Hi BFA team,

I am not sure if this is the right board for this question. Please forgive me if that should be posted somewhere else.

I experienced a technical problem when I started to use a tool from ranks.nl. With that tool you can analyze the keyword density of a web site. The result with web sites using the Atahualpa theme is completely wrong, it shows almost only HTML and CSS tags as keywords. If I analyze other Wordpress sites that use other themes, this problem does not occur.

The tool can be found at http://www.ranks.nl/tools/spider.html

I analyzed my blog at http://www.disturbedcooking.com and another Atahualpa site at http://www.ein-jahr-hongkong.de in order to check if it's a specific problem of my site. The result is wrong in both cases.

I compared it analyzing the blog of a friend at http://www.blackforestbbq.tv who uses also Wordpress but a different template. In this analysis the results looks correct.

Could you please have a look at this and maybe find out what could be the reason for this result? Thanks.

Sven

When I did an analysis of a couple of my sites (using ATA 3.5.3) they all came back correct (keywords being examined instead of CSS elements), when I analyzed your site using the tool, I see what you mean by the CSS being interpreted as keywords. Yikes!

I don't have any sites running any of the ATA 3.6.x version of the theme, so I can't say if it's an ATA version issue or possibly something particular on your site.

Best of luck in tracking it down!!

Edit: Forgot that I do have a test blog using 3.6.4 and when I analyzed that site (there are no plugins and I haven't modified anything) it comes back UGLY. A ton of HTML and CSS elements are thought (by that tool) to be the keywords. Not sure what's going on but the tool seems to not like the 3.6.4 version of the ATA Theme.

Thanks for the feedback but actually both blogs use ATA version 3.5.3 and not 3.6.x. So it seems that it is not a version issue?

Sven

Is your CSS external or internal?

You mean in the ATA settings? CSS and JavaScript are both set to "inline". Could that be the problem?

Sven

I just ran the test with css/js inline then external - gee guess what, there is a big difference.

My guess is this just reads all the counts all the individual words on the page and ranks them. It doesn't skip things in the <style> or probably any other HTML Scripts etc.

I wouldn't put too much faith in this 'tool'