Google counting only the first link to a domain – rebunked
OK, so not so long ago Rand put out a post saying that Google will count only one link to a URL from any given page. This is a phenomenon that was originally noticed by Michael VanDeMar but many an eyebrow was risen at both posts. That particular facial hair elevation trend has seemingly culminated by David Eaves’ post which supposedly debunks the theory. However, some problems were found in the way David has conducted his research (starting with the fact that it was done in google.co.uk which seems to behave like Yahoo did 10 years ago). Since we love to solve dilemmas around here and since argument from personal incredulity is not terribly popular in my neighborhood, I decided to conduct a separate test which will cover most of the questions raised in and around the above mentioned posts. (BTW the initial discussion is covered at SERoundtable, as usual)
So, in order to test the theory, i picked two sites that sometimes double as my furry lab animals and set them up so that site A links to site B with two links using different anchor texts. The phrases appeared only on a site A, they were not to be found anywhere on site B and if that site was to rank for any of those two phrases, it would be only due to the anchor text of the links pointing from the site A.
The testing process went like this:
- Setup the sites like described above and wait for Google indexing. After the links are indexed, check the rankings of the site B for the two phrases.Result: Site B ranked for the first phrase and not for the second phrase. The below image describes the situation:
- Switch between the position of links to site B. Now the previously second phrase appears above the previously first phrase on site A. Wait for Google to index the change and check the locations of B for the new phrases.Result: Site B dissapeared from the SERPs for the new second phrase (previously first) and appears for the new first phrase (previously second), like in the image below:
- Revert the situation to the initial state: Switch back between the phrases, wait for indexing and check the rankingsResult: Site B reappeared for the initially first phrase and dissapeared again for the initially second phrase, like in the image below:
OK, so to me this is a pretty solid data supporting Michael’s and Rand’s claims. The fact that the SERPs reacted to my changes in the order of links back and forth is kind of hard to debunk.
However, one question that kept popping up is whether nofollowing the first link will cause Google to index the second link even though it is placed lower in the code. So i performed an identical test to the one described above, just with the first link nofollowed throughout the test. The results are pretty interesting – Google did not change its behavior due to the nofollow attribute applied on the top link.
Even when i switched between the first and the second link (preserving the state where the top link is nofollowed), Google still counted only the top link and not the bottom one.
Besides the fact that nofollowing the top link will not get you around the Google ignoring the second link, it is interesting to see that Google will still count the top link even though it is nofollowed. Maybe it was due to the fact that the nofollowed link was the only link on the web using that phrase as an anchor text ? Who knows… sounds like a theory worth testing.
As for the general settings of the experiment, the keywords I have used belong to the “semi-promiscuous” kind, meaning that they are made up of words that mean something but as a combination are not commonly used. The phrase #1 had about 40 results and the phrase #2 had about 270 results in Google SERPS. It took about two to three weeks for Google to index each of the changes, meaning that each set of results persisted for that time. It also means that the testing for this post took about a month and a half. Hence the posting frequency 
As i write in every post describing an experiment, the conclusions should be taken with a grain of salt, Google is a black box that cannot be 100% scientifically tested (I am not talking about you Matt), i was not wearing my lucky underwear when performing the tests, etc.
PS. In the course of the described experiments, i did come up with a method that can help circumvent the described situation and force Google count all the links from a single page to a target domain. It was even independently confirmed by an SEO who knows what he is talking about (unlike me) and is being successfully used on some of our sites. However, I was asked not to blog about it, so if you want to know more and I know who you are, either personally or from Sphinn/Twitter/Plurk/etc., contact me through the form and I’ll drop you a hint 
Tags: first link, Google, nofollow, page design, SEO, sphinn, testing
86 Responses to “Google counting only the first link to a domain – rebunked”
Interesting results, I think this shows exactly the direction google is going with links.
By David Heine on Jul 3, 2008
Your test is about cross-domain links, but now I’m wondering about multiple internal links, too. I haven’t rigorously tested for this, but it seems like different anchor text in the content area (coming after the menu area links in the source code) can have an effect – it apparently can trigger negative ranking adjustments for keyword stuffing, going back as far as the Florida update.
By tedster on Jul 4, 2008
So that’s the secret, I need to wear my lucky underwear when performing my SEO test. It always felt fresh without any. *LOL*
By Bubsy SEO Challenge on Jul 4, 2008
What were the time frames of the tests between changes in page content? What are the crawl frequencies and cache update frequencies of the pages involved? Did you put control links on the page to show that all the links were being processed?
As described, your test is no more valid than the others.
By Michael Martinez on Jul 4, 2008
Definitely an experiment sheding a few more rays of light onto that phenomenon. Now, if you’d only know me enough to send me said hint
By Bert on Jul 4, 2008
thanks for all the response
@tedster thanks for stopping by. i am not sure i understand th internal linking test. You get penalized AS IF you were doing keyword stuffing or it expedites the keyword stuffing penalty if you already are keyword stuffing ? In any case it is definitely interesting to see the difference in behaviour between the cross-domain vs. internal linking.
@michael – the time frames are stated in the article – 2 to 3 weeks between each cache update – crawling happened every 2-3 days. I did indeed put control links and control text to verify that links were processed and that the site is ranking for the control text. As for your last sentence, I guess that there is only one way to conduct a valid test – being Michael Martinez. Some of the people can be wrong all the time and all the people can be wrong for some time, but can all the people be wrong all the time ?
@bert I am sorry dude, but it has happened more than once that techniques were abused and their benefit taken away by Google. however with some experimenting and creative thinking, I’m sure you’ll get on to it…
By Neyne on Jul 4, 2008
I did an experiment with an orphaned page on my site.
The page was completely isolated apart from half a dozen blog posts on nofollow blogs, using unique anchor text. The test is 3 months down the line and Google still hasnt indexed the page and the page will not come up for the anchor text used in said links.
That was enough proof for me (and a recent google declaration) that nofollow links were completely discarded from the site link profile.
Your test contradicts mine! Which is a pain in the a**!
By matt inertia on Jul 4, 2008
@matt hey, don’t take the nofollow conclusion as a done deal. In order to test that, I have to design a specific test with that in mind, trying to isolate nofollow as the sole difference between two links. My “conclusion” about nofollow is a byproduct of this test and should serve only as a question for further testing. There was a difference between your test and my test – i nofollowed a link to a page that previously had a followed link to it. It is also an offpage link and you are testing internal links, it is possible that nofollow is treated differently in these two cases…. anyways you get the idea…
By Neyne on Jul 4, 2008
Neyne – as to MM, ignore him, it’s safe to do so 9 times out of 10, and the times he’s not speaking out his ass it’s been said elsewhere anyways.
As to the nofollow, I think you’ll find that if you leave the links in place with the nofollow intact, Google will drop the destination page from that search shortly. I tested that exact thing 3 weeks later (search for [brogginoodle] to see the post I mean), and I can tell you that starting from scratch, if the first link is nofollowed then none of them have any effect, and that if a followed link points to the destination page and is then removed, there is a slight delay before Google drops that page from that search. It was an unintended side effect I ran into when MyBlogLog click tracking caused a followed link using that anchor text to get indexed after the post went popular. About a week after those followed links disappeared the page stopped ranking again.
By Michael VanDeMar on Jul 5, 2008
Hi Michael,
Thanks for stopping by. That is what i thought also and i debated whether to write about it in this post or to wait some more and then make a new post out of it. The truth is that this situation persists now for 3 weeks, but it is possible that due to different crawling rate, this will change soon. i am keeping my eye on it anyways…
What is really interesting is the fact that after i switched the links around, the ranking status also changed, even though the first link was nofollowed. So when i elevated the second link above the first one, removed the nofollow from the first (now second) link and applied nofollow to the second (now first) link, the ranking dissappeared for the followed and reappeared for the nofollowed link. I think that there is more to this than just the time factor…
By Neyne on Jul 5, 2008
Great work as always, Neyne!
We’ve been experimenting with ways to circumvent this as well, but I would love to hear more.
If I had a vote, I would crown you best up-and-coming SEO!
By Hugo on Jul 5, 2008
You might be able to use javascript, iframes, or some other wizardry to tell Google that the first link is not to be crawled… if that’s necessary. I’d be interested to hear which particular wizardry is working for you. Probably would get more links/sphinns if you drop it into the post
or at least the hint.
One thing I can say is I can’t imagine Google venturing to include more than one link to a domain in its link graph. Deciding which link is the best, least keyword stuff, most descriptive link is hard to do algorithmically.
I imagine as well that domain relevance could possibly shift the decision in one way or another, beyond a threshold of Google’s choosing. Let’s ask them. Google I mean…
-Fred
By just fred on Jul 6, 2008
sorry, meant to say “more than one link [per page] to a domain in its link graph”
By just fred on Jul 6, 2008
@michael martinez
1. You have posted a comment in a wrong blog post. Since moving it involves running several SQL queries on the database, I will first copy your comment here and then respond in a new comment:
By Neyne on Jul 6, 2008
OK, so now to my response:
- the queries as you entered them (with hyphens) do not give results in Google. I assume you meant without hyphens or possibly in quotes.
- when searching for either of the phrases in quotes, Google shows zero results and suggests to search without the quotes. So I am going to deal with unquoted queries
- when searching for [seo theories galore], Google gives two results from your blog, grouped at #1 and #2. Neither of those results are your homepage – the page you have linked to from xenite.org. Furthermore, both of those pages seem to have text which is relevant for the query, they have the word [galore], and there is the expression [seo theory] in the name of your blog, where [theory] is associated with [theories] and is marked in Google SERPs accordingly. Furthermore, none of your pages appear in SERPs for [seo tests galore]. Even when performing the above queries with the site: operator on your site, there are only those two pages showing for [seo theories galore] and none for [seo tests galore].
Now I am not sure what you were trying to prove here, but the results of your “test” show that Google will rank pages from your site that are relevant to your query and will not rank pages from your site that are not relevant to the other query. The page that you linked to does not appear in SERPs for either of the queries, so another finding is that Google does not consider anchor text when ranking pages ?
Besides the obvious conclusions that the test has helped us reach, there is another one I would like to point out: when performing anchor text tests, I suggest you do it for keyphrases that do not appear on the site that you are testing. Messes up with the sterility of the test.
I was expecting from someone that constantly looks down upon others to bring a bit more to the table, but as they say “the size of a disappointment is proportional to the size of an expectation”
Cheers.
By Neyne on Jul 6, 2008
Great to finally see some strong data on this instead of relying on theory. Great post.
By Paul @ Webdistortion on Jul 6, 2008
“the queries as you entered them (with hyphens) do not give results in Google. I assume you meant without hyphens or possibly in quotes.”
No, I provided the queries as I used them and when I click on the links I see results. I cannot guarantee that you will see results as Google is not very consistent.
“- when searching for either of the phrases in quotes, Google shows zero results and suggests to search without the quotes. So I am going to deal with unquoted queries”
That is because Google has not finished fully indexing everything. The indexing process is sort of staggered.
I’ll post some screen captures on SEO Theory.
By Michael Martinez on Jul 6, 2008
Okay, I have now posted the screen captures and explanation of what I did here.
By Michael Martinez on Jul 6, 2008
As I responded here, my opinion is that your test is fundamentally flawed. It does not prove your claim and it does not disprove mine. I stand by my result (which continues to be valid even now) that Google will not pass a second anchor. My testing sites pass all of the “supplemental page” conditions as you have defined them, even though I have not personally tested the difference between pages that are crawled less frequently, so i cannot even support your statement that there is a difference between the two or that there even is still such thing as supplemental index. In any case, you have not disproved my findings, and there is no new information you are providing, so we will finish this discussion. Thanks for your input, which was, although flawed, thought provoking.
By Neyne on Jul 7, 2008
Branko, this is very interesting, I like your method!
I wish you could perform more experiments faster, lol
By eric shannon on Jul 7, 2008
Other than “rebunked” not being a word (debunked is) the information you provided is solid evidence that Google does, include links, beyond the first one.
By Matt Keegan on Jul 7, 2008
What if the links are linked to different pages on Domain B?
Link one from Domain A: link to “page 1″ on Domain B
Link two from Domain A: link to “page 2″ on Domain B
If these are relevant, deep links, would they both carry value?
By Scott Thomas on Jul 7, 2008
Very nice write up. I would be interested in the same question that Thomas asked regarding different pages. We seem to have good results from something similar but have not conducted any real research.
By Arnie on Jul 8, 2008
@Matt Keegan – the English language completely mutates 66% every one thousands years quite consistently. New words have to start somewhere, there is no committee that controls their formation.
@Scott Thomas – yes, they both carry value in that case. No reason why they shouldn’t.
By Michael VanDeMar on Jul 8, 2008
This makes a lot of sense as google may see this as link overkill therefore it disregards the next link. It still probably counts on Yahoo and MSN. Easy solution is just to link to a different page on the second link
By search machine on Jul 8, 2008
Just yesterday i reviewed a website which had a second ranking position for a keyword. I mentioned the keyword in the post and used a nofollow on the site’s link.Today it has the first position for this keyword, this made me a bit confused about the “nofollow” value.
Great test by the way, one step towards resolving G’s secrets.
By edwinsdesignlab on Jul 8, 2008
Awesome experiment and test Branko – thanks for sharing this one with us! Interesting to see how the nofollow was treated too.
By Gab Goldenberg on Jul 9, 2008
Did you test what happens if the first link is a completely useless phrase like a “.” or a word like “here”, “click here” etc. ?
By Web Marketing Tourism on Jul 9, 2008
Last year I was checking the competition of a client.
I taped one phrase on Google, I found a website A that was Ranking on the first page but this website A has not the keyfrase on its content but there were many others website that were linking to website A with anchor text of the keyfrase i was checking.
Before I did not undertand it but now I do.
Thanks for your experimente.
By Rafael Montilla on Jul 14, 2008
Interesting test I was thinking about this the other day. When you say links to the same domain do you mean different pages on the same domain, or the same URL on the same domain?
By Dave on Jul 15, 2008
Great post, I would also like to know about internal links compared to external links.
By Bill Ross on Aug 20, 2008
Hello,
As this experiment was of great interest to me and i was still not 100% sure of its resutls – I conducted my own test, and can confirm that indeed only the 1st key phrase counts on google.
Check out http://articlesdepot.info/entry/20/ for more details.
By Ahoy on Aug 24, 2008
Hi, I found your blog on this new directory of WordPress Blogs at blackhatbootcamp.com/listofwordpressblogs. I dont know how your blog came up, must have been a typo, i duno. Anyways, I just clicked it and here I am. Your blog looks good. Have a nice day. James.
By James on Sep 19, 2008
Great post. Does this matter for links coming from external sites also, as far as only counting the first link that goes off site to a specific domain? And is it page specific or domain specific?
By questions on Sep 26, 2008
This is a great post. This is one of the few posts that tries to back up an opinion with some real evidence. Based on what I’ve read across about two dozen different sources, there’s more evidence that supports the idea that only the first link is counted.
I’m encountering this on a site I’m examining right now. It appears that its ranking is suffering due internal run-of-the-site links in a site-wide menu that uses poor anchor text. While I don’t have the liberty to run tests on this site, what I did find interesting is that pages that are not part of the menu are: 1) ranking better for their keywords and 2) have internal links with good anchor text elsewhere on the site.
By web designer on Oct 24, 2008
This is the best post on your site! I wonder if you link to different pages on the target site if both links will then be counted?
By Beauty on Nov 3, 2008
Ha! We did our own tests similar (exactly) to what you did when this whole thing surfaced and we came to the same results as you did. Except for the nofollow thing, we got very different results as I had speculated and as was noted by someone else I know. Google has an order of operations when crawling a page. First it crawls and removes links to the same URLs (for obvious crawling efficiency reasons), then it goes through and removes nofollowed links. So that being said, if you nofollow the first link in your code then none of the other duplicate links will get crawled.
Plus I also know about the work around and was advised by a commenter at SEOmoz to have my comment manually edited out of Rand’s post on this topic and to not let too many people on to it. It does work great and I have talked about it at a few local speaking engagements.
By Organic SEO Guy on Nov 18, 2008
Awsome post on this one. i love your illustrations! thanks for this addition.
By webdesign on Nov 26, 2008
You would figure that they would remove any value of other anchors on one page to the same target page, as that would be considered spam in my book.
At first, the nofollow part in the article was a little confusing to me. I figured by nofollowing the duplicate first link, it would just skip to the next one, but that was not the case. After seeing what Organic SEO Guy wrote about the behavior of the Googlebot, it made more sense as to why the other link was potentially not counted.
I am really starting to like these experiments
Keep them coming!
By wvo conversion on Dec 3, 2008
This post is one of the best posts I have read all year, no honest, however as I stated on a comment on another post you really should post more often. I take it you are 2 busy to write SEO articles which I can relate to, more soon!
By search engine optimisation on Dec 18, 2008
Great post. Thanks for sharing!
I also have something about SEO to share. Check out at my website please.879
By SEO on Dec 27, 2008
Thanks. It will stop spamming.
By seo on Jan 22, 2009
I was wondering about this myself. I was working on the assumption that they only count one, glad to of found this article.
Very scientific
By banner exchange on Feb 5, 2009
I love it when people do some research before posting. Thanks for the info. Keep it up
By Eric on Feb 11, 2009
im changed language of my site
of asp to html
waiting to actualization
lol!
By transportadora on Feb 12, 2009
Good post, I tested it as well, and you are correct: only the first link entry is counted for!
By Day Two Webdesign on Feb 15, 2009
Yes, Only first link one count but Google show all links from domain on Webmasters tool – Hemang
By SEO Company Singapore on Mar 20, 2009
This is an excellent article. You rarely see testing this thorough.
By Pozycje on Mar 25, 2009
Tny for tip.
By Handyzubehör on Apr 1, 2009
I have some doubts about Google only follow the first link of a domain in a page.
Maybe if you use a same anchor, Google can only follow one link, but I think if you set diferent anchors it will follow all links.
Great article, intersting view!
Thanks.
By otimização seo on Apr 3, 2009
Very interesting experiment about something I’ve been wondering about for a while now. I’ve been experimenting with Squidoo and have been narked to see that every lens I create has got 100 or so links in addition to the 2 or 3 I’ve created. I’ve wondered whether to create several links to each of the pages I’m targetting just to increase the percentage of links going to my site. By the results of your experiment, I guess Google would still ignore all of my duplicate links?
By John on Apr 6, 2009
This is really a great Information about seo . Thanks for such a good information.
By website design on Apr 24, 2009
Thanks for the information on website links, but does the same apply if the link destination remains the same while the link text is varied?
By website builders on May 9, 2009
nice to see some science, and not just spam and black hat.
By mirus seo on May 23, 2009
The very good post
the congratulations of this blog .. I will return several times
to check the content!
By Moda on May 26, 2009
Thanks for this great article. I will definitely come back for more of these interesting stuffs.
By Website maken on Jul 3, 2009