Adding more content - fearing diluting PR and getting SI probs

[privatefleet]

Hi all,

Have been having a bit of a problem with many of my pages appearing in the supplemental index. After lots of reaseach, I've come to the conclusion that it's come down to content/PR ratio - ie I haven't got enough incoming links to pass on enough PR to keep many of these pages out of the SI.

So I have two questions:

1) does this mean it's pointless adding any more content and I should just concentrate 100% on building IBLs?

I've been furiously trying to build links for the last month or two and it hasn't helped all that much. Also as quality link building often relies on fresh and quality content, there's a conflict there. If I create pages as 'link bait' there's a danger that if no-one links to this page it will just compound the problem and further dilute the PR

2) if I do add content, is it best to add fewer bigger pages rather than many smaller more targetted pages. In an ideal world the latter would be preferable as it means one can target strongly for certain keywords but maybe by putting all the content on one page it conserves more PR juice?

With this 2nd question I'm thinking about a glossary with a couple of hundred terms defined and described. Originally I had thought to have a little page (and possibly image) for each item. This would mean I could use the image, page title, keyword density etc to do some good on-page SEO. However will this just dilute PR so much that none of the pages make the main index. The alternatives (A massive page with hundreds of definitions and thousands of words, or 26 pages split by letter etc) don't sound too user friendly.

Thanks in advance

Site is Private Fleet - New Car Broker obtaining fleet discounts on new dealer cars in Australia : Private Fleet : if that's at all relevant.

Dave

[crankydave]

Right off the bat the first problem I see is the duplicated meta descriptions. This can easily push pages into the SI. You need for them to be unique for each page.

Dave

[Peter (IMC)]

Maybe I'm too sleepy at this moment... I don't see any pages in the supplemental index and neither do I see the same meta description in all pages.

What Google shows in the snippets when you do a site:www.domain.com search aren't the meta descriptions.

[Webnauts]

I was just wondering, how could you figure out that you are in the supplemental?
I found for this domain privatefleet.com.au, 922 pages being indexed, which 726 of them are in the supplemental index.

There are many reasons that you can be in the supplemental index. That requires an investigation and then planning how to deal with those issues.

THAT HURTS!

Read here why: http://www.webproworld.com/340560-post25.html

[privatefleet]

Thanks for the replies guys...

Most of the meta descriptions are unique but some are duplicate so I'll go through and make sure each is 100% unique. Thanks crankydave, that's an easy one to fix but I doubt it will solve the prob completely.

I found this tool - Supplemental Index Ratio Calculator which is pretty handy.

It's actually very interesting. I was testing out a few competitor sites... eg if you run it for

Used Cars - New Cars - Search New & Used Cars For Sale - carsales.com.au you get

Google has a total of 24700 pages indexed from Used Cars - New Cars - Search New & Used Cars For Sale - carsales.com.au
1690 are in the main index
23010 are in the supplemental index

This means an amazing 93% of carsale's pages are in the SI

and then run for Used Cars & New Cars for Sale | Car Sales | Car Reviews | drive.com.au (a direct competitor)


Google has a total of 136000 pages indexed from Used Cars & New Cars for Sale | Car Sales | Car Reviews | drive.com.au
70300 are in the main index
65700 are in the supplemental index

Which is only 48%

However, the fact is that CarSales get's heaps more traffic than Drive (over twice as much I think) despite having a fraction of the amount of pages Drive has in the main index.

So whilst anyone would want as many pages as poss in the main index, it doesn't seem to be the be all and end all (disclaimer: of course both these websites will attract traffic from lots of other sources other than organic searches)

Anyway, I'm digressing - anyone got any comments regarding my original question - about whether adding fresh content will further cause SI angst and if you do, what's the best way to do this and avoid these issues?

Thanks again

[privatefleet]

oh and thanks for the link to that tool, Webnauts - tried to download a few weeks ago but ran into problems...

I'll see if I can do it again cos it sure would be valuable

[cass-hacks]

Originally Posted by privatefleet

Have been having a bit of a problem with many of my pages appearing in the supplemental index. After lots of reaseach, I've come to the conclusion that it's come down to content/PR ratio - ie I haven't got enough incoming links to pass on enough PR to keep many of these pages out of the SI.

Correct.

From Fall weather forecast

PageRank is the primary factor determining whether a url is in the main web index vs. the supplemental results

From Infrastructure status, January 2007

Having urls in the supplemental results doesn’t mean that you have some sort of penalty at all; the main determinant of whether a url is in our main web index or in the supplemental index is PageRank.

There are cases where Google decides that the content of a page is barely worth indexing in which case it may end up in the Supplemental index but generally speaking, PageRank is the deciding factor.

Duplicate content, by itself and duplicate titles/descriptions/keywords/blah/blah/blah have nothing to do with a page being in the Supplemental index.

Having duplicate content may indirectly cause a page or pages to go Supplemental but it would be a PageRank dilution issue instead of being directly caused by duplication.

Originally Posted by privatefleet

So I have two questions:

1) does this mean it's pointless adding any more content and I should just concentrate 100% on building IBLs?

It is NEVER pointless to add more content!

But, regarding the Supplemental index independently, inbounds are what you need, along with making sure that whatever PageRank you have is funneled to where it needs to be to keep your most important pages in the Main index.

Originally Posted by privatefleet

I've been furiously trying to build links for the last month or two and it hasn't helped all that much. Also as quality link building often relies on fresh and quality content, there's a conflict there. If I create pages as 'link bait' there's a danger that if no-one links to this page it will just compound the problem and further dilute the PR

There are ways to keep a page from accruing PageRank from other pages in your site.

Also, if there are pages that are indexed that don't really need to be, "About us", "Copyright", "Privacy" or other types of pages that people really aren't going to be searching for, they are not doing you any good in the index so why not disallow them in robots.txt and nofollow links from your site to them?

No sense in wasting PageRank on pages that aren't likely to show up in searches anyway or aren't likely to contain content that people are likely to search for.

Add to that, making sure the pages you really care about "get their's" before other pages will help keep your more important pages above water.

Originally Posted by privatefleet

2) if I do add content, is it best to add fewer bigger pages rather than many smaller more targetted pages. In an ideal world the latter would be preferable as it means one can target strongly for certain keywords but maybe by putting all the content on one page it conserves more PR juice?

Unless you are talking about a lot of pages, it won't make all that much difference.

But all things being equal, fewer bigger pages would be better.

Originally Posted by privatefleet

With this 2nd question I'm thinking about a glossary with a couple of hundred terms defined and described. Originally I had thought to have a little page (and possibly image) for each item. This would mean I could use the image, page title, keyword density etc to do some good on-page SEO. However will this just dilute PR so much that none of the pages make the main index. The alternatives (A massive page with hundreds of definitions and thousands of words, or 26 pages split by letter etc) don't sound too user friendly.

Even were PageRank not an issue, I wouldn't chop up a glossary anyway, unless you wanted more page loads for ads or something. I'd put them all on one page and then put a in-page-link menu at the top of the page so that a visitor can see a list of all the terms and quickly go to ones of interest.

At least that's how I see things, your mileage may vary.

[isharli]

Please clarify...does one need to have more inbound links to pages that are categorized as supplemental in order to get them out of that category?

[Webnauts]

I started a discussion in my own forums about the supplemental results. You might would like to check it out too: Do you have pages in the supplemental results? - SEO Workers

[cass-hacks]

Originally Posted by isharli

Please clarify...does one need to have more inbound links to pages that are categorized as supplemental in order to get them out of that category?

Of course links directly to the Supplemental index pages will help pull them out faster but links to other pages within the site, depending on navigation, will likely help, although at a slower rate.

[dvduval]

I truly believe that if you add good and original content that is good enough that people enjoy reading it, you will slowly build readership and traffic. I think you should think about your readers first, and search engines second. You may only have a few at first, but that will change over time with a continued effort.

[cass-hacks]

Originally Posted by dvduval

I truly believe that if you add good and original content that is good enough that people enjoy reading it, you will slowly build readership and traffic. I think you should think about your readers first, and search engines second. You may only have a few at first, but that will change over time with a continued effort.

If one is willing to wait it out or, "Build it and they will come." that is perfect advice.

Consider that any brick and mortar business can take up to 3 years to show profitability, the Internet is often no different.

From my experience what I have learned is that it is a good idea to get enough PageRank to get a few important pages ranking well and then leave it to "nature" to do the rest.

But, while nature is doing its thing, see what you can do about getting more eyeballs on your pages independent of search engines, i.e. social networking and link sharing sites.

In the long run, the more eyeballs on a given page, the more chances to earn links naturally.

All that said, just like there are short cuts one can take in the meat world to TRY to reach success sooner, short cuts exist in the virtual world but there are easily as many risks by taking that route in either realm.

[Webnauts]

Originally Posted by cass-hacks

Of course links directly to the Supplemental index pages will help pull them out faster but links to other pages within the site, depending on navigation, will likely help, although at a slower rate.

Common Causes of Supplemental Pages:

• Pages due to canonicalization problems, e.g duplicated content, too much content similarity;
• Pages with low or no content;
• Orphaned web pages. Pages that no one links to, including your own;
• Error pages, if a site does not use If-Modified-Since, Last Modified and/or Expires rules;
• Poor website navigation;
• Loading pages with irrelevant keywords (keyword stuffing);
• Too low PageRank;
• Long URLs, especially with long parameters, starting with a question mark (?) and being separated with an ampersand (&) and are not rewriten;
• Suspicious pages for spamindexing, like non-unique and irrelevant to page content heading tags, meta tags, or linking to bad neighborhoods, etc.

So do you want to say here that linking to supplemental pages is the best solution? If the pages have too low PageRank I could understand and would agree, if they are important pages though. But what if the problem is one of the other cases?

[Webnauts]

Sorry. Duplicated my mistake my above post.

Mods please delete this one. Thanks.

[cass-hacks]

Originally Posted by Webnauts

So do you want to say here that linking to supplemental pages is the best solution? If the pages have too low PageRank I could understand and would agree, if they are important pages though. But what if the problem is one of the other cases?

What I was replying to was a question about getting more inbound links. More inbound links directly to a page in the Supplemental index is more likely to have a positive effect since "PageRank is the primary factor determining whether a url is in the main web index vs. the supplemental results"

As for the other causes,

Pages due to canonicalization problems, e.g duplicated content, too much content similarity; doesn't seem to mention anything about about Supplemental pages unless I missed something.

Pages with low or no content; doesn't seem to either.

Error pages, if a site does not use If-Modified-Since, Last Modified and/or Expires rules; again, no mention of Supplemental indexed pages being the result.

Actually, none of the links you list say anything about those problems causing pages to be moved to or start out and stay in the Supplemental index.

The link to the page about too low of PageRank would be a good place for Google to mention the association but they didn't.

The only references by a Googler regarding a cause for a given page being in the Supplemental index that I have found is Fall weather forecast and Infrastructure status, January 2007 although that is not to say other info from Googlers isn't out there somewhere, just that I haven't found it.

[Webnauts]

If you claim now that what I posted above is not correct , to be specific about the causes of the supplemental results, then I feel very sorry for you.

But before I add something else here, may I ask if you are a professional SEO?

[cass-hacks]

Originally Posted by Webnauts

If you claim now that what I posted above is not correct , to be specific about the causes of the supplemental results, then I feel very sorry for you.

But before I add something else here, may I ask if you are a professional SEO?

I think what I wrote speaks for itself, no?

It is not a matter of me claiming anything. I was just making an observation.

You, on the other hand, are making claims that information from the Google Webmaster Help system is proof of something yet that does not seem to be the case.

Do any of the pages you linked to speak about the topics mentioned respectively causing pages to go Supplemental?

As for you feeling sorry, I'm not sure what your personal feelings have to do with a technical discussion.

"Professional SEO"? I get paid to do what I do. It is not be hard to find people who claim to be SEO experts yet don't seem to know the first thing about what it is they claim to be and usually are the first to ask why their site selling SEO services don't rank. In other words, titles mean very little.

Would it make any difference were I to say I was a Professional SEO? Unless there was some certification process, I don't see the point.

[Webnauts]

First I had no intention to offend you. I did not know if you are a SEO or not.

So here is my next question:

Do duplicated pages go into the supplemental results or not?

[cass-hacks]

Originally Posted by Webnauts

First I had no intention to offend you.

And of course no offense taken.

Originally Posted by Webnauts

Do duplicated pages go into the supplemental results or not?

There seems to be no single answer other than, "it depends".

Of two pages with the same content, the only thing that can be said for sure is that more than likely only one would ever show up in SERPs for searches for which they might both apply.

If the two pages are on different sites, it depends on their respective PageRank as to whether or not either page is Supplemental or not but being a duplicate in and of itself would not be a direct factor although it could be due to a knock on effect if there isn't sufficient PageRank to go around in the first place. But, in the case of there being insufficient PageRank, it seems to pretty much be a crap shoot as to whether or not a given page is Supplemental or not.

If the two pages are on the same site, in the case of both pages being indexed at all again, it seems to depend on PageRank, or Google realizing they're duplicates and either not indexing one at all or sticking on in the Supplemental index where it isn't hurting anything anyway since it is a duplicate anyway.

More often than not though it seems Google simply chooses one to index and totally ignores the other. Consider the case where / and /index.html both exist without one being redirected to the other, not very often are they both actually indexed.

That said, I would actually agree that duplication can have an effect on a given page's Main/Supplemental chances but usually only when it is indexed at all and when there is insufficient PageRank to begin with.

[crankydave]

Originally Posted by Peter (IMC)

Maybe I'm too sleepy at this moment... I don't see any pages in the supplemental index and neither do I see the same meta description in all pages.

What Google shows in the snippets when you do a site:www.domain.com search aren't the meta descriptions.

Check the source code Peter. Yes they are. No not all pages are duplicating the description. Don't think that's what I said but many are.

Dave

[crankydave]

Another thing I noticed, may or may not be causing problems, but I noticed a lot of links with session ID's. It's possible that these are being considered duplicate pages.

Dave

[Webnauts]

Originally Posted by cass-hacks

And of course no offense taken. .

Great!

Originally Posted by cass-hacks

If the two pages are on the same site, in the case of both pages being indexed at all again, it seems to depend on PageRank, or Google realizing they're duplicates and either not indexing one at all or sticking on in the Supplemental index where it isn't hurting anything anyway since it is a duplicate anyway.

We recently messed up our redirects in our forums. Aren't the 3 pages found in Google index the same?

http://forums.seoworkers.com/web-content-accessibility/google-accessible-web-search-for-the-visually-impaired-t26.0.html - Google Search

How comes that they are all indexed since they are tripple? And do you mean that there is no danger of diluting PR? And what would happen if I would organize two links for the duplicated pages? Don't look at the green bar, as those new pages are recently duplicated, so you can tell much about that.

[Peter (IMC)]

Well,.. I just wrote a long post, then the system had an error and didn't save the post... oh well, I'm not going to try to write it all over again, but will just get to the point right away.

You´re paying for links and since Google hates paid links, that´s probably the reason that most of your internal pages don't have PR. Matt Cutts says in his blog (Infrastructure status, January 2007):

the main determinant of whether a url is in our main web index or in the supplemental index is PageRank

Do your self and your wallet a favor and stop paying for those links. If you spent over $ 500 per month on it, you may be able to afford a company that helps you get your site into newspapers, magazines, etc. It may get you a lot of visitors and also links, the natural type that Google loves so much.

[Peter (IMC)]

Originally Posted by crankydave

Check the source code Peter. Yes they are. No not all pages are duplicating the description. Don't think that's what I said but many are.

Dave

I really was sleepy I guess,. I checked like 3 pages and they were all different,.. I guess I just had luck to find 3 different descriptions.

[espmartin]

I believe in total white hat techniques. But what I don't agree with is "bowing down" to
Google, JUST because they want to keep the "Paying For Links" business to themselves
(ie: Adwords)

You´re paying for links and since Google hates paid links...

Keep Matt Cutt's happy with "Sponsored Links/Ads/Supporters" - whatever you want to label
your link marketing, but if you want to buy a quality link from a quality site that would
make your readership that much more grateful - by all means buy them!

Even better, if your site is an authority in your niche (aren't we all?) - and you can
monetize your webmaster labors, buy all means sell some real estate off your site.

Now, I do say this knowing that you have to "play nice" with Google, and I'm in no way
telling y'all to shoot yourself in the foot, but...Come on.

Be wise, be creative, be marketable!

[Peter (IMC)]

I don't agree with Google telling people how to code their pages. However, I do believe that Google has every right to use code they way they feel is best for providing the best search results.

The issue is not the paid links them selves, but the lack of doing other things when people buy links. Paid links are no better in getting natural links than link exchanges are.

The site in question has a PR5 on the homepage. What would that PR be if the paid links are not taken into account? The toolbar does take them into account as far as I know, but the algorithms probably deal with it differently. In anyway, it seems that Google is not passing a lot of PR to the internal pages the usual way.

The paid links aren't causing the problem, but they do hide the real problem, which is a lack of natural links. So I say, take out the paid links and use the money for other marketing techniques.

[Webnauts]

Originally Posted by privatefleet

Hi all,

1) does this mean it's pointless adding any more content and I should just concentrate 100% on building IBLs?

I guess you would like to read this article Got Supplementals? Accepting PageRank is Only The Beginning » Half’s SEO Notebook which it explains pretty well the causes of the supplemental results and how to deal with them.

[cass-hacks]

Originally Posted by Webnauts

We recently messed up our redirects in our forums. Aren't the 3 pages found in Google index the same?

Google seems to err on the side of indexing pages first and then determining if they really should be there second. The duplicate pages will likely be dropped fairly soon depending on the general crawl rate of the site.

If a site has canonical problems, both www and non-www being served, which would be a classic case of duplication, rarely are both versions indexed or if they are, indexed for very long.

What is more damaging are pages that aren't quite duplicates, such as many affiliate retail sites have, that aren't similar enough for Google to determine they are essentially the same. Even though there is very little difference between them, Google will likely try to index them which then would more likely cause PageRank dilution due to multiple pages which likely would be applicable for basically the same searches and since only one, or at most two, with one being indented, would likely show up in any search results, any additional pages would essentially be wasted.

[privatefleet]

Come bcak after the w/e to see a lot of discussion - thanks guys! Only one person has actually attempted to answer my original question but nevetheless the discussion regarding why I have this SI issue is probably even more valuable so thanks!

Thanks for your feedback cass-hacks regarding the glossary but don't you thing this could result in a ridiculously big page - say 200 paragraphs (definitions) of 100-200 words each? Maybe Google would even penalise the page because of it's size?

Thanks for the 'tip re nofollowing' unimportant pages - did try this a month or 2 ago so hopefully it's having an effect of some sort.

Re duplicated mata descriptions - that's an easy one to fix albeit tedious so I'll get onto that.

Re "Paid Links"? Who me? OK I admit there could be the odd one - have also paid some directories for inclusion and that sort of thing. However the vast vast majority of my links are not paid for.

I guess what everyone is saying is it comes down to getting more organic links but that's where I'm struggling. For example, I created a page - Top 5 Car Dealer Tricks... and how to avoid them. : Private Fleet : paid a professional to write it, bought some nice images then spent a few hundred getting the article distributed on PRweb, Netscape etc, featured it in our e-newsletter, basically did whatever I could to get eyeballs to it hoping links would result. However the experiment failed (on a ROI basis anyway) and I got bugger-all links. All I got was a letter from the car dealers association asking (very nicely actually) that I reconsider the page as it paints dealers in a bad light! This leaves me struggling to see how it could be as simple as that (in my industry) - create quality content and links will come.

[cass-hacks]

Originally Posted by privatefleet

Thanks for your feedback cass-hacks regarding the glossary but don't you thing this could result in a ridiculously big page - say 200 paragraphs (definitions) of 100-200 words each? Maybe Google would even penalise the page because of it's size?

It used to be the case that search engines would only index a certain amount of a given page. I forget how much it was but as far as I can tell, there is no longer a limitation.

That said, if you are talking about 200 paragraphs X 200 words and guestimate 10 characters/word, that would be about 400k. That might be a bit much for reasons beyond just what a given engine can't deal with.

If you are concerned about page count AND page size, can you break things down into categories of definitions?

Only you can decide if that is possible but I think if possible, putting as many related definitions on a page as possible would be better for both your site and its visitors but then again, it depends on where and how the definitions are used and how they are implemented.

One thing related, I have seen a lot written about how to get one's definitions to be used by search engines when using the define: search operator but have seen no single method consistently succeed. You might try executing a few define: searches on some of the terms you plan to provide definitions for and see how others are getting in the SERPs for those searches but this is quite far off topic.

Originally Posted by privatefleet

Thanks for the 'tip re nofollowing' unimportant pages - did try this a month or 2 ago so hopefully it's having an effect of some sort.

It is interesting that you chose to only use nofollow, which I would say is the best of both worlds.

You don't use any PageRank on a nofollow'ed page but at the same time, if someone else links to it, you benefit because of links on the nofollow'ed page back to your site. It effectively becomes a one-way PageRank valve.

On the other hand, if you both nofollow AND disallow, that page essentially becomes an island and any PageRank it accrues from outside stays on that page so just nofollow'ing in your case is actually a better idea than what I suggested in the long term although in the long term your site will hopefully have accrued enough inbound links that you won't have to worry about it anymore.

Originally Posted by privatefleet

Re duplicated mata descriptions - that's an easy one to fix albeit tedious so I'll get onto that.

Considering that Google often ignores the description meta tag and creates its own snippets for search results, at worst you would be ignoring a possible plus for determining relevancy.

Pages that show up as "Similar results" do not mean they are Supplemental, only that their being listed in the given search would not help one choose between them.

It would be good to fix them in any event though simply because of the "Similar results" causing pages to be "hidden" by the similar results link being less clickable.

Originally Posted by privatefleet

Re "Paid Links"? Who me? OK I admit there could be the odd one - have also paid some directories for inclusion and that sort of thing. However the vast vast majority of my links are not paid for.

So far Google doesn't seem to penalize sites that have bought links although if you consider a link being devalued as being penalized, I guess it would seem the same. So, buying links at worst is a waste of money. On the other hand, if you are selling links on a significant scale, that maybe possibly perhaps wouldn't be good but you don't seem to be in that position anyway.

Originally Posted by privatefleet

I guess what everyone is saying is it comes down to getting more organic links but that's where I'm struggling. For example, I created a page - Top 5 Car Dealer Tricks... and how to avoid them. : Private Fleet : paid a professional to write it, bought some nice images then spent a few hundred getting the article distributed on PRweb, Netscape etc, featured it in our e-newsletter, basically did whatever I could to get eyeballs to it hoping links would result. However the experiment failed (on a ROI basis anyway) and I got bugger-all links. All I got was a letter from the car dealers association asking (very nicely actually) that I reconsider the page as it paints dealers in a bad light! This leaves me struggling to see how it could be as simple as that (in my industry) - create quality content and links will come.

I wouldn't say the experiment has failed yet although maybe it has within some time frame you have imposed. Personally, with what could be considered a "negative" piece and looking at how well articles like those do on social networking sites like Digg, Stumbleupon, Shoutcast etc. you might give those types of promotions a try as well. At least they are free.

[Peter (IMC)]

Re "Paid Links"? Who me? OK I admit there could be the odd one - have also paid some directories for inclusion and that sort of thing. However the vast vast majority of my links are not paid for.

What do you mean "odd one"? 90% of the links that Google shows are paid links.

[privatefleet]

Thanks cass-hacks - all great advice there. You're right, I'll try and categorise these glossary definitions and split it up into 10 pages, say.

And Peter, well you know as well as anyone that the links that Google shows are usually a very small proportion of total links and tend to be somewhat arbitrarily selected. I know what you are referring to but all those 'paid for' links are coming from the same domain anyway so I'm sure G would have only ever counted it as a single link.

Going by my favourite backlink checker - Webmasters tool (SEO) : Host and IP Addresses Backlinks Checker or through sitemaps, I can assure you that only a couple are paid for.

However I like your advice re newspapers, magazines etc. Maybe going offline could be the best way to get new links...

[cass-hacks]

Originally Posted by privatefleet

Going by my favourite backlink checker -

Wow! Cool tool!

I've always used Yahoo Site Search but the tool you linked to adds a LOT of useful information!

Thanks greatly!!

[Peter (IMC)]

Originally Posted by privatefleet

Thanks cass-hacks - all great advice there. You're right, I'll try and categorise these glossary definitions and split it up into 10 pages, say.

And Peter, well you know as well as anyone that the links that Google shows are usually a very small proportion of total links and tend to be somewhat arbitrarily selected. I know what you are referring to but all those 'paid for' links are coming from the same domain anyway so I'm sure G would have only ever counted it as a single link.

Going by my favourite backlink checker - Webmasters tool (SEO) : Host and IP Addresses Backlinks Checker or through sitemaps, I can assure you that only a couple are paid for.

However I like your advice re newspapers, magazines etc. Maybe going offline could be the best way to get new links...

I guess I have a different definition of what a paid link is...

Your tool tells me the same thing: "Way too many paid links."

All those newspaper sites are also paid links if you ask me. Even those links from that review site are paid links.

How many of your backlinks have the word "sponsored" or something similar close to it?

Now please don't get me wrong. I don't think these paid links are hurting your website, but I do believe they hide the real problem. I tried site explorer of Yahoo to find links that are considered more natural, but it is just page after page with paid links. And in my opinion (but that´s just my opinion) site wide links aren't counted as 1 link just like that. If, for example (and I don't know if they do) Google has some kind of factor that calculates percentage of links that are paid, your score would be 90% or so.

By the way, the site lacks internal links. You think it is possible to get an average of 4 internal links per page? (Not counting menu links and footer links).

On another note. If you´re considering more content, try to be more interactive. (Web 2.0) Perhaps a video here and there? More pictures would be good too. All there is to link to now is huge amounts of text.

[privatefleet]

Thanks Peter,

Point taken. I suppose when i have bought links in the past the fact that some of these have been sitewide has made the proposition more attractive so those are the ones I've gone for.

Per domain (ie if you took every single domain that has a link to our site and weighed up those domains that have been paid against those that haven't) I'd still like to stress that any paid links would be a tiny percentage.

When you say 4 internal links per page, can you tell me exactly what you mean? I should interlink pages more from within the text of other pages?

Web 2.0 it certainly is not but do have some plans to add forums etc and get a bit more interactivity and user input. Thanks for that suggestion

[cass-hacks]

Originally Posted by privatefleet

When you say 4 internal links per page, can you tell me exactly what you mean? I should interlink pages more from within the text of other pages?

I am curious about this as well. If Peter could explain the benefits he sees of in-context/content links as opposed to menu/footer links, that would be very interesting!

Also, would in-context/content links need to be in replacement of or could they be in addition to menu/footer links?

I would guess that in-context/content links could add relevancy as well as the opportunity for use-specific anchor text as opposed to what would be more generally found in a menu but would it have to come before a menu or other link for it to be effective or would it have to not exist in the menu on that page?

I've heard that if multiple links on the same page point to another page, only the first anchor text is "used" although that may be old info that no longer applies.

[Peter (IMC)]

Internal links that are in the menu, footer, etc. repeat them selves in all pages. It is only logical to assume that search engines deal with these kind of links differently. And even if they didn't, it's more links, but with the same anchor texts.

Considering the huge amount of textual content in the website, you can do something like Wikipedia does with changing many words into links to other pages. Wikipedia doesn't look at relevance but you can and choose to link between pages only with words that are related to the subject of the page.

Another thing you can do is add "breadcrumbs" to all your pages. These are the links at the top of a page that show you where you are in the site. For example:

home >> porsche >> Cayenne

This not only helps the visitor, but you automatically have a lot of related links in your pages. If you have to do this manually, it's a big job,.. but very much worth it.

Another thing you can do, for example here: Porsche cars review : Private Fleet :

At the bottom you see the 4 links to different models. Add those links in each of the pages they link to (just take out in each page the link to the page it self.) You can put them under the header: "Other Models of Porsche:"

These kind of internal links are very valuable. Implement it all and you'll be so happy with the results you'll be wanting to know my paypal email address to not feel bad about receiving this kind of advice for free. These kind of links do not just help with search engines, but they also help your visitors and you may see the average number of pageviews per visit go up and conversion rates as well.

[crankydave]

Originally Posted by cass-hacks

Considering that Google often ignores the description meta tag and creates its own snippets for search results, at worst you would be ignoring a possible plus for determining relevancy.

Google can and does use the meta description for filtering purposes. Duplicating the tag can and does get pages filtered and/or pushed into the SI even though they will not rank a page on what is the the description.

Dave

[cass-hacks]

Originally Posted by Peter (IMC)

... And even if they didn't, it's more links, but with the same anchor texts.

This reminds me of a friend who plays the lottery. He always buys two of the same set of numbers in case he wins but is not the only one to win. By buying more than one ticket, he gets a larger share.

What I wonder about though is how does Google treat anchor text from multiple links on the same page to a same different page? If the anchor text are different even though they point to the same page, does Google consider the different anchor text or only the first links anchor text?

Even if Google ignores anchor text beyond the first of a given link, the bread crumbs and context links are excellent ideas! But, if you can slip some different anchor text in at the same time, making the links more context rich insitu the benefits would seem even greater!

[cass-hacks]

Originally Posted by crankydave

Google can and does use the meta description for filtering purposes. Duplicating the tag can and does get pages filtered and/or pushed into the SI even though they will not rank a page on what is the the description.

That would seem to contradict what was recently posted on GoogleWebmasterCentral, Official Google Webmaster Central Blog: Improve snippets with a meta description makeover

Specifically,

And it's worth noting that while accurate meta descriptions can improve clickthrough, they won't affect your ranking within search results.

If they don't effect ranking, why would they have any effect on indexing?

[crankydave]

Originally Posted by cass-hacks

That would seem to contradict what was recently posted on GoogleWebmasterCentral, Official Google Webmaster Central Blog: Improve snippets with a meta description makeover

Specifically,

If they don't effect ranking, why would they have any effect on indexing?

Notice how they carefully used the word "ranking" and not "filtering".

They are basically saying the same thing that I posted.

Dave

[cass-hacks]

Originally Posted by crankydave

They are basically saying the same thing that I posted.

I must be missing something, where does that blog post mention anything about filtering or supplemental index?

Could you explain the connection that you see?

[kgun]

You find the word "ranking" with page search (CTRL + F). You do not find anything directly related to "filtering" and / or "supplemantal index" on the page.

The semantic web is about meaning.

[Peter (IMC)]

Originally Posted by cass-hacks

This reminds me of a friend who plays the lottery. He always buys two of the same set of numbers in case he wins but is not the only one to win. By buying more than one ticket, he gets a larger share.

What I wonder about though is how does Google treat anchor text from multiple links on the same page to a same different page? If the anchor text are different even though they point to the same page, does Google consider the different anchor text or only the first links anchor text?

Even if Google ignores anchor text beyond the first of a given link, the bread crumbs and context links are excellent ideas! But, if you can slip some different anchor text in at the same time, making the links more context rich insitu the benefits would seem even greater!

Multiple links on a page to the same page is not a problem, but at the same time it means the page links to just 1 other page. If you do that just to add more anchor texts you have to keep in mind that the anchor text is not just important for the page it links to, but also to the page the link is in. The anchor text needs to be relevant to both pages.

[kgun]

Originally Posted by Peter (IMC)

The anchor text needs to be relevant to both pages.

Like mosaic where the final picture is more valuable than the sum of the individual parts.

It is more and more difficult to fool Google, and if you try, the long run effect may be negative.

[crankydave]

Originally Posted by cass-hacks

I must be missing something, where does that blog post mention anything about filtering or supplemental index?

Could you explain the connection that you see?

That's exactly the point.

Here's what I posted...

...even though they will not rank a page on what is the the description.

From the blog you cited...

...they won't affect your ranking within search results.

There is no contradiction.

Keep in mind that "ranking" and "filtering" are two entirely different things and the blog does not state that it won't affecting filtering or what index a page is placed in.

Again, no contradiction.

Dave

[cass-hacks]

Originally Posted by crankydave

Keep in mind that "ranking" and "filtering" are two entirely different things and the blog does not state that it won't affecting filtering or what index a page is placed in.

Again, no contradiction.

The blog article does not state a LOT of things. That does not mean that all the things it does not state are true.

But, maybe the issue is what does "filter" mean that you are using? Do you mean by filtered pages only shown after the "view similar results" is clicked on?

If that is the case, the problem is not with the meta-description or whether or not the page even had one. The problem is that as good as Google is at comping up with its own snippets, it didn't have enough to work with to come up with a unique one for the page in question. That is not a problem of the meta-description or even not having one, that is a problem of poor page content.

On the other hand, if that is not what you mean by "filtering" could you please define it because I've seen so many people use the term "filter" in so many different ways that it is hard to figure out the context.

Sorry, I forgot to add, regarding Supplemental. If a meta-description effected whether one was in the Supplemental index or not, and one changed the meta-description and autmagically moved out of the Supplemental index and started ranking for relevant searches, that would seem to effect ranking then, i.e. no rank before, rank after, no?

[Webnauts]

Originally Posted by cass-hacks

That would seem to contradict what was recently posted on GoogleWebmasterCentral, Official Google Webmaster Central Blog: Improve snippets with a meta description makeover

Specifically,

If they don't effect ranking, why would they have any effect on indexing?

Not everything is about ranking.

Two of many examples:

Google in their webmaster guidelines say:

Originally Posted by Google

Check for broken links and correct HTML.

Site wide I have no broken links and my pages HTML validates 100%.

Originally Posted by Google

Make sure your web server supports the If-Modified-Since HTTP header. This feature allows your web server to tell Google whether your content has changed since we last crawled your site. Supporting this feature saves you bandwidth and overhead.

I do that too.

Does all that all mean that I will rank better?

Google is using sort a of bayesian filters, to help them get rid of their spam.
So the primary usefulness of description tag in Google is to avoid the possibility of dupe filter hitting your page.

We had several clients so far being penalized with -30 penalty only because the did not use site wide unique meta description tags.

Therefore I do not see any contradiction there. They just provide tips how to write quality description tags.

If you have doubts, add irrelevant or non-unique meta tags on your site, and come back and tell us what happen.

[Webnauts]

Originally Posted by cass-hacks

Sorry, I forgot to add, regarding Supplemental. If a meta-description effected whether one was in the Supplemental index or not, and one changed the meta-description and autmagically moved out of the Supplemental index and started ranking for relevant searches, that would seem to effect ranking then, i.e. no rank before, rank after, no?

When you abuse the meta tags, either you go to SI or from my experience you get punished with the -30 or -930 penalty. If you know what these penalties are, you will understand why you rank well again after you fix the problems.

[cass-hacks]

Originally Posted by Webnauts

Not everything is about ranking.

True, but that is not what is being proposed or questioned.

The question is, does a given page's meta-description effect "filtering" or a page being in the Supplemental index.

Originally Posted by Webnauts

Check for broken links and correct HTML.

Originally Posted by Webnauts

Make sure your web server supports the If-Modified-Since HTTP header. This feature allows your web server to tell Google whether your content has changed since we last crawled your site. Supporting this feature saves you bandwidth and overhead.

What do either of those issues have to do with "filtering" or a page being in the Supplemental index?

Originally Posted by Webnauts

So the primary usefulness of description tag in Google is to avoid the possibility of dupe filter hitting your page.

According to whom? Besides which what is this "dupe filter" you keep talking about?

So far, all I have seen is many people on the outside saying a "dupe filter" exists while those on the inside say no such thing exists.

Barring a word from Google describing how a "dupe filter" exists, do you have any conclusive repeatable proof that such a filter exists?

Besides all that, how does having a meta-description help in preventing such a filter? Why not the keyword meta-tag also?

You could probably say, "Google says they don't use the keyword meta-tag." but at the same time, why don't you accept what Google says they do use and what they do use it for, without trying to infer all sorts of additional functionality that isn't even implied?

Originally Posted by Webnauts

If you have doubts, add irrelevant or non-unique meta tags on your site, and come back and tell us what happen.

Why would I need to do that? Search results are filled with examples.

How about the site that is the topic of this thread? How many pages have exactly the same meta-description? Why aren't all the pages with the same meta-description Supplemental or "filtered"? At the same time, pick a page that is in the Main index, copy and paste a string of text from the page into a search and see how the page does.

Since there seems to be a reasonable sampling of pages without the same meta-description as have the same that are not "filtered" and not Supplemental, how can you point to meta-descriptions and say they are the cause of problems?

Another site posted in this forum, Gonna Luv It - Sterling Silver Jewelry most of the pages with the same description yet not all of those pages are "filtered" or Supplemental. Similar situation, some pages indexed, some not, some with unique meta-descriptions, some without but in any case, there would seem to be no distinction in how Google treats them.

I can't find a link to it now, although I can if need be but in this site's case there are pages in Google's Main index from the site that are the output of a dynamic image generator that is nothing but seemingly random byte code with not a single meta-description to be found. But yet, it is not "filtered" and it is in Google's Main index, i.e. not in the Supplemental. Not even a chance of having a meta-description yet it hasn't fallen foul of Google indexing it and even having it show up in searches with reasonable competition if you select the right string to search on.

Originally Posted by Webnauts

When you abuse the meta tags, either you go to SI or from my experience you get punished with the -30 or -930 penalty. If you know what these penalties are, you will understand why you rank well again after you fix the problems.

I know what most of the -30 and -930 penalties are, they are attempts at providing an explanation when one has none.

Google, by their own admission does manually apply "penalties" but not as often as people would like to think because to accept that there is no actual penality in effect would be to have to accept the only other option, which is that an algo-shift occurred and a site now ranks where it should according to the new algo.

A -30/-950 penalty is convenient because then it is Google that is the one that did it and not the webmaster doing something or not doing something so the webmaster is off the hook and it is bad Google that messed up. But the odd thing is, while it is Google that placed the site's pages where they rank, it is what Google finds there, which is the responsibility of the webmaster, that put the page where it is.

That is not easy for many people to accept so thinking they are being "penalized" for something they didn't do is just a convenience.