhile and my pageviews would gradually increase again. Nope. After about 4 - 5 months I retreated to the old design. Page views went back up to the previous level over night. Some changes just don't work. Have you gotten any feedback from your visitors, emai, etc, about the change?
|
martinibuster
msg:3922845 | 7:04 am on May 30, 2009 (gmt 0) |
Are you using an XHTML Doctype? Did the site use one before the redesign?
|
shallow
msg:3923054 | 2:29 pm on May 30, 2009 (gmt 0) |
martinibuster, yes I am now using an xhtml doctype, and was not before. I'm very interested to know why you ask. Is one doctype better than another? Thank you everyone for your replies.
|
martinibuster
msg:3923096 | 4:03 pm on May 30, 2009 (gmt 0) |
Many layouts using xhtml (often because it's based off a WP template) display badly on IE 6 and IE 7. Visitors to your site may only be seeing the left and right side nav/gutters while your conent is pushed down a thousand pixels below the fold. That's what the most mangled xhtml layout will look like. While IE 6 and 7 typically accounts for only 15 - 20% of site visitors, that may be enough to be tipping your bounce away rates in the negative direction. Just a theory. If this is indeed the problem then you may want to preview it in those browsers to see if that's the problem. If it is, then you can either tweak the doctype to HTML 4.01 loose or else tweak the code.
|
Robert Charlton
msg:3924369 | 4:22 am on Jun 2, 2009 (gmt 0) |
Re XHTML doctype, you might want to check out this thread... Why most of us should NOT use XHTML
[webmasterworld.com...] ...and re other doctypes... FAQ: Choosing the best doctype for your site
[webmasterworld.com...]
|
simonuk
msg:3924662 | 10:58 am on Jun 2, 2009 (gmt 0) |
The smallest of changes can often radically alter user interactions. I spend a lot of time dealing with the usability of web site designs and I can't emphasize enough how critical A/B testing is. I have seen websites that look like a 4 year old built it have conversions 80% higher than the professional built sites which replaced it. I have seen changing the colour of a advert in the content area from red to green radically altered the way the visitor used the top navigation. The correlation between the two should have made no difference but it radically altered it. Anything and everything is possible and unless you've spent years researching usability then you should be A/B testing even the smallest change because even the smallest little change can and often does have a huge impact on the site. Google's Website Optimizer is a good place to start.
|
shallow
msg:3924879 | 4:35 pm on Jun 2, 2009 (gmt 0) |
Thank you! I'm going to try some color changes and discuss what I've learned with my web developer.
|
cgrantski
msg:3925590 | 12:33 pm on Jun 3, 2009 (gmt 0) |
Apply 4Q to your site and ask people directly. You can get to the root of it far faster than with endless inferring from their footprints.
|
jackgordon
msg:3929765 | 5:57 pm on Jun 9, 2009 (gmt 0) |
Forgive me if I am getting the definition of "bounce rate" incorrect, but because of your website has recently been redesigned, could it be that existing customers/ visitors do not recognise it so they decide to leave?
|
|
