| 3:04 pm on Dec 30, 2006 (gmt 0)|
Noticed this for a few of our sites yesterday as well. I gave up on Google a long time ago. Google is the monster from bad dreams...the boogeyman.
| 3:57 pm on Dec 30, 2006 (gmt 0)|
What about the Title / meta description tags? Are they unique too?
| 5:09 pm on Dec 30, 2006 (gmt 0)|
In discussing the Supplemental Index, I haven't noticed anyone talk about freshness for a while. I have a site where the archive section has lots of otherwise kosher pages in the Supplemental Index with PRs in the 2-3 range, but they are also unchanged for over a year or more. Tweaking those pages is on my New Years Resolutions list.
| 10:50 pm on Dec 30, 2006 (gmt 0)|
internetheaven - a few more details will help your question to be answered:
-what steps you have taken to eliminate duplicate content
-how have you arranged your crawlability
-what is quality of your inbound links
-any deep links
-how long has the site been up
-any changes since it was launched
| 11:04 pm on Dec 30, 2006 (gmt 0)|
|What about the Title / meta description tags? Are they unique too? |
|In discussing the Supplemental Index, I haven't noticed anyone talk about freshness for a while. |
Hmmm now I've always disagreed with that. Many of the highest ranking pages in this type of industry (advice/articles) are the oldest and unchanged. What use is a news archive that changes each article every day? ;) I have only found the freshness helps with my commercial websites (i.e. products/services). As many people have pointed out in many threads, Google treats different industries/topics differently ... I would go a step further to say that even the keywords searched for dictate the "type" of search/algorithm Google uses.
|internetheaven - a few more details will help your question to be answered: |
-what steps you have taken to eliminate duplicate content
Yes, I checked that out, there are a few copies of my articles about the place with scrapers but only a few and none of them have backlinks/PR/etc.
|-how have you arranged your crawlability |
Don't understand that question, we've already been crawled, and now they are supplemental. There is a main page which links to some of the articles and those articles link to each other. Each page is a max of three steps from main page, most are two.
|-what is quality of your inbound links |
High, very high. We only have one-way inbounds and nearly all have PR1+
Not as many as I'd like, but yes.
|-how long has the site been up |
Ah, now there are two answers to this. Firstly the domain is over two years old. The "site" has only been up for one year.
|-any changes since it was launched |
Yes, lots. Main page re-written several times, same with inner pages such as contact/terms/about etc. of course the articles haven't changed because they are date specific.
| 11:37 pm on Dec 30, 2006 (gmt 0)|
I used to have many supplementals. Now have zero.
The main thing I did (and don't know if this is what helped):
1) Reduced the boilerplate template code to a minimum.
2) Made sure the content was as high as possible in the html.
3) Decreased the common internal nav links - especially on lower-content pages.
4) 'No indexed' a handful of lower-content pages.
5) made sure each title and description were unique
6) got rid of Keywords Meta tag on most pages (I suspect this had no effect).
All of these things were learned here :) thanks guys.
| 11:53 pm on Dec 30, 2006 (gmt 0)|
Are the supplementals ranking?
Are the supplementals the latest pages or old pages that have been replaced?
Do you always get supplemental on the results, or does that depend on the search query? [ ie is the site supplemental for all results ]
Has the site tanked at any point which you can relate to an event?
Crawlability - you say you have several revisions of your site. Have you checked the index paths are consistant with e.g. "/"?
How have you organised the navigation? Is it heirarchal and running down through theme levels?
Do you have any "boilerplates"? ... see Adam Lasnik's recent response on duplicate content.
Do you have the content repeated on more than one page of your site?
Can you reveal anything more about the history of major events on your site?
sailorjwd - Were your results affected by these supplementals, and did the changes fix this? Are you able to indicate how the recovery worked it's way through?
[edited by: Whitey at 12:11 am (utc) on Dec. 31, 2006]
| 12:54 am on Dec 31, 2006 (gmt 0)|
It may or may not be current or accurate, but what's the PR that shows for the homepage, and how many outbound links are there on the homepage - both within the site or to other sites.
Also - for those pages that are Supplemental, how many links are there to those pages? Are they linked to fom 1, 2, 3 pages or more? And what PR shows for the pages that are linking to those Supplemental pages?
| 1:36 am on Dec 31, 2006 (gmt 0)|
"Were your results affected by these supplementals"
It took me a year or more to get rid of the supplementals. I haven't had supplementals for the last 10-12 months.
If a supplemental pops up in the future I would quickly no index it and/or block it with robot.txt and try to fix the problem, and create a new page.
I'm pretty sure my supplemental problem was a combination of a huge template (frontpage) and pages with too little content. followed by some dup descriptions.
| 3:46 am on Dec 31, 2006 (gmt 0)|
Do anyone else have the sense that Google is up to something new with the Supplemental Index in the past maybe two months or so?
My subjective sense is that if we had the ratio of Suppementals to all URLs from last summer, and could compare that to the same ratio in present time, we would see a jump of major significance. Couple that with recent statements from Matt Cutts and Adam Lasnik recently to the effect that Supplementals aren't always "bad" in every case, and I'm beginning to wonder what's up -- what we think we know may no longer be the whole story.
| 4:08 am on Dec 31, 2006 (gmt 0)|
|recent statements from Matt Cutts and Adam Lasnik recently to the effect that Supplementals aren't always "bad" in every case, |
Isn't that supported by the same page showing in a site: search as supplemental, but in a keyword search as non-supplemental (and rank well), both instances with the same cache date?
At any rate, wouldn't the situation I described make it pretty hard to come up with a count of, or ratio for supplemental pages?
I wonder if a page might be shown as supplemental based on a comaprison to other pages on the same site, but not necessarily when compared to pages on other sites.
| 4:26 am on Dec 31, 2006 (gmt 0)|
I don't know about you but Google is driving me crazy. In a different thread, I was inquiring about a one to two month old site that I created with hundreds of unique pages each with at least 800 words of unique content... as well as links from PR6 sites and below (one way links).
About supplemental results, my site went from supplemental results then to main index and back to supplemental results all within 24 hours (I searched site:www.mysite.com).
In addition, the cache was 12/28/06 and now it is back to early November on all my pages. Why?
Oh well, as I said before, the Google dance continues even though the terms have changed. I hope fewer and fewer people will be willing to dance. It is time for another engine to end this ridiculous Google dominating status.
[edited by: JayDev at 4:27 am (utc) on Dec. 31, 2006]
| 6:15 am on Dec 31, 2006 (gmt 0)|
The Supplemental Index is made up of a URL + a specific cache date. There can be a regular result of the same URL on a different cache date -- usually a more recent one.
|wouldn't the situation I described make it pretty hard to come up with a count of, or ratio for supplemental pages? |
You said it -- that's why I wrote that I have a subjective sense. I cannot conceive of a way to nail down such a ratio.
Here's some more detail that prompts me to suggest that something new is afoot with supplementals.
1) I'm looking at a 1400 page client site that had maybe 20 supplemental urls last spring. Now it has 450. Most of these urls 1) would make very weak entry pages into the site, and 2) suffer from some titles and descriptions that are similar to other pages (10 page breakdowns of long articles, e.g.).
One more really strange observation -- for the snippet of these multi-page supplemental articles, Google is using lines 2 and 3 of a 4 line footer. This is for pages that hold 1,000 words and up.
2) On another site that offers many pdf downloads, only those pdf files are supplemental -- and all of the pdfs are supplemental.
3) On a third site, a previous version of the site has been kept online, linked from the new home page as the "archives". Many deep backlinks go to the archive section -- and yet over the past two years, more and more of those urls are now tagged as Supplemental. It's now up to maybe 60%
All of this looks to me like there may be some new criteria for tagging urls as supplemental, but I can't quite get a handle on it yet.
I should add that these newly supplemental pages are getting some long tail search traffic, and that their parent websites are thriving now even more than a year ago. In short, the new supplemental assignments seem not to be hurting at all, and may even be part of bringing in a better quality of Google traffic.
|the cache was 12/28/06 and now it is back to early November on all my pages. Why? |
1. different data centers?
2. one of those Google mysteries that only we notice because we watch so closely?
| 6:52 am on Dec 31, 2006 (gmt 0)|
|2) On another site that offers many pdf downloads, only those pdf files are supplemental -- and all of the pdfs are supplemental. |
No outbound links from those PDFs, right? That would make them a PR dead end.
| 7:09 am on Dec 31, 2006 (gmt 0)|
Not sure if this fit's into your thought framework, Tedster, but i thought I'd throw it in:
Why does one of our sites show supplementals and old cached pages, yet have new caches of the same page:
Site 1 - it crawls around 9,000 pages a day all dupe content fixed, but lot's of sub pages with addditional content, maybe confused with the leading article page. btw - lot's of quality deep links and a PR 5/6
Site 2 is mostly clean with no supplementals and most pages indexed, and no sub pages of the main article, with few inbound links.
Why is Google picking on 1 and not 2. It may strengthen your view on your first example.
My hunch is that it's creating internal duplicate content and confusion with the sub pages - just wanted to run it past you. btw - we're in the middle of an experiment on this network to remove the sub pages from indexing completely.
Marcia What is convincing you that this is an issue?
I'm seeing some good SEO's putting outbound links to quality sites and ranking well - any definitive/evidenced connection?
[edited by: Whitey at 7:15 am (utc) on Dec. 31, 2006]
| 1:34 pm on Dec 31, 2006 (gmt 0)|
Yes, something is up, something is always up. Right now, Google is consolidating. It is shedding pages in its main index. Google is absolutely winnowing -- as well it should. Here's my take at this point: If you have 1000 pages and half of them are mostly strays, Goog is going to slice you down in the index to about 200 pages. If you have 1000 pages and only a quarter of them are strays, perhaps they might keep 300-400 pages in the main index.
To remain with the 1000 page website as an example: a stray page is one which receives less than one percent of internal links. If you have a website of 1000 pages and less than 10 pages refer to that
page in an internal link, how important could that page be? Drop: I call it a stray. I think this is the minimum new standard.
My sites are undergoing the same experience tedster described in his last post. tedster, by the way, thank you for your participating in this thread.
My sense from observation is that Google is moving to reward websites with wider, more richly linked* footprints as opposed to the narrow footprint site where there is a clear set of home pages on top of 90% secondary support pages of low quality (stray).
In the opening example for this thread, internetheaven described a site top heavy to the home page supported by mostly stray pages. And so Google's latest filtering is following my observations and applying it to internetheaven's example.
For what it is worth, another observation based on IH's example is that 200 pages is the new entry ticket for relevance in Google's world. If a new site does not present 200 pages of quality content and make a wide, not narrow footprint, I don't believe it has a shot of cracking into its market.
*more richly linked per page internally and externally
| 5:47 pm on Dec 31, 2006 (gmt 0)|
"For what it is worth, another observation based on IH's example is that 200 pages is the new entry ticket for relevance in Google's world. If a new site does not present 200 pages of quality content and make a wide, not narrow footprint, I don't believe it has a shot of cracking into its market".
If this is the case Google is finished. What most people are searching for is detailed information, something they are not going to find on a "shotgun site". When I search for product or service the last thing I want is to have to search through 200+ pages of dreck to find the nugget I'm looking for. Content is one thing, but throwing up thousands of pages on a topic that could be covered in 50 and giving them preference could only have one reason.....adsense.
| 6:15 pm on Dec 31, 2006 (gmt 0)|
Maybe Google is shifting some of the job previously handled by so-called "sandbox filtering" over to the Supplemental Index for purposes of more efficient computation.
| 1:55 am on Jan 1, 2007 (gmt 0)|
I just thought I would throw this into the mix. I manage a small 32 page site for a mom and pop business that does very well in it's search queries for it's niche. It's only 32 pages and about 9 months ago it started going supplemental slowly until only the index page was non-supplemental. Still maintained it;s ranking even tho none of those supp pages showed up for even the most extreme search. I thought I had all the bases covered. No reason for the supplementals. Except no deep links really. All went to the mysite.com/-
Well, last month g1smd finally drove it into my punkin head that I WAS missing one thing. I was on a microsoft server with no isapi rewrite installed.
I got it moved over to an apache server last month and installed the base href on every page.
Looked today and four pages are out of supplemental. Looks like when they get crawled now, they come out shortly after. Cache dates are dec. 23rd.
Thanks g1smd! I owe you one. Dang! Why didn't I understand that whole thing sooner.
| 3:03 am on Jan 1, 2007 (gmt 0)|
AY-YI-YI! Okay, I'm desperately trying to get some pages out of supplemental, or at least get them to a state where if they're ever re-indexed by Google they'll pop out. I have links, (not deep links yet), no dupe content issues, no www vs non www isues, but so I can easily work on the site offline, I use all relative linking.
Now after reading Texasville's post, I've been trying to figure out this whole base href tag thing! If I'm using all relative links in on my site, is not having the base href tag really going to make a difference? in the SERPs? Thx in advance!
| 3:09 am on Jan 1, 2007 (gmt 0)|
BushMackel, does your site resolve for both www and non-www?
| 4:26 am on Jan 1, 2007 (gmt 0)|
It does. However, 99% of my content is on subdomains, which is what I'm really interested in appearing in SERPs. I.e., all my content is at widget1.widgets.com and widget2.widgets.com. So...I don't really care if www.widgets.com gets high up on the SERPs. HOWEVER, it should be noted that www.widgets.com is the only page that has a PR and appears naturally in SERPs, (ie not supplemental).
| 6:18 am on Jan 1, 2007 (gmt 0)|
One of my sites is mostly supplementals. For several months, a very similar competing site was not in the supplementals - probably because they had more incoming links. Within the past week or so, their site just went completely supplementals (though their alexa traffic hasn't fallen).
Also another of my sites went completely supplemental (for a long time it wasn't supplementals despite having very similar structure/content to the site that went supplemental).
This points to the "supplementals are increasing" theory.
| 2:58 pm on Jan 1, 2007 (gmt 0)|
Don't know if this will help but here goes:
I just noticed that I accidently copied the 'borders' folder for frontpage onto one of my website folders.
Google indexed the bottom,right,left.htm files.
All are supplemental.
Here are the charactistics of these border files:
1) contain little content
2) are obviously duplicated in many pages
3) have no links pointing to them
4) they have links going out
| 3:22 pm on Jan 1, 2007 (gmt 0)|
So what you are saying is use absolute linking instead of relative links to get out of supplemental?
| 3:26 pm on Jan 1, 2007 (gmt 0)|
Yeah, I guess that's my question too. Or if you're using relative links, should you be using the base href tag?
| 9:22 am on Jan 1, 2007 (gmt 0)|
< This post was moved here from another location >
For more than a month now the site: operator returns only supplemental results for my site. the site is pretty big (more than 50K of unique pages), and more than 4 years old.
This is quite bothering since the internal Google's search box
in my site returns search results from the supplemental pages.
the thing is that my ranking was not harmed at all in the regular web search.
did someone had similar a problem? is it the famous "site operator is broken"?
[edited by: tedster at 6:06 pm (utc) on Jan. 1, 2007]
| 6:23 pm on Jan 1, 2007 (gmt 0)|
I moved your message here because, as I think we can see, your observations are very much on this topic, and they might help us understand better what is going on. You are definitely not alone.
One particular thing you wrote really jumps out at me: "my ranking was not harmed at all in the regular web search."
Does you mean that on regular search, your rankings also do not show the green "Supplemental Results" tag -- but in your Google-powered site search, those same URLs do show as supplemental? If I have that right, do both places have the same cache date?
| 6:35 pm on Jan 1, 2007 (gmt 0)|
erm, are you saying that there was no possible fix for your particular website on a microsoft IIS server?
you couldn't find the IIS equivalent fix?
| This 81 message thread spans 3 pages: 81 (  2 3 ) > > |