homepage Welcome to WebmasterWorld Guest from 54.211.7.174
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

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

This 87 message thread spans 3 pages: < < 87 ( 1 2 [3]     
Experiments in keyword rich links to Home
CainIV




msg:3756887
 6:21 pm on Oct 1, 2008 (gmt 0)

Hi guys, I am doing an interesting experiment on two of my more throw away domains. The experiment is testing to try and determine more information about how linking to the homepage affects rankings. The testing involves various controls - linking to the root domain from the nav only using 'home', linking from the nav using 'main keyword', linking from nav using 'variations' of keyword, linking from content only (while nav links saying home) to home using 'keywords' etc, etc.

First, I should mention some points about the domain.

4 years old
Owned by me
Dedicated IP
Canonical comdomized
HTML only
Ranks top 5 in Google.com for main, second and third keyword phrases.
Total of 90 pages, all unique content (written by me)

Testing was done over a 3 month period, with grace periods in between testing.

Here is so far what I have found. Might tell us a little about the threshold and re ranking filters

1. Linking home from every page in content using the same keyword caused 6 page drop in rankings.

2. Linking home using keyword in nav on all pages caused the same drop.

3. Link home from every page in content using variations caused a 3 page drop.

4. Linking home from the first 10 pages listed in google.com for site:domain.com/* brought increased ranking (from 5th to 3rd)

What is really interesting is that I gotten this down to the 'by page' factor. When I *slightly* cross the threshold and add links to two extra pages, and then wait until they are cached, I tip the scales and drop, to page 6.

What is further interesting is that linking home from content using variations of keywords WAS quite effective to a point, after which the site plummeted.

As well, this might point to a 'hard line' being crossed in terms of threshold, at one point I had the website going between position 4 and 51-60 for the same keyword every second day (flipping back and forth)

My test will be about trying to -950 the website by being ridiculously deliberate in nav linking, and then seeing if I can reverse the results by removing those (and how long it takes for the trust to be reinstated to the website)

Any thoughts?

 

Marcia




msg:3803125
 3:05 am on Dec 9, 2008 (gmt 0)

There's specific mention made in several published patents of both anchor text and changes, but here we have what Jake found on the topic back in 2004:

[webmasterworld.com...]

doughayman




msg:3803734
 9:07 pm on Dec 9, 2008 (gmt 0)

This is all very, very interesting. I have a fairly thin site that had been ranking consistently on Page 1 for many terms, for many years. I do a lot of internal linking, not in an attempt to fool Google, but in an attempt to make my site more user friendly for its users. I recently added some pages that had links back to my home page, and the site dropped to page 6 for the effected anchor text keywords. I too, am going to conduct a test:

1) First, I'm going to rel="nofollow" all these new links back to
my home page;

2) If # 1 doesn't get back in the good graces of Google, I'm going
to remove the links on these pages back to my home page.

If # 1 doesn't work, but # 2 does, will I sacrifice the usability of my website to positively effect rankings ? The big question....

Note: a thread about the rel=nofollow attribute has been split
off. Please continue any discussion of that second topic here:

PR Scupting and rel=nofollow [webmasterworld.com]

[edited by: tedster at 9:47 pm (utc) on Dec. 10, 2008]

tedster




msg:3803831
 11:36 pm on Dec 9, 2008 (gmt 0)

You wouldn't name any nav to the home page anything different: That is the point.

Hence the value of the testing that CainIV is doing. A site that is suddenly flooded with keyword anchor text pointing to Home is a site that is being heavily optimized. But occasional variations in Home anchor text could happen for a number of non-manipulative reasons.

So where is that dividing line?

[edited by: tedster at 10:08 am (utc) on Dec. 10, 2008]

tootricky




msg:3803835
 11:38 pm on Dec 9, 2008 (gmt 0)

The dividing line is from page 1 to page 6!

I am of the opinion from my own drop (and a bit of fiddling afterwards)that in page links are less likely to be penalised as those in the side panels, footer (and perhaps nav). It is highly likely that Google uses some form of VIPS to make judgements on link value and relevance, so where the link is on the page, is as important as what text it uses to link with.

whitenight




msg:3803928
 2:13 am on Dec 10, 2008 (gmt 0)


Hence the value of the testing that CainIV is doing. A site that is suddenly flooded with keyword anchor text pointing to Home is a site that is being heavily optimized. But occasional variations in Home anchor text could happen for a number of non-manipulative reasons.

Manipulative?! Are we going back to "black" and "white" "evil" and "good" SEO?!
There's what works and what doesn't. Period
That INCLUDES what the algo (or individual reviewers) won't penalize for and what they will.

Btw - who said anything about "flooding keyword anchor texts" to the home page?!

Which leads to the MOST important question that has still yet to be answered...

Has ANYONE (except me) actually DONE Cain's test as given?

If you don't understand WHY I keep harping on this point, and why is has LITTLE to do with:
Past patents,
Past experiences with rel=nofollow,
past experiences with similar but different tests people have done on their sites,

then I give up.

[edited by: tedster at 10:07 am (utc) on Dec. 10, 2008]

tedster




msg:3804015
 5:33 am on Dec 10, 2008 (gmt 0)

Manipulative?! Are we going back to "black" and "white" "evil" and "good" SEO?!

Nope, just acknowledging that Google sees some optimization acitivities as clarifying and others as manipulative and worth a smack on the wrist.

who said anything about "flooding keyword anchor texts" to the home page?!

CainIV did, in the opening post. Reports 1 and 2 were about adding keyworded home links to EVERY page.

From CainIV's opening post:
What is further interesting is that linking home from content using variations of keywords WAS quite effective to a point, after which the site plummeted.

As well, this might point to a 'hard line' being crossed in terms of threshold, at one point I had the website going between position 4 and 51-60 for the same keyword every second day (flipping back and forth)

My test will be about trying to -950 the website by being ridiculously deliberate in nav linking, and then seeing if I can reverse the results by removing those (and how long it takes for the trust to be reinstated to the website)

That "hard line" would be interesting to pin down - is it a percentage, or a specific number? Would it be different for different anchor text? Lots of potential test to look into here, and they would potentially gobble up lots of domains.

The question of reversing out of a -950 is interesting. I have only one case that I can report on. The site triggered a -950 by adding two different keywords for home links in the nav area - splitting those keywords 50/50 between the pages. They made the change all at once, and reversed the change all at once (they needed the income and regretted testing on that domain), but rankings did not come back for about 3 months.

[edited by: tedster at 10:06 am (utc) on Dec. 10, 2008]

CainIV




msg:3804019
 6:01 am on Dec 10, 2008 (gmt 0)

sometimes the point of experiments is to experiment :)

I think that all I really did prove to myself was that there are certainly nav and footer link thresholds that cause a website - even a website into the top 5 - to spin out and down to pages 5 and 6 in ranking.

Another element I seemed to reveal was that Google appears to handle thresholds from links within content different than those within nav or footer sections - this is to say, that content links have a fuzzier 'trust' logic about them, and I could more links to content, faster and more focused on the primary keyword pointed at the root url without the same threshold being crossed.

I am sure many would have deduced this point by logic alone, and I am sure there is much more theory I could derive from this experiment.

But, alas now I am curious about nofollow as well, yet I am still at a loss as well as to how one can measure whether trust is passed through that type of link.

I do have other domains I could test this out on that have pagerank.

[edited by: tedster at 10:05 am (utc) on Dec. 10, 2008]

whitenight




msg:3804020
 6:02 am on Dec 10, 2008 (gmt 0)

Nope, just acknowledging that Google sees some optimization acitivities as clarifying and others as manipulative and worth a smack on the wrist.

After the many years fighting against MORALIZING SEO tactics, I'm simply tired of any "judgment" values placed on techniques, esp tests.
Goog does enough "Thou shall not"-ing under the guise of "morality" to last us all.

"It works well/sometimes/not at all"
or
"It doesn't work well/sometimes/not at all"

applies just as well without placing Goog's religious-like value judgements (aka FUD-pronouncemnts) that seem to PREVENT SEO's from testing and analyzing JUST ANOTHER SITE
that so happens to be named Google.com

That "hard line" would be interesting to pin down - is it a percentage, or a specific number? Would it be different for different anchor text? Lots of potential test to look into here, and they would potentially gobble up lots of domains.

Gee, how many domains? Wish someone had posted earlier some advice about the right amount of domains for testing... -.-

[edited by: tedster at 10:05 am (utc) on Dec. 10, 2008]

tedster




msg:3804027
 6:48 am on Dec 10, 2008 (gmt 0)

under the guise of "morality"

No moralizing here. There is one area where I feel the words "ethics" or "morals" apply, and that's in client communications. Talking a client into a very risky change without letting them know that you consider it high-risk is not ethical. But it's Google who decides what actions are risky, and deciding to probe the shapes of those hidden decisions is neither ethical or unethical - it's just practical.

The whole "over-optimization" penalty (-950) really messed with any "moral high ground" that Google hoped to portray. What is "enough" and what is "too much"? Google's not going to tell is where the line is, and so we are in an area of FUD. Unless you are content with the results you already have, then only testing can show you where the line is currently being drawn.

With regard to home page links, I built a site a year ago and launched it with a keyword home link in the main navigation, and a link that said "Home" as well. The site has only improved in ranking for that kw, and now 14 months after launch is stable at #2, with never a wobble.

I have a more recent site that was launched in the smae fashion, but it's too early to tell. And several older ones that all showed no ill effects from having the keywords there at launch. Funny thing here, adding a link on an established site that says "Home" for the user's sake has never caused a problem. Well, maybe it did for the kw "Home" - I never watched for that, LOL.

But if I'm working with an established site, I would only consider adding keyworded home links to the content area, never sitewide and not in the footer, either. And the best bet I think is only adding such links in the content area of new pages. Everytime I went into the nav area or the footer or sitewide fpor the additions, the rankings for that keyword took a drop (or in a few cases, showed no change.) The size of the drop has varied over time, but it always left the first page, and any drop is a drop.

[edited by: tedster at 10:04 am (utc) on Dec. 10, 2008]

whitenight




msg:3804043
 7:18 am on Dec 10, 2008 (gmt 0)

ok tedster,

let's use some mildly hypothetical numbers here.

Let's say you test something on a minimum 30 or so sites and it works on all 30 sites.
Better yet, let's say it's 29 out of 30 sites.
Statistically speaking that's a 97% "favorable rating" with a +/- 3.x% margin of error over nearly every single top 10 page available on Google.

94% - 100% "favorable rating" on any SEO technique is basically gold.
And SHOULD give you confidence (NOT FEAR) to test it on any big "corporate" client, no?

Obviously, as a corporate SEO, every statement you make should start with
"I have no 100% guarantee if it's going to work BUT..."
That goes for ANY changes one makes to the site, no matter how "fool-proof"

At a certain point, YOU, as the knowledgeable SEO who HAS tested something like this (on their many "tests sites") has to just say,
"This is what we are doing. It works. I've TESTED it enough to my satisfaction".
But the only way one can make a statement like that to a corporate client is if one KNOWS it works, right?

-------

As Cain has said MANY times, this test as given incorporates 'fuzzy logic' vs. any and every test, patent, theory on anchor texts, OOP, etc to date.

Any past experiences with those, in comparison, to how THIS test is done, is basically null and void (in MY opinion)

Shaddows




msg:3804106
 10:08 am on Dec 10, 2008 (gmt 0)

Did I bring nofollow into this? Sorry :) However, what I really, really wanted to know should have been clear from my suggested test methodology (I think Cain gets it at least). That is- can you change Gs interpretation of sitewide navs (esp footers) by gradiated change, using nofollow as an intermediate step.

Some comments/observations on the rest of this thread, so a bit off-topic but on-thread.
Black/White/Blue Hat. Come on people, you should all know better than this. In fact, I'm reasonably sure you all do, but just think everyone else doesn't.

Google does NOT own the internet. But they do own Google. Google writes the Google Algo. Google algo spiders and sorts the net. People use Google.

Now, no morality here. But if you want to play in Googles back yard, you need to follow their rules. NOT THEIR PUBLIC STATEMENTS- THEIR ALGORITHMICALLY DERVIVED RULES. If you do something that Google doesnt want you to do (AND CURRENTLY HAS WAYS OF DETECTING), a nice short hand is BLACK HAT. Its not moral. Its shorthand.

If you do stuff that works for G, G does not penalise, and is good for users, you are WHITE HAT. Its low risk, and probably converts well too. Again, its shorthand, not a moral point.

Reasonably consistent shorthand for 'it CURRENTLY works, but doesnt benefit users' is GREY HAT. G might penalise it in the future, they might not. It might not pass a manual review. Its medium risk.

The problems with this is that almost everyone who posts in the monthly threads with "my site has a penalty" declares themselves "white hat". They are using the shorthand as described above, but they are often wrong. I.e. They THINK they are white hat, but arent. Maybe they have used descriptive anchor text for internal links. Maybe they have a bunch of published articles using identical anchor text inbound. Whatever, they are doing something G does punish, they are just ignorant of it. Or, the algo knobs are tweaked and they are insufficiently diversified and rely on something that has just been turned down.

Now, I'd advocate a new thread on SEO verbiage and nofollow experience, but this thread has been cleaned up once already, and from experience I doubt the new thread would be as... well contended as this one.

Note - two threads have been split out of this discussion:
PR scuplting and the rel-nofollow attribute [webmasterworld.com]
Testing methodologies - learning how Google is actually working [webmasterworld.com]

[edited by: tedster at 9:45 pm (utc) on Dec. 10, 2008]

whitenight




msg:3804110
 10:13 am on Dec 10, 2008 (gmt 0)

Thank you Shaddows. Very well said my friend.

tootricky




msg:3805091
 1:04 pm on Dec 11, 2008 (gmt 0)

Anyone got 30 sites they can lend me? ;)

tootricky




msg:3810408
 2:29 pm on Dec 18, 2008 (gmt 0)

I have a 1000 page site that I will be testing something like this on. I have 6 keywords that the home page ranks for and will design a rotating footer that shows one of the 6 keyword rich links on each and every page of the site.

I have a feeling that this amount of internal linking to the home page will get the keywords smacked, but 6 keywords is as good a number to start as any.

diddlydazz




msg:3810626
 7:40 pm on Dec 18, 2008 (gmt 0)

I have been running tests on 3 new domains (competitive phrases)

Site A: Footer KR-LTH added after being fully indexed and ranking (top 30)
Site B: Footer KR-LTH present from the start
Site C: KR-LTH from content present from the start

Will post here when I have more data, although so far it seems to tally with what CainIV has experienced

CainIV

Thanks for posting this, I forgot how much fun experimenting can be ;0)

p.s I am keeping my eye on MSN and Yahoo too

[added]experiment started 5th november[/added]

CainIV




msg:3810638
 7:58 pm on Dec 18, 2008 (gmt 0)

Diddlydazz - thanks for helping me quantify some numbers and arrangements. I think these types of experiments shed further light, especially as it relates to thresholds.

I won't revisit morality / BH / WH too much except to say that both Tedster and Whiteknight are correct. It is important to both know what Google deems manipulative, and at the same time, have general guidelines in terms of your own personal preferences (in the case of your own domains) and a level of SEO that you apply for clients, if you serve any.

Let's face it, there are tons of loopholes, not only in Google.com but in life. Rules are meant to be broken by the most creative. Those who break rules and find lasting success are called geniuses.

My experiments are tactic-neutral. I don't care to evaluate them for morality, no more than a child evaluates the WAY in which they play with toys. You choose how to apply the information you can derive from this post.

I don't know about you guys, but when a big red Tonka Toy firetruck is put in front of me, I just wanna get out my screwdriver and take it apart. Isn't life grand? :)

diddlydazz




msg:3810658
 8:19 pm on Dec 18, 2008 (gmt 0)

especially as it relates to thresholds

There be gold in them there hills ;o)

I just wanna get out my screwdriver and take it apart

I hear ya, I think I may have been born with a screwdriver in hand.

pageoneresults




msg:3810683
 8:48 pm on Dec 18, 2008 (gmt 0)

Those who break rules and find lasting success are called geniuses.

Depends on how you define success. I know a few broke arse geniuses. ;)

Personally? I don't think you can do too much messing around with that "Home Page" link within primary navigation elements.

5 Comprehension - 5.1 Writing style
[w3.org...]

Strive for clear and accurate headings and link descriptions. This includes using link phrases that are terse and that make sense when read out of context or as part of a series of links

While all this "keyword rich" dabbling may be having "some effect" SEO wise, how does it affect the visitor? I mean, you're talking about changing the visible text that users since the dawn of the Internet are used to seeing: Home Page or Home. I think that is playing with fire from a user perspective, let's not forget about them, or us.

There are certain things that most users have come to expect in most instances. That is primary navigation and the anchor text associated with it. I'm not expecting a 3, 4, 5 keyword rich link that leads back to the home page. Not from primary navigation anyway, that's messing with the wrong part of the equation if you ask me. That is where your internal "inline linking" comes into play along with external links. You don't want that stuff replicated across your own internal sitewide links, I wouldn't think so.

I see three links to the home page...

1. Header
2. Footer (text navigation)
3. Copyright (absolute bottom visually)

The header is typically an image link using both a title attribute on the a href and an alt attribute on the img.

The footer is typically anchor text that reads Home or Home Page depending on how hokey you want to be. I think Home is fine, that is "universal" language. A keyword rich text link is not.

The copyright notice is a great place to use a sitewide keyword rich link. But only if your company name contains those keywords. :)

No, there are more refined ways to link to one's home page using targeted anchor text. Internal navigation is not one of them, not for the home page.

cheesy snacks




msg:3810834
 12:38 am on Dec 19, 2008 (gmt 0)

The footer is typically anchor text that reads Home or Home Page depending on how hokey you want to be. I think Home is fine, that is "universal" language. A keyword rich text link is not.

The copyright notice is a great place to use a sitewide keyword rich link. But only if your company name contains those keywords. :)

No, there are more refined ways to link to one's home page using targeted anchor text. Internal navigation is not one of them, not for the home page.

Excellent post. Basically are we saying it is blatantly obvious to google that if you are using your keywords in your 'home' links from internal pages you are more likely or not attempting to manipulate the search, and that excessive use of these keywords could trigger a filter.

Whereas using the keyword 'home' or 'homepage' is completely fine in the eyes of google as that is what has been as the standard since the internet began.

gouri




msg:3814405
 11:06 pm on Dec 25, 2008 (gmt 0)

This thread has provided some very good information regarding linking to the homepage.

I was wondering if you could maybe give me your feedback on this.

a) Linking home too many time poses a threshold penalty, especially when this element is changed and historically was not present at onset.

b) Removing or adding these types of links can trigger a performance hit on keywords for the domain.

At the moment on one of my sites I have www.buildingcoolwidgets.tld in the footer, which has been there for some time and is something I think many people have in their website's footer, the domain name linking to the homepage. In my header I have written Building Cool Widgets as text only. On every page it shows Building Cool Widgets. If I made Building Cool Widgets into anchor text that linked to my homepage from every page on the site, would that be considered too much linking to the homepage as mentioned in statement (a)? It would be a new addition. Also, could it help to rank for Building Cool Widgets or Cool Widgets or would it create a situation mentioned in (b)?

Also, in the navigation menu, I have about nine links. Most of them have anchor text that describes what the page that it links to is about. For the homepage, however, I have the link as Introduction. Would making the link Building Cool Widgets (Introduction) be something that I can do or would it not be good use of keywords in the navigation menu? I am not trying to change the anchor text of all the links in the navigation menu, just one of them. Maybe this could help in the rankings for Building Cool Widgets. All pages link to each other so that would be several links that has Building Cool Widgets in the anchor text.

[edited by: Robert_Charlton at 11:47 pm (utc) on Dec. 25, 2008]
[edit reason] changed .com in example to .tld [/edit]

tedster




msg:3814901
 4:12 pm on Dec 27, 2008 (gmt 0)

It sounds like you are close to the threshold that CainIV is testing for. No one knows exactly where the "too much" border line is set, or how it changes over time. But it does sound like you might be walking the tightwire here.

If this isn't a domain that you can afford to have trouble with, then I'd suggest you move on to other kinds of work for a while. On-page tweaks and in-site anchor text can only take things just so far. On the other hand, if you can afford to experiment a bit more, you may learn something useful about just where the threshold is for your site.

gouri




msg:3814925
 6:37 pm on Dec 27, 2008 (gmt 0)

I added a link to a homepage from an inner page using anchor text that the homepage is trying to rank for. I noticed a decrease in my ranking for the keyword. Then I changed it and the inner page has cached the page with the version that does not include the anchor text.

The rankings for some long-tail keywords whose ranking went down went up but for the keyword that the homepage is targeting the rankings have not moved up. Is it possible that there is some sort of penalty when the anchor text link was added to the homepage from the inner page and the penalty is still there even though the cache version of the inner page without the anchor text has been cached?

tedster




msg:3814927
 6:53 pm on Dec 27, 2008 (gmt 0)

That is quite suggestive, even for just a single case. How far down did rankings drop? Were the higher rankings in the top ten?

gouri




msg:3814928
 7:00 pm on Dec 27, 2008 (gmt 0)

The rankings dropped about 20 positions. It was around 50 before the drop and around 70 after. The homepage actually targets two keywords and for the other one also there was about a 20 position drop.

This made me think that there is some sort of penalty because both keywords went down, and they went down around the same amount.

Gemini23




msg:3814956
 8:49 pm on Dec 27, 2008 (gmt 0)

Interesting thread... what about if the link to the homepage is by the main website homepage url being on each page. ie.. in the position of a 'footer'. Does that count as being detrimental? or can it be seen as just 'advertising'?

tedster




msg:3815145
 4:57 pm on Dec 28, 2008 (gmt 0)

I doubt that would be a cause for a penalty, Gemini. The issue is not with the url, it's with the keyword anchor text. Footer links are common on many sites for a usability feature after scrolling. Without any keyword stuffing issue, there's not a problem.

For gouri's situation, it's hard to say what's going on when the rankings involved are lower than page three. There's so much shuffling in that area, that it's not easy to call it with any certainty. I suspect Google does all kinds of statistical testing on those pages, and that testing can create ranking churn without any penalty being involved. However, seeing two keywords drop the same amount at the same time does suggest a penalty.

Mostly, I'd think the home page needs more PR to do well - off-site help, rather than on-site or on-page tweaks.

gouri




msg:3815231
 9:18 pm on Dec 28, 2008 (gmt 0)

I agree, I think I have to focus on building more backlinks. One thing that I think I could do on-site is make the header into a link instead of just plain text.

This 87 message thread spans 3 pages: < < 87 ( 1 2 [3]
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