18 replies to this topic

[Randy]

A bit more sophisticated approach to delivering a correct breadcrumb, etc is to use visitor history to build the breadcrumb trail, which could also be used to customize other elements of the page of course.

It's a hack, but the way I've seen this done is to uses javascript to set a cookie via onClick events. The JS cookie records the Category someone is in, assuming they're in one, and can be populated by server side scripting. Then when a product page loads the first thing it looks for is the cookie. If it's set use the info there to set the breadcrumb and anything else you may want be changeable, so that people get a way back to where they were before. If not, give them the default information from the page they actually landed on.

FWIW, the reason for using a cookie via onClick instead of just checking the referrer info from the browser is because mainly because of Norton's security suite products, which block referrer data. If you wanted something close but not quite as effective you could certainly do it that way. This would however take an extra database hit most likely.

[BBCoach]

A bit more sophisticated approach to delivering a correct breadcrumb, etc is to use visitor history to build the breadcrumb trail, which could also be used to customize other elements of the page of course.

Yuk! Why not have the dB schema designed so that any breadcrumbs generated would be by design instead of tracking user clicks?

Did my design suggestion help you qwerty?

[qwerty]

It probably would, but as you wrote,

BTW, this system is of course custom made and no CMS package out there will do what I'm talking about.

The client isn't willing to spend the money on a new CMS, so it looks like we're going to try some A/B testing. We'll pick a product and set up redirects from all but one of its pages to the remaining page, keep that up for a month or two, then drop the redirects for a couple of months and see if there's a significant difference in performance.

[BBCoach]

The client isn't willing to spend the money on a new CMS

Totally understand. It took us 6+ months of coding before we finished. The key is to build this from the ground up. I don't see how one could "wire" it into an existing system. Many days of brainstorming over what was essential SEO and how to best implement without losing thousands of established (legacy) inbound links before writing one line of code.

Anyways, good luck with your A/B testing.