< This post was moved from another location>
Unique article has gone supplemental
I have just noticed that one of my articles has gone supplemental (just this single article).
I have some other pages that has gone supplemental, but I never have had one of my articles going into the supplemental index before.
The strange thing is that Google shows an ancient version of the article (cache date: august 26 2006) in the SERPs, and seem to have forgotten all about the new version -- the article has been indexed several times since august 2006.
The articles content is original and unique and the article has a unique title, a unique META-description and so on...
On november 12 2006 I published a new article (lets call it article B) and moved a small section from the now supplemental article (lets call it article A) to article B and removed this section from article A.
The cached version from august 26 2006 has the mentioned section in it's content.
Could this have caused article A to go supplemental?
[edited by: tedster at 7:19 pm (utc) on Jan. 1, 2007]
Thanks for the precision and detail of your report, OutdoorMan. That kind of thing really helps move us toward better understanding.
Roughly how big a section did you move from the old article to the new one?
|Roughly how big a section did you move from the old article to the new one? |
It's difficult to be accurate on that one (without going on counting letters and so), but I roughly estimate that I moved between 5% and 8% of article A to article B (the whole section).
I developed that particular section into a brand new article -- the reason why I moved it from A to B.
... Maybe I should add that the file size of the document is about 55 kb (Google estimation).
To elaborate on my earlier post for those that had questions:
There was a fix available on the IIs server. It needed isapi rewrite installed and the host wasn't really willing to do it. Isapi rewrite prevents dupe content by eliminating such things as accessing a url many different ways. ie:
www.example.com and www.Example.com or www.examPle.com. All of which makes dupe content issues. Not even mentioning, www.example.com/Widget and www.example.com/wIdget. I finally understood there were a thousand different paths to one url.
I had actually thought a 301 redirect solved all all my dupe content issues. Not so. G1smd set me straight on that. Apache servers do NOT have those issues.
After being moved to the apache server I discovered they do NOT have mod rewrite installed so doing a 301 redirect in my .htaccess was useless. Ergo, the base href installed on every page. It seems to work. Pages coming out of supplemental finally. I also use nothing but absolute linking in the site and have for a long time.
Is having relevant links a problem? Could it put your pages in supplemental?
Outdoorman - When you say "supplemental" are you seeing this status on search results, on the site:tool or both. And is that page ranking?
The reason i ask, is that the supplemental status and result can vary according to the query method.
Just for the record we have [ 3 out of 4 ] sites that are "non supplemental" with both the site:tool and in the regular results. Caches are all up to date. These sites rank well and have:
- 4 sites with same layout design
-Unique content between pages ie each page is different [ meta title, description, body content]
-keyword density on any term is very low
-clear navigation linking theme pages to themed pages
The 4th site is the same as above, but does show a lot of "supplemental results" which it should not as it is exactly the same as the former three, except in a different language.
Then we have other sites [ 3 ] which have an inconsistant "supplemental status", showing old cache dates in the results as well as new ones. How Google has selected one over the other has got me beat. However, these sites may have imperfections, with :
- 3 sites with different template designs from 1st set
-common blocks or phrases of content [ e.g. 20 -30 characters the best deal in <tagged> ]
-keyword/phrase density is thus higher
-greater elements of boilerplate content [ than above ] - maybe 5-6 lines
-confusing navigation so that themes might be difficult for Google to interpret.**
** interstingly Google is sometimes throwing up irrelevant results on the latter in the serps
*** these sites provide more consistant results where the navigation into these pages is less confusing, although i emphasise "more" not always.
Hopefully, this may lend some clues as to what's going on.
As mentioned elsewhere, we've converted one of the sites to a similar format as the successful sites, to see if it makes a difference. This was done about 1 week ago, so we'll wait to see if there are any changes.
[edited by: Whitey at 10:36 pm (utc) on Jan. 1, 2007]
|Outdoorman - When you say "supplemental" are you seeing this status on search results, on the site:tool or both. And is that page ranking? |
The page is supplemental when I use the site:operator.
And the page is not ranking anymore when I search for the page title -- before the page had a #4 position for the page title. Now it doesn't even show up at all.
Edit cause: bad explanation in first line.
[edited by: OutdoorMan at 10:36 pm (utc) on Jan. 1, 2007]
The title tag is important. One of my sites has pages that are supplementals purely based on the title tag (for a short time I was generating the same title tag for a lot of pages due to a mistake). The pages with different title tags are ok.
|The title tag is important. One of my sites has pages that are supplementals purely based on the title tag (for a short time I was generating the same title tag for a lot of pages due to a mistake). The pages with different title tags are ok. |
My titles are unique. Article A has the same title as it allways have had *. Article B has a different and also unique title.
So the supplemental issue that I'm experiencing is not caused by identical titles.
Edit: * except that I have changed the title from: "example.com ¦ My page title" to "My page title" in september 2006 -- and the cached version is currently showing the old "september title".
[edited by: OutdoorMan at 10:46 pm (utc) on Jan. 1, 2007]
I don't mean to challenge the solution you've implemented, but I work only with microsoft servers an i know that filepaths are not case sensitive,
conversely, from my very limited knowledge of the unix o/s family, files paths could be case sensitive,
However, its very complex, well you're on apache now, but if you want to go on an IIS machine again
have a look at this below
P.S Mods please note the link is to a microsoft developers network blog, if its not okay please delete, cheers
It's really hard to find *nix/Apache hosting that *doesn't* have .htaccess and mod_rewrite enabled, even the cheapest. I've never heard of it, except for one company 7 years ago.
No matter what server your site is on, search engines WILL treat the file path in a case sensitive manner -- that is the way the web works. So among other issues that Windows server people need to watch out for, case sensitivity is definitely on the list, or else you may see your Page Rank and backlink influence split into several little piles when you really have earned one bigger pile.
Microsoft made a BIG error in making an "Internet Information Services" server that does not adhere to the actual technical conventions of the Internet. Many websites are suffering low-level troubles, and sometimes worse, from this bad decision made many years ago.
The best practice in my view is not to indulge in any capitalization games at all -- keep everything lower case.
I was not aware that ISAPIrewrite could eliminate case sensitivity issues on a Windows server. Does it do this one url at a time, or can a global rule be set up?
>>>It's really hard to find *nix/Apache hosting that *doesn't* have .htaccess and mod_rewrite enabled, even the cheapest. I've never heard of it, except for one company 7 years ago.<<<<<
Yep. Shocked me too. Same host. The site owner likes these guys so I am stuck working with them. Used to they had an owner/engineer that was top notch. Knew everything. He sold out last year to some guys that know nothing. I had to send them instructions on how to do the 301 redirect. lol.
However, if you are on virtual hosting...almost no one has rights to the .htaccess file. I have clients on three different hosts running on apache and none give me access to the .htaccess file. If you know of a good reasonably priced host that does...shoot me a note.
>>>I was not aware that ISAPIrewrite could eliminate case sensitivity issues on a Windows server. Does it do this one url at a time, or can a global rule be set up? <<<
I can't answer that Tedster. Perhaps g1smd can fill us in when he drops back by. That was the advice I was given in another thread on here.
Has anyone ever noted that pages within the subdomain are typically more likely to be labled supplemental as opposed to those in the root domain even though both pages may have equal PR, inbound links, content etc..etc..?
Just wondering why this occurs.
For the same page we have:
1. site:oursite.com/ ( see page amongst results for Widget 1 )
"supplemental" and cache date of 26 December 2006
2.site:oursite.com/widget1/ ( see site:tool search on specific URL )
"non supplemental" and cache date of 29 December 2006
These pages show in the serps for general search under unique keywords on this page
Is anyone else seeing this type of behaviour?
[edited by: Whitey at 5:34 am (utc) on Jan. 2, 2007]
Just an observation that may be useful... while watching Google in serious flux tonight with my sites ranking for it's main keywords moving all over the place, I decided to try a long-tail search. For a 3 word phrase, my site comes up number 2 for a page that is marked supplemental, and BELOW it in the serps are less relevant results that are NOT marked supplemental.
And on a side note: On my main page, about a month ago I had no description tag for a short time. During that time aparently G indexed me, because right now this page shows up in G with a description taken from DMOZ. Oddly, for the search "blue widgets" I see the main page with the current title and description (from the description tag), but a search for "green widgets" brings up the same page with the old DMOZ description. Weird!
< This post was moved from another location >
Recently I have posted about my sites coming out of what has been a 2.5 year supplemental nightmare. More recently, I have noted that now, after about 1 month of seeing no sup results, there are a few creeping back in, especially on my larger sites.
The thing about it is; the results using the site: command do have sup results in, yet when I search for unique text on these sup pages they are in the results and not labeled as supplementary.
There is a heck of an amount of heavy crawling by googlebot going on for my sites atm and this has not been seen for a long long time. Perhaps this to is related to the "new" - for me its new - non sup results too.
Has anyone ever come across anything like this before?
Is googlebot doing a heavy crawl elsewhere atm?
[edited by: tedster at 4:21 pm (utc) on Jan. 2, 2007]
|when I search for unique text on these sup pages they are in the results and not labeled as supplementary. |
There have recently been several mentions about this. Have you noticed whether the cache dates vary between the two situations?
25 Dec 2006 02:27:28 GMT on both caches O_o
As well as all of the others I have looked at for my own site, there seems no differance in the time it is the same cache.
Thanks. That's the first time someone has authenticated this "new" behavior -- that a search result in regular search may not show the Supplemental tag even though the same url and cache date is tagged as Supplemental in the site: operator results.
I have another thing to hold on to... I have sites in spanish, and when I click "seach pages in spanish" the Supplemental Result doesnīt show. I have unique content sites, being inside the Supplemental bag.
>>> same cache dates for supplemental and non supplemental
I've been seeing that for a while now.
> For a 3 word phrase, my site comes up number 2 for a page that is marked supplemental, and BELOW it in the serps are less relevant results that are NOT marked supplemental.<
I donít think a lot of people are seeing this. Scrapers are learning to back door their way into the index by duplicating more obscure pages of peopleís sites. Youíll also see this as more pages of legitimate sites get tossed into supplemental results. Iím also seeing article sites where the entire content was once supplemental now without any supps. Almost every page in these sites is duplicate content.
It almost looks like Google is freeing up space for other projects. Anyway they started turning up the notch on something two months ago and its gone too far. Itís really ridiculous. Iím getting a variety of old cache dates dating back to September when on Dec 20th everything was up to date on my site.
Sorry if this throws a spanner in the works, but we had already started research on relative/absolute linking a year ago to check this out. We made several of our sites use relative, some with relative and base tag and some with absolute. We also split those groups into some with google sitemaps and some without. Obviously its our time consuming research so I'm not going to go into great detail but the main conclusion was that it doesn't make a difference. It made a different in correctness of crawling with some SEs but not for anything else like ranking, speed of crawling, supplementals or anything like that.
I just thought I'd throw in that alternative opinion because it seems like alot of people have just stated that "changing to relative+base means you get out of supplemental index!" Based on 3 or 4 pages of their site coming out once the change was made. Maybe it was the fresh "created on" date that did it? There are a thousand alternative explanations to why those pages came back after the change.
Hi Internet heaven, is this particular site you're talking about an ecommerce site, listing products for sale, possibly product not manufactured by you?
Are the supplemental pages product description pages, allbeit with unique descriptions by your business?
If the 2 above are true your site's supplemental issues, nothwithstanding its evident good quality, seems to fit what i percieve to be a growing trend in google implementation of its supplemental index
I'll be more specific:
one the regular search (google.com) the results are totally normal.
No supplemental issues, no drop no nothing (in fact we see increase in traffic and ranking).
On the other hand, in the internal site search powered by Google or by site:mydomain.com keyword
the returned results are all and only supplemental -the returned supplemental pages are well deserved to be supplemental (old forum posts, dup issues long being fixed by 301 redirect)
the "real pages" do not show AT ALL!
another thing worth mentioning: there's a dash sigh (-) in my domain name.
I just started seeing that as well. One of my sites shows all but four urls as supplemental when I do a site: command, but my logs show plenty of traffic coming in from these supposedly supplemental urls - and if I track back in Google, they DON'T show as supplemental (when not generated by the site command) If anything, traffic is up. So I'm completely kerfloozled at the moment.
Here is what I see day in day out, doesn't matter about unique content or any other gibberish, it's all about external links to pages, get em, you will stay out of supps, don't and you eventually will end up in the supps.
Linkbacks that generate Page Rank is what Google was born with and it still is the underlying factor that drives all Serps in Google.
What I've seen on my site is a PR5 on home page, lots of PR4 pages, then it seems to skip PR3 and jump to PR2 for a lot of pages. Pages that have been in supplemental due to no meta description tags (or duplicate tags) have PR0 (white bar) but are in most cases coming out of supplemental.
1. What does it mean when a page has the meta description updated to be unique, then gets cached with the new description, but stays supplemental with a PR0? Does it take time to drop the supplemental status after the page has been fixed?
2. When looking through pages indexed in Webmaster Tools, I have a couple hundred pages listed that are not supplemental, and they do seem to be in some order of relevance with my home page at #1, but they are more random after the first 10-20 listings or so. Then I have a bunch of supplemental pages listed, most of which are now disallowed by robots.txt, so I would expect them to go to supplemental, then I have more non-supplemental pages listed.
3. Why does this site not have any PR3 pages listed? When clicking from a PR4 page, is it normal for the next page to be PR2?
Is there anything to be learned regarding the pages listed after all the supplementals? I've seen several opinions on this, but none of them seem to be in agreement.
We had a bunch of pages come out of the supplementals a few weeks back, just to see them go back in to supplemental results. From watching them, it appears as if pages that had recent cache dates were out, now the pages have old cache dates again ( a few months old) and are back in the supplementals.
| This 81 message thread spans 3 pages: < < 81 ( 1  3 ) > > |