homepage Welcome to WebmasterWorld Guest from
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Visit PubCon.com
Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

Long slow slide for our primary search term

 7:14 pm on Sep 9, 2007 (gmt 0)

I run an information site (with a bit of ecommerce thrown in) on the theme of, let's say, "widgeting." I have several hundred pages, regularly add material, and have over 7,000 inbound links (according to Yahoo). We're completely whitehat and I don't do much SEO beyond making sure that "widgeting" is in titles, etc.

My site, until around March/April, ranked between 5th and 9th for "widgeting" on Google, but usually 5th or 6th. This was always lower than I'd hoped, but the competition is stiff and because we were still on the first page I got about 200 visitors a day just from the word "widgeting" plus many more for the long tail. Now all of our Google traffic is long tail. There's no body.

In March I changed the site over to Wordpress and put in 301 redirects from all the old locations to the new ones. Wordpress has been great because it makes it to easy to add new material (I'm talking maybe a dozen new articles a month, plus links to related news stories). After the change we slid to 8th/9th, which was okay. Then three months later we were 11th, then 15th/16th, and now six months later we're 19th. I'm seriously thinking that we'll slide onto the third page or lower before long.

Some of the slide can be put down to Wikipedia gravitating to the top of the results (they have two results showing). But some of the pages above mine are just one-page wonders and at least one is very spammy, with a drop-down list that seems to be there only to pack more keywords into a smaller space, incoming links from unrelated pages (hidden links tied to a stats counter), and a blaring audio message.

I keep thinking that I must be doing something wrong without intending to. I'd welcome any pointers or suggestions.



 7:38 pm on Sep 9, 2007 (gmt 0)

Can't say for sure without looking at how you've got WP set up, but the first thing I'd check is how many URLs you've got pointing to the same content. Calendar? Archives? Most Popular posts? Make sure you give Google only one path (URL) into your content.


 8:06 pm on Sep 9, 2007 (gmt 0)

This thread may help: WordPress And Google: Avoiding Duplicate Content Issue [webmasterworld.com]

It's easy to find, since its in the Hot Topics [webmasterworld.com], which is always pinned to the top of this forum's index page.


 11:47 pm on Sep 9, 2007 (gmt 0)

Thanks. The vast majority of the site uses Wordpress Pages rather than posts, so duplicate content isn't a problem there.

With regard to Posts, there is no calendar and the archive pages and posts listed by author pages just contain a brief excerpt, so I'm guessing that this isn't a duplicate content issue.

I also have my htaccess file set up to redirect all non-www queries to www. The code is:

RewriteEngine On
RewriteCond %{HTTP_HOST} ^example.org
RewriteRule (.*) http://www.example.org/$1 [R=301,L]

I got this from the web and it seems to work, although I've no idea if it's properly done.

Any other thoughts?


 11:49 pm on Sep 9, 2007 (gmt 0)

Also, I have a "print this page" link on every page but the home page, and it has rel="nofollow" in it. I've been pretty careful to avoid duplicate content.


 12:35 am on Sep 10, 2007 (gmt 0)

Then I'd take a look at : "widgeting" is in titles,

Just how often does 'widgeting' appear? Is it repeated in the titles? Are there any synonyms that could be used, both for titles and for link text?


 1:58 am on Sep 10, 2007 (gmt 0)


On the homepage I have: "Mydomain Curly Widgeting: Learn Widgeting Online". On internal pages I have "Mydomain Blue Widgeting: Post Title" (where the post title will sometimes, but most often not, contain the word "widgeting").

I generally try to keep the "Mydomain Curly Widgeting" prominent in order to brand the site. It's effectively the site name or brand name of the project.

That's the same format I used previously, before shifting to Wordpress. Of course it's possible that Google no longer likes that format.

Of course the site that's currently ranking #4 for "widgeting" (15 places above me) has the following as its home page title:

"Widgeting, Home, what is widgeting, types of widgeting, widgeting and green doodads, widgeting as a complementary doodad, fluffy doodad, mind-body doodads, domain.com"

Seems spammy to me, but Google seems to like it.

[edited by: Haecceity at 1:59 am (utc) on Sep. 10, 2007]


 4:46 am on Sep 10, 2007 (gmt 0)


I am experiencing more or less the same with you - Google SE ranking sliding down slowly. The only different is I do buy/exchange links from related websites and submit articles to popular directories.

From what I see, the problem is not about what we did wrong - it's what we didn't do and it's what our competitors do.

We just got 'displaced' by the new websites, gradually.

Try have a look on the new websites that gained rank recently in your industry, do you observe:
1. Old expired domain being reused?
2. Website with lots of purchase quality links?
3. Incoming links with exactly the same 'keyphrase' link?

From what you wrote earlier:

But some of the pages above mine are just one-page wonders and at least one is very spammy, with a drop-down list that seems to be there only to pack more keywords into a smaller space, incoming links from unrelated pages (hidden links tied to a stats counter), and a blaring audio message.

I guess your answers will be a big YES - which is the same with my case.

There are PLENTY of these websites ranking above me nowadays - displacing me from the top and taking up all the traffics. Frankly, I think they deserve the traffics. afterall, they are the ones who are willing to take risks to run the biz on tight margin and being filtered from Google. We never knew when will Google turn the knob and filter off all the websites with massive paid links.

The question now is - should we follow the trend and go out hunting for paid links? or should we just pray for google to ban those with paid links?


 6:30 am on Sep 10, 2007 (gmt 0)

Hi tahiti

I'm sorry to hear that you have similar problems.

I've never bought a link, and I don't even do reciprocal links. I link to a site only if I think it offers something relevant and of value to my readers. My inbound links are organic -- I haven't requested a link in three or four years. People just keep linking to my site, although that's slowed down recently because of my lower profile.

I'd include "it's what we didn't do and it's what our competitors do" as "doing something wrong" as long as the things the competitors are doing are ethical! I'm not interested in going for short-term benefits through questionable means.

I'm actually comforted by the fact that it doesn't seem as if the things you are talking about are going on in my field, except with one site.

Try have a look on the new websites that gained rank recently in your industry, do you observe:
1. Old expired domain being reused?
2. Website with lots of purchase quality links?
3. Incoming links with exactly the same 'keyphrase' link?

1. Well, the spammy site I mentioned is newly registered back in January '07, so that doesn't apply there. The other sites that have crept up over mine are generally longstanding.

2. With the spammy site there's something *really* fishy going on. The parent company has released a free "hit counter" which people have been adding to their sites. What they perhaps haven't realized when adding this counter is that it contains a hidden link to the spammy site. The hit counter coincidentally shares a name with the parent company of the spammy site, as shown in the whois record. Neat trick. It shows how easy Google is to fool.

I've reported this to Google but I'm not holding my breath.

With the other sites it certainly isn't the case that they have lots of identical inbound links. I'm in a pretty ethical field and the spammy site I've been talking about is a rogue operator.

What is notable is that several of the sites above me have just one page on the topic in question (while I have several hundred pages). One result contains precisely one paragraph on the topic (although it's not really on-topic) -- apart from (non-relevant) navigations links there really is nothing on the page but one paragraph (which doesn't mention the keyword) and the word "widgetings" in the page title and "widgeting" in two links. Makes me sick. I've put many hundreds if not thousands of hours into creating quality information on "widgeting."

3. It's certainly the case with the spammy site / hit counter scam that the links all have identical link text.

Anyway, apart from that one spammy site it seems that the other results are clean, as far as I can see. Most of them haven't changed in years. Maybe that's my problem.


 8:52 am on Sep 10, 2007 (gmt 0)

Where do you feature on an allinanchor:widgeting search and how long has the ecommerce featured on the site?


 11:11 am on Sep 10, 2007 (gmt 0)

Hi Glengara.

Interesting question. Today I'm 17th for allinanchor:widgeting and 18th for simply "widgeting". The two sets of results are identical down to #13.

I'd be interested to hear what conclusions you'd draw from that. I've never ben clear about the allinanchor function.

The ecommerce has been there for many years. The site's been running since 2000 and I think it was the following year that we started selling some stuff.

We started carrying adsense in November last year.


 11:47 am on Sep 10, 2007 (gmt 0)

Small discrepancies between "normal" and allinanchor results are par for the course, wide ones can indicate a "problem" but in your case there doesn't appear to be one.


 1:00 pm on Sep 10, 2007 (gmt 0)


Well, two random questions from me as well...

First one isn't that important but... in WordPress while the Title is usually generated quite well by default, the description ( META ) is sometimes either repetitive throughout the pages, duplicate ( the template and perhaps an SEO plugin both adding their own ), or sometimes of course they're non-existent.

Second is, with WordPress, especially if you let it send the word out whenever you add an article and have the RSS feed linked from your homepage, there's a huge chance that your content is being syndicated by someone who really shouldn't be syndicating it.

I'd say that they're probably a higher authority on the topic, or more trusted than you, but in fact just yesterday I saw a blog I look after... being outranked for its own main keyphrase by a spam site that included the page in an iframe with its title and description copied onto the source page for good measure.


Does either apply?
I mean any of the META description problems, or being syndicated / spamdicated by one site too many.


 1:07 pm on Sep 10, 2007 (gmt 0)

I really appreciate these thoughts. Keep 'em coming!

And I've learned something about the allinanchor query.

My incoming anchor text varies hugely. Most of the sites that have been above mine for a long time have "widgeting" in the domain name like "widgetingcenter.com" and "learningwidgeting.com") which naturally gives them an advantage that I don't in any way resent.

But these new ones don't have widgeting in the domain name.

My domain name, although evocative, has about as much inherent relation to "widgeting" as the word "Amazon" has to "books."

That's why I've been keen on branding the site as "Mydomain Curly Widgeting." Some people (quite a lot, actually) link to me with that full phrase, although as I said there's a lot of variety, as you might expect.

Interestingly, I just noticed that all but three of the sites above mine have widgeting in the full URL. Almost all of those at the top of the second page have some variation of widgeting.html as the file name. These are, by and large, the one page wonders. That seems significant, as if somehow the algorithm is now favoring sites in the following order:

(1) those that have widgeting in the URL and then
(2) those that have one page devoted to widgeting if that page has "widgeting" in the file name, and then
(3) those (like my site) that only have "widgeting" in the page title, text, incoming anchors, etc.

In other words the ranking or sites is now more focused on the search term being in the domain and (secondarily) the file name.

There are a couple of exceptions to this in the 17 sites above mine, but they mainly seem to follow that pattern. With at least one of the sites that doesn't fit into the pattern it's easy to see why they have good ranking: they're called "Acme Widgeting" (where that's pretty much a household brand name) and the site is able to simply have their initials as domain name. With the profile they have you'd expect them to rank highly.

And there are two Wikipedia results, with massive pagerank.

Another that ranks well on the first page but doesn't have widgeting in the domain has it twice elsewhere in the URL and has the spammy title that I describe in the post just above tahiti's first post above. I guess Google just can't resist the spam.

Just a hunch.

Does my core hypothesis have any relevance to what you're seeing, tahiti, or to anyone else?


 1:51 pm on Sep 10, 2007 (gmt 0)

On the domain/file name thing....in the past G has experimented with what to include in their allinanchor calculations, I know for a fact file names were one and suspect domain names were too, and while they don't show up anymore that's not to say they've lost all "value"...


 3:14 pm on Sep 10, 2007 (gmt 0)

Hi Miamics,

Good thought -- things I hadn't really considered.

With meta tags I've always been a slacker. Before the update I basically had none except on my home page. I now (on Wordpress) have a meta tag plugin that allows me to have unique description and keywords for each Page/Post, but I'm embarrassed to discover I haven't made much use of it. I'll get onto that immediately.

Syndication: I did a search for unique phrases from recent posts and found no evidence of wholesale pillaging, fortunately. I'll certainly keep an eye open for that in the future.

Even though I never had much in the way of meta descriptions it's quite possible that that's the problem.

But any other suggestions will be seriously considered and greatly appreciated.


 2:59 pm on Sep 23, 2007 (gmt 0)

I'm now pretty sure that it was in fact a duplicate content issue. I noticed on Google Analytics that example.com/page/ was being treated as a different page from example.com/page, and I've installed a plugin that redirects to the page without the trailing slash.

That was a few days ago and the site's a little higher than it was. Does anyone know how long it takes for sites to recover position after resolving duplicate content issues?


 5:54 pm on Sep 23, 2007 (gmt 0)

1. Old expired domain being reused?

I have noticed this also. In my industry, old domains that may have once been used for political purposes, educational purposes, etc, have been bought and turned into blogs completely unrelated to their original use, ranking very well. Inbound links rarely use any anchor text that would support these rankings. Seem to be ranking based on established PR, trust, etc.


 5:36 am on Sep 24, 2007 (gmt 0)

Hello Haeccity. I too own a WP based site and had these problems about a year back, which I have fixed.

There are other posts here which talk about this alot, but here is a short checklist of what you need to make sure you have in place:

1. Use the meta desc plugin and double check that every live page has only one unique meta d take from the first 152 characters from the page content or hand written. Double check that there is in fact only one meta description by viewing the source code of live pages.

2. Make sure that ayn tagline on the pages LAST not first eg "specialized widgets for the common person - My Curly Widgets" Where possible, stay away from taglines altogether.

3. Pages are NOT posts. Posts are based on one set of data from a row, where pages can inlclude mroe than one post and can be paginated. I chose to stop any pages from being indexed and focused entirely on posts only.

4. Index.php can be a real pain in the butt in WP isntallations and spidering, and so can non ww urls, non slash ended pages '/' Makes sure you cover all bases and test using header checker:

Index.php root 301 redirects to blog home.
Non www urls 301 to their www counterparts.
Pages that have no extension ending with no slash are rewritten to the 'slash' version if they are landing pages or folders.


 5:56 am on Sep 24, 2007 (gmt 0)

maybe you need to embark on an aggressive link-building campaign.

Global Options:
 top home search open messages active posts  

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved