SEO tool shows %55 similarity between 2 pages. Is this a problem?

[talya]

Recently, my site took a hit for a term we just began optimizing for. We were on the 4th page in google SERPs(not so great but not so horrible for such a competitive term, especially after 1 week), but today, we are not even listed for that term.

I was wondering perhaps that the reason we took such a hit all of a sudden and had our homepage disappear completely in the SERPs for that term was due to google thinking we have duplicate content. We have a similar category URL that DID show up in the results, though much further down on page 10.

Basically, our domain name includes the keyword, and we changed around the homepage to optimize for that keyword even though we have a category page optimizing for similar keywords. The URL for the cat page also includes the keyword and is focused solely on those items, whereas the homepage also has many links to subcats for different products, though they are further down the page, and the title tag H1 etc are all optimized for the term we want to rank for.

I used a similar page checker tool and compared the homepage to the cat page and it said it was about %55 similar.

Do you think this could be a reason for the homepage to be dropped for that term? Could it be that google sees the pages as so similar that it has to choose one page as more relevant than the other and only list one?

Any suggestions?

[tedster]

not so horrible for such a competitive term, especially after 1 week

This sounds to me like the common cycle early in a new website's experience. The early rankings are a kind of gift, test, or experiment (take your pick on terminology) and then they go away. Long term rankings then begin to establish themselves over time.

Related threads:
Good Early Rankings - but they disappear [webmasterworld.com]
Filters exist - the Sandbox doesn't. How to build Trust. [webmasterworld.com]

Those two threads and more can be found in the Hot Topics area [webmasterworld.com], which is always pinned to the top of this forum's index page. There's also a currently active thread that touches on the same topic: week old website ranked #14 in google for main keyword [webmasterworld.com]

So no, I don't think you're working against a duplicate filter right now - you're just building a history of trust with Google and going through the kinds of treatment that new sites often see.

[talya]

It's interesting though that my site is actually not new at all and has built trust over several years. Depending on the season and what we are trying to sell, we change the title tags around a bit and the content to optimize for our main keywords.

So that's why I'm so perplexed... If I switch title tags and make some changes to a page, does the page all of a sudden lose value with google? I didn't think so. That's why I'm suspicious of the similar page problem.

[Marcia]

You might want to check out where on the pages the similarities seem to exist. Are they at the very top of the page, the first thing the crawler fetches? in the footer? in the navigation? in the main body text? If in the main body text, at the beginning, middle or very last of that section?

The general percentage I've usually seen mention is either 70% or 80% (don't recall exactly which), but looking at the concept of "footprints" on pages for identifying near-duplicate documents, the location and format (whether regular text or global navigation) could well be factors to consider.

Some plug-ins for FireFox are handy for checking that out, as is looking at the text-only Google cache for the pages.

[talya]

I looked at the text version in google's cache, and the similarities are mostly having to do with the global nav, although there are huge similarities in the title tag. Also the H1 header is the same. Everything towards the top is pretty much identical (again, due primarily to the global nav), and only when we get towards the bottom do things start changing with actual product differences.