Welcome to our first Whiteboard Friday of the new year. It's 2012 and we're going to kick it off by examining the intricacies that revolve around anchor text. Although, this may seem like a very basic topic, we are going to cover some lesser known aspects of anchor text that is sure to satisfy even our more advanced SEOs. Enjoy and don't forget to leave your comments below!
Video Transcription
Howdy, SEOmoz fans. Welcome to another edition of Whiteboard Friday. Happy New Year. This is the first Whiteboard Friday of 2012, and today we're talking about anchor text, which could seem like a basic topic. But, in fact, there are a lot of intricacies that we should cover. Let's get right to them.
What I have drawn here is a web page, and it says, "I just found this great website on Portuguese cooks. You should check it out." Now, this, this text in blue with the underline, that links somewhere, and that link points to another page. Let's say it's a page over here, a very nice page on Portuguese cooks. It has some pictures on it. I don't know what it's got.
What it's saying to the engines is not only eye this page and this website, I'm voting for this other page over here, and I want to pass over some PageRank and link juice. I want to pass over trust. I want to pass over the domain diversity, whatever the signals, the keyword agnostics signals are, but I also want to say that I particularly like this web page about Portuguese cooks. That's what I think you, search engine, should interpret and take away from it.
Of course, this anchor text with the keyword embedded in it becomes a very strong signal to search engines, and as we all know, this is one of the strongest signals that Google and Bing interpret, Bing maybe even stronger than Google. Because of this, lots of people go down a path of trying to acquire links that say the precise keyword that they want.
Of course, this is a challenge because most natural links on the Web don't generally do this. They will say things like your brand name. They might say something about your site. They might use your personal name, if they're linking to a blog or something. But it's rare, it's uncommon that they might say "Audi 87 engine parts for sale" or "best deals on holiday gifts." These types of anchor texts, the things that people search for, longer phrases, in particular, are very hard to get as natural links, and this is one of the biggest reasons that gray and black hat SEO exist because manipulating the search engines by acquiring lots of links that have these keyword matches pointing to your page can, in fact, do a great job of ranking you up, at least temporarily until the engines catch up and do something bad to you or to the people linking to you.
What I want to cover is some intricacies around this, some details that you may or may not know about anchor text, and those include: Number one, multiple anchors from the same page "do not" provide more value. What I mean by this is if this page said I just found this great website on Portuguese cooks, you should check it out and a bunch of other text, and then it said Portuguese cooks again and linked over to this page, not helpful. It does not add additional value. There is no reason that you should be going, "Oh man, I wish I could get four anchor text match links from this web page." No, that's not going to help you.
Multiple web pages will help you, but if they're from the same domain, that's not nearly as valuable as if they're from different domains. That leads us to the next thing, diversity of anchor text, diversity of the source. The root domain source of the anchor text links provides the strongest benefit, meaning if you can get lots and lots of websites, not just individual web pages but different unique web domains, linking to you saying "Portuguese cooks," chances are good this web page will do very well.
Number three, the fluctuating anchor text. This is something that people talk about all the time. They don't just talk about diversity of the link location across different domains, but they talk about diversity of anchor text itself, meaning, "Oh, I should have one that says Portuguese cooks and one that says Portuguese cooking and one that says cooks from Portugal. I'm going to vary up the anchor text a lot."
I'm a little skeptical about this, not because it's not potentially useful, and it should be a natural thing if you're going out and doing white hat types of link building and inbound marketing. But because the primary reason I think most SEOs do this is so as to not trigger pattern matching problems in the engines, meaning if every website that's linking to me says Portuguese cooks, that's suspicious, highly suspicious. That suspiciousness is the feature that people are trying to prevent.
So, I'm not so sure whether this fluctuation is all that important unless you're doing manipulative types of link building, in which case SEOmoz is not all that helpful for you. So, you're probably not watching this video.
Number four, the first anchor text in the HTML of a page is what Google counts, Bing as well. This was discovered on SEOmoz a couple of years ago. We ran some tests about it. We published the results. There was a lot of skepticism. I think Debra Mastaler from Alliance-Link wrote about it and said, "Hey, Matt Cutts, would you please confirm this?" And he did. He came out and said, "Yeah, that's how we interpret it".
So, basically, here's what's going on. If you see a web page and it says this website is awesome, it features highlights of great Portuguese cooks, now look, these two links are both pointing to the same page. I don't know why my handwriting is so terrible in 2012. I hope that repairs itself soon. That means not that the website is going to get credit for the anchor text website and the anchor text Portuguese cooks, but rather they are going to consider the anchor text website and ignore Portuguese cooks.
It's very frustrating, and something that you should think about when you're doing internal linking and you say, "Oh, yeah, we should optimize this link." If it's already in your menu, if it's already at the top of the page somewhere in a side bar and that's higher up in the HTML code, then that is what the engine is going to count. So, do be aware of that and same goes for anything that you're earning externally. If you've got the optimized anchor text for your website in the footer of the blog post where it talks about the author, Rand Fishkin is the CEO of SEOmoz, an SEO tools company, but I've already link to SEOmoz's home page somewhere in the blog post above, that "SEO tools company," that's not going to help anything. That's going to be discounted by the engines.
Number five, internal anchor text, meaning anchor text that comes from your own site, your own pages, it does help. It helps a tiny bit. You can see a little bit of benefit from that. I wouldn't focus on it too much because tiny is a small amount. That's probably the most obvious statement I've ever made on Whiteboard Friday. But nevertheless, tiny, small amount, therefore don't focus too much energy on this. Link naturally, internally. Link in such a way that people think your site is good, and, yeah, if you can work in your anchor text, great.
External anchor text is where it really helps, meaning websites that are not your own linking to you. That's where you really get value from anchor text, and you do need to worry about this a little bit. There should be some manual efforts, some efforts, whether that's guest posting and blogging, whether that's sponsoring an event, whether that's getting your biography featured or something like that, getting a badge embedded somewhere or a graphic embedded somewhere that links back to you in a certain way, you do need that anchor text link match. So, working on at least a little of that external anchor text is definitely worthwhile.
Number six, if a link uses an image, like this, so check out this awesome site on Portuguese cooks, and then here's a little screen shot of the Portuguese cooks website, and this is linking over. I tried to illustrate that in blue. This does not have any anchor text. It's an image. So what could the anchor text possibly be?
The answer is they use the Alt attribute. The engines use the Alt attribute that becomes the anchor text usually, not always. If there is no Alt attribute, sometimes they'll use something like the surrounding text, and you can sort of see and feel that association. Sometimes, they'll use page titles. Sometimes, they won't use anything, but they'll have weaker signals from those other areas of the page, that kind of thing.
If you are embedding images and you're linking back to yourself or you're getting links from somewhere or you're linking out to someone, you want to help them out, use good Alt attributes that describe the page that you're linking to. This is a great best practice just in general for screen readers and usability reasons. It's also good for search engines.
Then finally, number seven, no surprise, surrounding text can matter as well. Just as in this example where we said, "Hey, Portuguese cooks is mentioned right before the image," the engines may be using surrounding text of an anchor, particularly where the anchor itself doesn't have much value or context.
If something says, "Click here, you'll find some great information about Portuguese cooks," the engines might sort of glance around the page and look at the sentence, parse the paragraph, try and understand, "Hey, what do you think they're talking about here? What seems relevant?" This is one of the reasons why you can see that people who have earned not necessarily great anchor text can rank very well for keywords because it's often talked about. That topic is talked about when their website is talked about, and it becomes a brand association thing. It becomes a contextual association thing. This is a helpful thing to think about if you are earning links and you can't control the anchor text. Maybe, at least, you can get them to mention what you do somewhere near the link.
All right, everyone. I hope this edition of Whiteboard Friday has been helpful. I look forward to discussing more details about anchor text in the comments and hope to see you again all next week for another edition of Whiteboard Friday.
Happy New Year! Take care.
Is it correct to state that the sole exception to "first anchor text counts", is when you use named anchors deeper in the page (e.g. #cooks) where you use different anchor text to those anchors? How does that count in terms of importance?
I remember a study published on SEOmoz looking deeper into anchor names some time after you discovered that the first anchor text counts.
To be honest, I would be skeptical about giving an answer with 100% certainty on that topic. I think it's something that's long overdue for another test (or a few). Bing and Google may have changed their behavior around this, particularly as named anchors are now so often used for AJAX-type calls.
Yes, named anchors are indeed popular nowadays because of AJAX. I'm looking forward to another groundbreaking test to discover the current state.
I have clients with '1-page branded campaign websites' with lots of named anchors that structure the page. These websites are conceptually developped elsewhere, so I have to work with what I get.
'Only the first anchor text counts and the rest are discared' in the instance of a page linking out to another page using multiple anchor texts is a myth. How much weight hypertext relevance should pass depends upon number of factors as outlined in many Google Patents like:Phrase-based indexing in an information retrieval system.
One important factor is the relationship between pages and the sites on which the pages are hosted. So if you own multiple sites or are an affilate or partner of another site and link out to each other, then the weight the hypertext relevance will carry will be calculated differently. The weight can be more or less depending upon the strength of the relationship. I personally believe that the weight of the hyper text relevance deteriorates as the relationship between pages and sites become stronger even for the 'first anchor text'. One good example is side wide linking. Such type of linking creates a signal of very strong relationship between the two websites. Google then try to determine the nature of this relationship. Is it genuine or is it manipulative then calculate the hypertext relvance weight accrodingly.
Another important factor is how the anchor text is being used on a page. Google is pretty smart in determining semantic connectivity between words and phrases used on a page, set of pages, between linking pages/sites. Which means the weight of the anchor text relevance has also something to do with semantic connectivity between linking pages. So your first anchor text relevance may carry less or no weight if there is little or no semantic connectivity between linking pages. So for example you have a page which talks about 'dog as a pet' and you link out to a page of another website using multiple anchor texts:
1st anchor text: Tesco
2nd anchor text: How to pet a dog
3rd anchor text: Tesco Direct
1st and 3rd anchor text have weak semantic connections with the surrouding text and the linked to page. 2nd anchor text has strong semantic connection and therefore carry more weight. That is why a link from a topical community which contains your targeted keyword in the anchor text carries more weight than a link from a non-relevant website.
Great write-up! I didn't know this.
Have you done any tests that proof this thesis? Any links to such tests? I'd be interested to learn more about this.
I also believe the same that ‘Google picks and counts only the first anchor text’ is a Myth and my small experiments have also thought me the same. In my Opinion Google give relevancy to that link keeping the semantic analysis as one of the factor.
Very good points, Himanshu.
Matt did not confirm the "first link" hypothesis, in the comments to Debra on the link spiel post:
1st comment from Matt:
"Dudibob, no, I confirmed the converse: if the anchortext is the same, we’ll typically drop the second link.
This is the sort of thing where people can run experiments to see whether different anchortexts flow in various ways."
2nd comment from Matt:
"Your question is short, but the answer is more complex. Typically if the anchortext on the two links is identical, we would probably drop one of those links."
I've seen some discussion in search-patents on link mergers when there are links or groups of liks that are repeated on a page, and how they might be merged, and at least one of those (from microsoft, I believe) seemed to hint that they would record the different anchor texts used if they did actually differ.
But the factors that you raise, such as how "related," anchor text might be, from a co-occurrence stance under phrase based indexing, is an issue that I didn't see raised in testing of multiple links on the same page. The features involving a "reasonable surfer" weighting of links should also be considered.
A new children patent application based originally on the Historical Data patent was published just couple of days ago, and it describes how changes to anchor text pointing to a page over time might influence how much of an effect that anchor text might have on the ranking of a page. For instance, if the anchor text changes so that it is no longer relevant at all for the page, the link may no longer carry any weight. From the claims section of the new version of that patent application (US20120005199 ):
"39. The method of claim 35, further comprising: determining that the anchor text, associated with the link pointing to the document, is related to content of the document; subsequently determining that the anchor text is no longer related to the content of the document; and ignoring the anchor text when generating the score for the document after determining that the anchor text s no longer related to the content of the document. "
There is also discussion in those claims involving "freshness" of anchor text as well.
There are a lot of different factors to take into consideration, some of which we know about, some of which are hinted at by things like patents, and likely somethings that we have no idea about at all.
Many of those are possibly more important than whether a link is the first link or the second link on a page pointing to another page.
Hi Bill,
Rarely is there a case for a link to point to the same page from within the same page in my view, allowing multiple anchor text to pass would only encourage spam and low quality junk (in most cases) which is the one thing Google would not want to do.
I've yet to see any evidence from our own tests (albeit very brief and basic) that confirm multiple links with varying anchor text will pass relevancy for each anchor text and affect ranking. I hope this isn't the case.
I think it is the case. You see it a lot on ecommerce sites where they have an image and then below that, linked text both pointing to the same product page. I recommend the image alt text deviate from the text link anchor to hopefully pass some link authority for both potential queries.
my 2c
Hi Geoff,
There might be a number of reasons why there might be more than one link to the same page on a single page. The search engines are often a secondary consideration when it comes to many of those.
For instance, from a usability standpoint, it's not a bad idea to have a link to a home page via a logo, a link home at the top of a page, and a link to that home page at the bottom of a page, so that visitors don't have to scroll up to the top of the page.
Also, from a usability standpoint, it's not necessarily a bad idea when you provide a link at the top of a longer page to sometimes provide another link to that page (on site or off) at the bottom of the page. For example, if someone comments on a blog asking about the page that a particular link in a post might be, I'd rather include the link in a response rather than telling them to scroll back up to the top of the page and look for certain anchor text.
From a conversion/call to action stance, it's not a bad idea to provide a link and a call to action at the top of a page, and if the content on that page causes someone to scroll, to provide another link to the page, especially if doing so will make it more likely that someone will follow that link.
On an ecommerce site, when people display product images with links and text based links under those images (as spafinder noted), search engines would likely be better served by paying attention to the text based anchor text than whatever alt text the site owner may or may not have included with the image.
It's also a common practice for many site owners to provide a number of links to pages in a main navigation bar at the top of a page in a horizontal role, or or in a top left navigation, and also include footer links at the bottom of a page to a number of the same pages. That's usually not an attempt to spam search engines, but rather to provide a good user experience.
With Google likely using more than 200 different signals to rank pages, and Bing at least at one point likely using over 500 under a ranknet approach, it can be hard to isolate individual signals.
My response above wasn't a suggestion that people attempt to use multiple links on the same page to the same page for purposes of trying to manipulate search results, but rather a suggestion that there are many different factors and considerations that make even trying to perform such a test very difficult.
I loved your examples, specially that of the ecommerce, where search engines will most probably get the signal from the alt text rather then the text around the image.
And I would like to know about this "Ranknet appraoch" of Bing that you mentioned here, can you please elaborate it a bit?
I do think exact match anchor text strategies are becoming outdated with less emphasis on the anchor text, and the "natural" aspect of link building is certainly going to be my focus in 2012.
When getting content published on 3rd party sites I now only go for brand or domain links (or more to the point I don't request exact match) - and have seen that so long as both the linking page and your page are well optimised for the terms you are targetting, then you will see the benefits.
I've even seen evidence (though nothing concrete as yet) of a high authority site (a D.A. of 60) linkiing to one of my sites with a brand link - the linking page had a title tag, url, h1 and content for a term that was targetted on a sub page of my site - and though the link I got was to the home page, I saw the sub page's ranking rise for it's key term (which I am at this stage putting down to internal linking) with little effect on home page rankings.
I think the engines will follow the first two links to a domain... as long as the links point to different pages on the root domain.
Ex. Link 1 goes to the homepage, and link 2 goes to an inner page on the same domain.
Am I wrong?
I thought it was additional links to a *page*, not domain that were ignored.
Surely the engines are using the anchor text to help determine the topic of a page, after all the object here is to return the most relevant page in the SERPS and domains can have lots of different types of content on lots of different topics.
Yes - you've got it Doug. We've never seen evidence that multiple links to different pages on a site are discounted.
Hi Rand,
"HAPPY NEW YEAR" 2012
The topic seems to be a basic one, but you really covered major factors to gain true value(link juice or PR flow) of your anchor text. I found point no.4 really helpful for me because I was confused about first anchor text in HTML of a page and how search engines treat other anchor texts pointed to same page or website.
Thanks for another great WBF!
Nice job Rand, although not very fresh content in it, But anyways here is my question:My site is linking to 3 domains, with two anchor texts each:
Site 1 : Anchor 1 & Anchor 2
Site 2: Anchor 3 & Anchor 4
Site 3: Anchor 5 & Anchor 6
So what you basically means is that Google will only ocnsider the Anchor 1, Anchor 3 & Anchor 5 for each of the sites? and rest of the Anchor 2,4 and 6 will be discarded?
And secondly, negating the value of the internal linking and in fact multiplying it with a zero figues seems a bit of a bold act. Would you like to talk about it in detail? I have doubts about this BOLD statement that you given in point 5.And last point, about surrounding text! Well as far as I ve heard from Amit Singal, he mentioned that this signal is used in the images mainly when there is no alt tag mention! I don't think this will have this big impact as you said. Because if you are right then it means that the BlogRoll link that most of the bloggers have, dont have much value because the text surrounding them might not match the link's anchor.
Hi Asad - let me clarify on those points:
In regards to the "first link counts" rule. That likely applies only to pages, not to sites. It's probable that engines don't count multiple links to the same domain as strongly as if those "votes" came from different domains, but individual pages on the same site can receive different anchor text from a page and have all of it count. It's when an individual URL gets multiple anchors that the ones after the first are discarded.
In terms of internal linking - I'm sorry if I misspoke. Re-watching, it looks like I may have been more strongly negative on this than I could have, though internal anchors are usually, IMO, passing <10% of the value of an external anchor and possibly <1%. I'd also worry that using internal anchors is a sure way to make your site appear "over-optimized" not just to engines, but to users. Internal anchor optimization is one of the times when I see SEOs hurting usability/user experience, and that's probably affecting your SEO more negatively than the anchors are helping.
If you can find good ways to be totally natural and user-friendly with internal anchors, though, go for it!
Hi Rand,
This might be a bit late but i'm a bit confused about the "over optimised" anchor text within pages of the same website and its effect on userbility/user experience. I say this because if, for example, i have a tourism website with well written content about tourist attraction in new york and have a page with detailed content for "statue of liberty" then its probably best to have majority of the links pointing that page from other pages to have "statue of liberty" as the achor test? will that count for over optimised anchor text?
I can't imagine that Google blindly takes the first anchor for granted. Especially if we think in some more "advanced" techniques like boilerplate recognition, template matching or even the "reasonable surfer", it doesn't make a lot of sense to me. I'd rather think that Google is trying to figure out which link is most likely to be clicked (hence probably carrying the most editorial value).Although I didn't find any confirmation from Matt Cutts on this topic. Would be really happy if someone would share :)
That's a very good point. And for those who know about it, it could even affect the quality of the content, as someone might make the effort for their first link in an article to be a keyword link and not care about the look/layout of future links to the same page within the same article.
I guessed it's a case of "damned if you do, damned if you don't" for Google. I'm assuming they made that information public (assuming they did - I think I remember Matt Cutts announcing it originally?) to stop people overdoing it within one piece of content, but the other extreme is they try too hard to make the first link exactly how they want it, potentially sacrificing quality.
The ideal for Google - like you say - is if they'd be able to tell which link is best, based on CTR, surrounding text, etc. I guess that would encourage the most natural linking within content.
One would presume that the link to any given page within the actual body content would be the link that search engines paid attention to - this would naturally be the place to look for where the linking page is saying "Hey, look, I'm the link that should be passing the weight through to that page" hence the previous thought I had about potential devaluation of links in navigational / side menus which I mention in my comment further down the page.
so how many people watched this and then Googled "Portuguese cooks"???
Great WBF to start the year, Rand!
So, does this mean that the websites that go the Zappos route with their footer links for SEO purposes are probably wasting their time?
Makes you wonder doesn't it? I've also noticed a lot of footer links on amazon
Stumbled on this post regarding 2nd link juice passing from an internal page ~ https://www.burndowneasy.com/does-the-second-link-anchor-text-matter/
I published this test back in 2009 so things may have changed since then. Hopefully someone has done something similar recently, and with many more samples to verify the results.
Glad you liked the article anyway!
David
The hardest part of whitehat linkbuilding: "Getting in touch with the webmaster!"
It's unfortuante that the biggest signal has to come from exact anchor text. I can't wait for the day that it's all context supporting your link.
LOL! Indeed.. finding an email address is one thing, but when you finally do and send them a well composed email, you hardly get a return back.. either they never got it or simpy chose to ignore you.
totally disagree with the multiple links statement (p6) as ive done extensive research only the other month that shows they links pass juice however the first link rule applies.
what it mean by? even i thought that multiple links from the same domain which might pointing different pages of the domain would not be worst!! can you plz be spacific on this ? thank you
Just yesterday I read a few articles by various SEO pundits claiming that in the next few months, anchor text will diminish in importance because the search engines are continuing to move away from keywords as a significant indicator and that content will be more important. And, that because of black hat linking problems, even link value will decline, again with the focus on content. (I love to read the various thoughts from people in the industry about what's to come.) Obviously this wouldn't happen overnight and I'm not suggesting jumping on that bandwagon, however, it would negate much of what we currently know, including this white board. Anyone have thoughts they'd like to share on how to walk the balance?
Could you provide us with a link please? thank you!
Hi Slava,
I can't happen to recall all of them -as I read a series of articles when I did a "SEO predictions 2012" search. Here's one that I recall specifically: https://www.cyber-key.com/SocialSEO/seo-trends-2012/ (And by no means am I endorsing any content; I like to theorize and discuss various concepts; there's always plenty of folks who like to pretend to be experts but don't have any brand recognition.)
Hey Andrea, SEO is all about cracking the search engine Algorithms and on basis of tests we come up to some basic rules and regulations or I must call it best SEO practices. People experiment and get different results out of it! and as we know that web is free to share everything!
With referance to your question, this is another school of thought that Google in deducing the value of anchor links and they have their experiment results and assumptions to back that statement. I am one of those people who think that Google is not going to leave this as a signal any time sooner but we have seen Google is now getting smart about guessing links and what it is about. I believe over the period of time Google go smarter about detecting anchor link and to see if they really are useful enough! (A little Hint here: Google got much smarter with Semantic Analysis that it was few years ago)
IMHO… Anchor Text is there to stay and play!
SEO is not "all about cracking the search engine", it's about adhering to search engine guidelines whilst providing excelled customer experience and value. Anyone trying to 'game' the system will eventually lose out in the long-run.
With any business/website, the paramount objective should be to provide value and drive ROI, when done correctly (as any business/website owner should whether a search engine existed or not), your brand/business/website (etc) will naturally perform well in the search engines.
That's all Google are about - relevancy, value, customer satisfaction, quality... Do things right, and you'll naturally reap the rewards. Try and "crack" the search engine, and be prepared for the long haul!
100 % agreed “Google are about - relevancy, value, customer satisfaction, quality... Do things right” but natural ranking need time, what if we need quick ranking may be within a month! I think it’s only possible to play little bit (tiny) as rand says “Tiny is very little amount”. This will give some temporary boost to show some thing to non SEO clients within main time we drive long term strategy and plan (ethical SEO Plans).
I agreed if we want to survive in long run we must be ethical.
+1 "Google is about - relevancy, value, customer satisfaction and quality"
I would agree that Google doesn't seem likely to abandon anchor text as a signal. However, if you spend as much time as I do looking at peoples' backlinks in OSE :-)...and I'm sure Matt Cutts does (maybe with his own whiz-bang tool)...you'd be pretty convinced that it's heavily used as a spam technique.
So I wouldn't be surprised to see Google (a) lessen the impact of a high number of links with the anchor text, compared to other factors, and (b) look at the % of links that do NOT have the company name or domain URL as the anchor text as one of the signals of spam and possible penalize the site.
I referenced this study in a comment above, but I'll do it again...it's about 18 months old, and not a ton of data, but seems to indicate that over-optimizing your anchor text can hurt your rankings. And no, this 2nd link to that same page won't give SEOWizz anchor text for "this study" :-p
Great start to 2012, Rand. I learn something every Friday - like the definition of tiny. As my career has moved from IT Management to SEO I'm amazed at the amount of knowledge required to be successful in this field.
Good recap of anchor tags. Key takeaway: dont spam keyword heavy anchor tags, diversivy, use the alt-tag,consider the surroundings of an external link.
In our testing, the first link principle does count strongly in most cases insofar as the semantic weight of the anchor text itself being passed and impacting the destination page.In terms of pagerank, all links pass pagerank - even in theory, a link to the same page.
Apologies if this was already addresed, but there have been a lot of comments :-) You mentioned that it's not really worth worrying about whether your anchor text fluctuates, UNLESS you use manipulative link-building tactics. What about simply asking another site operator to include the specific anchor text in the link he gives you? I wouldn't personally consider that manipulative, yet it could certainly end up in a lot of the same anchor text links, unless you vary it in each case...
Hi Randfish
Thanks for providing important information
I would like to conform if my webpage have more than 2 ancher taxt in the same url so is it beneficial for web page are not
You said only the first link is counted by google. What if google is considering the pages to be the same, but technically the could be different. e.g
linking to all of these from one page.
www.example.com
example.com
www.example.com/
www.example.com/index.html
will all the links then count.
This information is fantastic and based on what's being said here, goes to show why a business name along with a domain name containing all targeted keywords is so powerful. Not only do they get the domain name juice, there's also no manipulation going on for perfectly matched anchor text. If my business name is Green Widgets, and my domain and that's also the key term I'm shooting for - I win.
What would we do without SEOmoz?
Hi Randfish!
Thanks for this post. It's nice to know about anchor text important. Yes whenever we left a comment with anchor text, it gives you good link juice. It 's strongly say to search engine to see it. Internal or external , link is link :) yes that tiny one also helpful.. External Link is your gold mine untill it's on good page. Thanks for this whiteboard friday.. have a great day.. :)
Cheers,
NetPrro.com.au
Nice post, very helpful. Will check out for sure.Blog is well describing and clear written easy to understandand informatic..blog is very helpful….
Of course, well-chosen anchor is the primary thing when you insert the link, if we want our link to bring us any effect.
Good WBF Rand, and happy new year!
I've found a good example to explain anchor text is Adobe Reader - when you Google 'click here' they rank number 1 due to the millions of people linking to the page with the words '<a>click here</a> to download Adobe PDF Reader' as the anchor text.
Cheers,
Andrew
Hey Rand, Happy new year!I'm a begginer in SEO and I've a question.
Now for example, my website xyz.com has a page on which I've linked to different websites like "SEOMoz tools are helpful!" in which the bold word is linking to seomoz.org how is it going to help xyz.com
Hi SEOers!
This was a great WBF for myself and I have learned a ton. I am new around here so I am not sure if I can ask this question here or not, but I will, because it is better to ask forgiveness, then not know.
I was doing an analysis of the keywords "fort mill sc real estate", and while my website is not on the first page of google anymore (I think since Panda it has slipped away), there is a competitor that I can not figure out how they rank so high. (I am using Market Samurai, as an analytical tool), they have what I would consider total spam sites linking to them with textual links, but with so many other irrelivant textual links, I can not believe theG ranks them so high and counts this website. For the life of me, I can not figure out why besides these two spam sites back links. The only other thing I see them having is an exact match with URL.
My question is not how to rank better or anything, my question is does someone else see something I do not. The site linking to them is a PR3 named digital4sight .co .za and the anchor text is the same as mentioned above.
Thanks for any insight!
2012 has started very very well with this awesome WBF. I liked number 4: the first anchor text is what google counts. amazing helpful. happy SEO to everyone.
Just a few months ago there was a White board friday where it was state that partial anchor text matching is more powerful than exact matching https://www.seomoz.org/blog/beyond-exact-match-anchor-text-to-next-generation-link-signals-whiteboard-friday
And now on your point three you said exactly the oposite. My question is, if we want only to improve the ranking of one keyword, what's the best way? exact or partial anchor text matching?
Video does not work......like many of the other tools on this site!
It worked OK for me! Been using this site for a month now and have found the tools very useful
Newbee to SEO.....Will anchor text help out your site or only the site your linking to, or both? Also couldnt it be bad to have people leave your site by clicking on the anchor text?
Anchor text should be used to accurately describe the page you're linking to, this should be relevant and appropriate to the end user, the visitor.
There is much less weight placed in anchor text now however and more emphasis placed on the relevance of the page, context used and the content surrounding the links.
I wouldn't worry too much about ensuring keywords you wish the page to rank for are contained within the anchor text, especially with the recent introduction of the over-optimization penalty from Google. Just ensure you are creating quality content that your target audience want.
Hi Thank your for sharing this type of post and also video because perfect knowledge about this seeing video post. once again thank you for share this information.
Late to the party on this one, just signed up and this is the first video I watched as it was relevant to a project I was working on. Good content, and lots of great content. I look forward to seeing other videos on the site! Thanks!
These videos are always great to watch
Hello Rand Fishkin
Hope you have a grate day!
I have seen this video after passing the SEO moz test today.
Is these kind of hyperlinks are still worth for high SERP's ? As today's concept is completely based on content and popularity and penguin hits directs hyperlinked websites.
Please let me know about it.
The content was informative and helpful ..Thank you Randy...
Hi Rand,
Thank you for sharing this vid. I know it's a little old, but I have a question regarding internal anchor text. When linking to an important internal page from, say, a blog post, would the anchor text be discounted since elements in the header and navigation menu already link to internal pages?
Best,
Dan
Whoops...apparently you don't accept private messages so I'll send it to you on Google Plus :p
Thanks so much for this post. I've honestly been wrestling with a competitor who I think uses what you refer to as "grey hat" or "black hat" SEO. They have hundreds of links from forums, blog posts, and profiles pointing to a website with specific anchor text. The annoying thing is it works SO well. They rank for every single keyword they have as anchor text in pretty much one of the top 3 spots on Google. I'm wrestling with what is "wrong" with this practice since in the last couple years I haven't seen any of these rankings drop at all. I'm going to PM you with the website of one of the examples so you can see.
Post is so Good to analyzw Importance of Anchor text and SEO Should consider these in pragmatic manner..
Just a Question ..
In regards to the "first link counts" rule ..
if our website Architecture not allow to take benefit of first anchor text as we generally put in Header part, logo, Top Menu,
So is it good or bad to put "nofollow" and then take the benefit from 2nd anchor which put in surrounded text ..
Please Guide ...
error on this video...! :( i can not watch...
Hey Rand,
Great First Whiteboard Friday of 2012! I'm going to think more carefully how I do my anchor links and text.
This informations is very well usable I am looking forward to it. Thanks
Regards,
Ahmed Adnan
Hi Rand
Great WBF as always. Just one question though (sorry if it has already been asked!) Will Google pass equal value to two links on the same page, each using the same anchor text, but pointing to different sites? Or will it lend greater value to the first link?
E.g.
"anchor" link to SiteA
"anchor" link to SiteB
Thanks
PageRank / link juice will be distributed equally via external links on a page but many other factors will come into consideration which will include authority and relevancy which could essentially mean one external link may pass more weight than another.
Excellent video Rand.. however, multiple instances of the anchor text "portuguese cooks" will not be an issue if that just happens to be their brand right, or company name?
I am all smiles after this post, coz seomoz confirmed what I have been testing & observing from my optimization techniques & search results. especally points 1,2 5 & 6
Hi All,
I have a question rather than a comment...
How you will analyse two keyword that which one is more important, the one which is more generic and the other one is derived from that generic keyword along with the addition of the location?
The other question is How keyword proximity can help me out in getting the better ranking for my clients?
Hoping to have a reply soon....
Hi Rand,
Very slightly off topic, but do you think that as social links gain increased importance, that the text surrounding the links will do too? Obviously not all links from social media have anchor text and so one would think that should search engines wish to recognise the increased significance of these links, then they must increasingly look at this surrounding text as a means of identifying what the linked to content is about. In fact is there any evidence that they already do this?
It's always great to see the basic essentials such as anchor text reviewed in a concise, comprehendible way. From what I gathered, you can have the same exact anchor text attached to your inbound external links as long as those links are from non-spammy sites. Where does Google draw the line between spammy and non-spammy sites in order to disregard a massive amount of identical anchor text links? I would think that the distribution of anchor text duplication lies within a game of trial an error on the SEO's part.
Rand - what are your throughts on randomising infographic embed codes to grab a varied range of exact / phrase match anchor text? I generally create about 10 different versions of the 'Created by' text, each with a slightly different set of text and links...
- Matt
For images, what's the SEO difference between ALT and title attributes, or are they treated the same?
I've seen a lot of conflicting opinions about which is better, and some people say use both, but then that creates another debate about whether to use the same text for both or different.
Typically, the image alt attribute should be used as alternative text in the event the browser does not display images - these should be descriptive of the image and not be used for stuffing keywords.
The image title attribute should be an accurate title for the image and is there tooltip style text pop that appears when you hover over an image - this ideally, should differ from the alt attribute and again, should not be used to stuff keywords.
The title attribute does not bear any direct connection to ranking signals whereas the alt attribute will be considered more by search engine crawlers to analyse image content and rank accordingly.
Link titles (text) carry little/no weight to benefit SEO. My speculation is that so many web users simply duplicated the title link as being the exact same thing as the anchor text that I think search engines just said the heck with it and now pretty much ignore it.
Link title text can be more valuable if anchor text of "click here" or "this link" or something vague is used....however, this anchor text, IMO, should NEVER be used in the first place ;-)
One other note---images may benefit more from link titles than plain text since end users may have no idea what clicking on the image may take them too...it's a wayfinding sign for end users to know what happens when they click.
Here's some valuable reading on the topic as well by Ann Smarty. She states it eloquently, "If the title tag can’t provide more information, then don't use it."https://www.searchenginejournal.com/how-to-use-link-title-attribute-correctly/7687/
Hello! I would like to ask something that i tried and i suspect it's a keyword cannibalization.
CASE
From the same external domain (for example testbookmarksite.com - a bookmark site) to give 2 external links to my page.
First link: <a href="mypage.com/index.php" > Anchor text </a>
Second link: <a href="mypage.com/index.php#test" > Anchor text </a>
Is it a tactic that might has the effect of Keyword cannibalization?
Great post as usual Rand, and I agreed with you that anchor text might look easy to understand but there are certain myths about it that not every SEO understands (even myself I admit)
What's your thoughts about using alt attributes as anchor text? How much weight does it have compared to a normal anchor text and will Google consider this to be a rather white hat link even if it might be paid (e.g. many websites link to their sponspors using image banners)
This is really great blog post for all SEO guys who are focusing on link building. I have experience to create 3 to 4 hyperlink on specific external web page and getting ranking for first link. Google is trying to provide best search result on each keyword. That's why it makes changes about external signal on frequent bases. I have questions about external links.
Brilliant post. In-depth for both beginners and the advanced SEO specialists. Thank you!
Nice WBF Rand,
Is there any information regarding length of anchor text as seen by the search engines?
With alt attribute as anchor text the assumption would be something around 140-150 characters. What about in the absence of alt attribute and looking at surrounding content?
The timing here is interesting for me in that a new client (this week) has 151 linking root domains and they are all associated with previous web developer. When looking at what he did with linking and anchor text it seems clever, but the anchor text is very long - 30 to 45 words, 250 to 400 characters. Then he repeats the anchor text as content as well.
So, I am curious if his overlong anchor text in any way added value?
No, we have no intentions of doing same ;)
Good overview.
For me the most interesting point was that issue of "first anchor text counts" when you have the existing link in your menu bar. Never thought about it like that.
So this begs the question, say I'm using a long tail term to boost up my rankings before hitting a short term high competition term which is being used in my menu bar (take for example using the term - winter holidays - in the menu bar and then winter holidays in january in the content). In order to get the content based anchor text link register by Google, is my best bet then to add a nofollow tag to my menu link?
By adding nofollow tags, would Google then go down to the second link or just ignore that too?
Or am I over thinking this?
Great video Rand but as usual I have some stupid questions
1. Google can find pattern! we all know that (Exactly same anchor text on many websites) but does Google bombing are not applicable now a days?
2. As you say finding the exact keyword anchor text is not possible naturally! The only way to get this kind of link is guest blogging?
3. If a page have 4 outgoing links abc on top paragraph and xyz on last paragraph
Abc.com with anchor text “abc”
Def.com with anchor text “def”
Ghi.com with anchor text “ghi”
Xyz.com with anchor text “xyz”
what’s the ratio of PR juice pointing to “abc” and “xyz”?
1. In my experience, Google Bombing has often been conducted when a large group of people agree to all link to the same page regarding a particular subject. The likelihood that all anchor text is exactly the same, is unlikely and in which case, would still play a role in ranking said page for desired target. Even in the event that the mass building of links to the same page with the same anchor text, I would expect that page to fly up the rankings albeit, these days, more than likely only for a temporary period while the algorithm kicked in and rejigged the rankings accordingly.
2. I wouldn't be surprised to see the devaluation of guest blogging / author byline links this year (if at all possible), it's a process that is (as with anything) becoming more and more abused. It is extremely difficult to obtain natural keyword rich anchor text backlinks these days and guest blogging does remain one of the most effective means of acquiring backlinks of any great value, the key is to build relationships with respected, reputable leaders in your field and guest blog from trusted, authoritative websites. It's only a matter of time before the likes of Google really step up a gear in their combat to spam and end up literally devalueing millions of splogs accepting junk guest content purely to manipulate search engines and drive traffic/numbers.
3. The flow of PageRank is distributed evenly between external links, so 25% of 'juice' to "abc", 25% of juice to "def", 25% of juice to "ghi" and 25% of juice to "xyz" (this does not take into consideration the flow of internal PageRank bear in mind) so that's probably the easiest method of viewing the process. The more external backlinks (without the nofollow link relationship) on a page, the more PageRank that will be released via those links.
Hope that helps.
Yep 2nd and 3rd points are cleared but 1st point still has some confusion. Thanks for your help I will do some research on it and hope find some solution.
Thanks
Hi,MR...，Im a chinese,im not good at english.Did this video comes up weekly? its prety good
Yes it is.
You can also check all old whiteboard Friday videos here..
https://www.seomoz.org/pages/search_results?q=white+board+friday
thanks!
If I am building links for www.abc.com
I recived 3 links with specific anchor text from a page - www.xyz.com/123.html
1st link is ( abc --> www.abc.com )
2nd link is ( 123 --> www.abc.com/123.html )
3rd link is ( 456 --> www.abc.com/456.html )
Should I get benifit for all 3 pages on given text anchors?
Yep! All the anchors are pointing to different pages, so each of those pages should receive some benefit.
Oh yes! cuz Rand's statment will be true only if all the 3 links are coming from xyz.com/123.html and pointing to 1 page only for example if all are pointing to abc.com/321.html
In your case obviously Google will ... (Similar Page and not Domain is the keyword)
Rand,
You mentioned that at some point google may catch up to you if your anchor text appears to be unnatural by consistently using exact match targeted anchor text. You are also not very fond of different variations of anchor text based on your targeted keyword. What alternative do you recommend we use in place of a exact keyword match, or varied anchor text?
FYI SEOWizz did this short study about a year and a half ago which seems to indicate that having an unnatural percentage of your links have keyword-focused anchor text can hurt your rankings.
true. a friend got some sites slapped by panda, other ones not. upon inspection, the ones slapped had over 60 percent targeted anchor text, the others didn't (sorry, for some reason my mobile isn't letting me capitalize.)
Nice video Rand, I think people make the mistake when they utalize the same anchor text 1000s of times over and do not diversify the link profile when building natural links, then aggain you need a good percentage of the target term.
I also agree with the comments about having link placement higher on the page with the first crawling of the link.
"Tiny is a small amount" I'm going to use that one! Great back to basics refresher there - nice to have a bit of focus on links and anchor text when the community seems to be spending a lot of time worrying about the pros and cons of social media and its impact on search.
Further to the above would have been good to include a brief bit about partial match anchor text and how that's working at the moment.
Thanks for the video Rand.
Is there any quantitative studies analysing the anchor text on a large scale of 1000+ domains' to see the correlation between anchor text and rankings?
We've noticied less weight being given to exact match anchor text on newer campaigns but still see exact match anchor text maintaining strong rankings on older campaigns.
We used a much larger dataset for our analysis in 2011 - https://www.seomoz.org/article/search-ranking-factors - and found, similarly, that exact match anchors aren't as powerful as they were historically (correlations were lower).
In regards to point 3, are you trying to say that you shouldn't always go for anchor text links and try and be even more natural by doing banner links (for the alt tags) and nofollow linking (forum and blog commenting).
Sorry if I've missed the point here, but as you said there are a lot talking about anchor link variation (very much so in the white hat arena) and I'm just trying to further understand your skeptisism.
If you're doing white hat, I generally wouldn't worry - natural anchor text will happen on its own, and if you do manage to get lots of great exact matches, don't sweat it, as the engines will likely count them.
Great WBF
Question - If an image with a link and completed ALT description comes before a text based link, does Google count this first? I sometimes have a banner with a link and description, followed by a text based external hyperlink in H1 tags, so I'm wondering if I should be looking to drop the banner link if I feel the text link is of higher importance from an SEO perspective.
Best
David
I'd guess it probably does, but this would be worth a test, IMO.
Thanks will do
David
As soon as the both link and anchor text are natural SEs don't care much about it, IMO
Excellent video Rand, exactly what I've been preaching for some time and glad to hear it clarified and confirmed from SEOmoz. So many will insist on linking multiple times within the same page to the same page with varied anchor text adamant that anchor text from all links will be picked up and used as a signal.
My only thought on this previously was wondering whether search engines may devalue navigation menu linking or allow inbody anchor text to over-ride navigation menu anchor text/image alt text possibly - it sounds like that isn't the case from your video - which too, is also good to know. A good 95% of sites online (probably) use keyword rich anchor text for internal linkage which is almost humerous that almost all of it is discounted(?) for pages that are linked to with navigation/side menus.
With this thought in mind, when we re-developed our agency website and subsequently developed our own CMS, we built the page hierachy andset up of the page internal linkage in such a way so that not everything was all linked to from the same page which allows far more scope for passing anchor weight internally between pages.
For example, our services page hierarchy makes no use of drop downs and will only link to top level services from its page, from each top level service page, it will link to sub sections of that service, with always a link back to it's parent item creating a highly 'relevant' internal linkage structure.
I'm glad we built our system like this now and is good to hear that my decision to do so may well result positively based on what you confirm in the video. I only hope the "tiny" amount of internal anchor text passed warrants the way we have built our platform - heh.
I too have heard rumblings/opinions regarding the search engines devaluing the anchor text and/or links from main nav. And, while I do believe that the search engines can "chunk" the page to figure out what's "template" and what's "content", I just cannot believe that the main nav links aren't seen as important, and here's why.
A well-constructed, white-hat site would have its most important pages linked to from its main nav. Its LEAST important pages wouldn't be in the main nav. If I were Google, and wanted to have a measure of what the most important pages on such a site were, I'd definitely consider the main nav links as an important indicator.
Having said that, I wouldn't necessarily believe that main nav links get an extra BOOST, but of course having them on every page means you accumulate a lot of internal link juice from them, which would give them a natural boost anyway.
It depends on the size of the site... (size matters!)
If you have a 5 page website, then Ok - link all of your pages from the main nav.
But if you have a 50 page site (or 50,000 page site), there's no way you can link all your "money" pages in the main nav. You'd have to link to category pages in the main nav, then and add sidebar or body text links to drill down to deeper pages - that's basic website architecture.
Also, navigation on large sites requires lots of buttons in the main nav. Which means they'll be short and possibly one word - like "services"... when the optimized version might be "Graphic Design Services". To use the longer, optimized version of your keywords, the body text is the perfect location.
Certainly I didn't mean to imply that all important pages should be linked to by the main nav--just that the main nav links tend to indicate the major category pages for a site, and those are certainly pages that Google should consider as "strong" pages from the site.
Re buttons in the main nav--if they're image buttons, you can add ALT text and make the anchor text (and title text, for essentially mouse-over longer explanations) the long keyword-rich descriptions you need :-).
If you've got many tens or hundreds of thousands of product pages, you're going to end up with body text links to those "leaf" pages, simply because your major category pages will have lists of the products in them...in the main body. And there, as long as you're careful to link to the detail page first with an image + tasty ALT text, THEN product name or whatever, you'll carry your best anchor text to the detail pages.
Hi Rand! I just love these videos!
What you forget to mention, is about Social Media link, that they are very important. Mainly for keyword that appears in the Freshness SERPs.
I want to learn more about this, because there is no so much information about it. I just KNOW that is important, and the keywords that are surrounding the link, are like the anchor text of a regular link. And I think, that Google is counting the comments when people share on Facebook or Twitter as the anchor text.
Bye to everyone, from Patagonia!
Excellent video. You covvered some great concepts about anchor text links. I think a lot people still don't understand the idea of only linking once to any given page and when there are two links, it is the top link that counts. Understanding the ideas you presented in the video can really help someone change a good link building campaign into a great one.
Great video, thanks. Must admit (and now a little annoyed that I didn't) know google only counts the first link to page. Had been including a couple of backs in some article marketing I have been doing, both pointing to the same URL (different anchor texts). While it's depressing to know I have wasted 50% of my links, it is also good that I can easily increase the backlinks to other domains i own by doing fewer submissions.
Carl
Good point @Rand brought up, but he did not say if the other backlinks hurt you, and since the ultimate goal I would hope to be is to get a user to click a link, it would not hurt to have multiple links. Now if you are only interested in the SEO effect, then that just seems gray hat no?
Hey Rand,
Great Wbf as usual. Couple of questions:
Is there any additional benefit in externally linking to a high authority relevant page or root domain using the first exact anchor text from within your content. I mean would search engines reward you in any way for linking to another trusted website using your keyword as an anchor text?
Second, in an earlier WBC you talked about using partial match anchor text. Would you still apply the same principle here? For eg instead of "Portuguese cooks" as your first anchor text in 100% of instances, you could also use "best Portuguese chefs" and similar variations?
Regarding outbound links to high-authority contextually-related sites: I don't have recent data, but about 5 years ago I did an experiment with two nearly identical websites: one was for Fiji honeymoons, the other was for Tahiti honeymoons. Both with pretty strong rankings to start with, both almost exactly the same amount of content, and both built from the same exact templates (generated from a database of hotels, built dynamically from the database and then self-scraped and remote published via FTP automatically to distant servers on completely different IP C blocks to ensure they were treated as unrelated sites). Then, for the Fiji site, I added a sidebar on the right with about 20 links to super-strong Fiji pages: visitors bureau, the CIA page for Fiji, a Wikipedia page, etc. Bang....big boost in rankings & traffic almost immediately.
But that was 5 years ago...so that's pretty stale data. However, I haven't seen any signs that this has changed. And if you think about it, a site that also offers the user links to related sites that Google VERY MUCH trusts is likely to provide a good experience for the user. There's a Matt Cutts Q&A quoted here that supports this. Now, is this likely to be a massive factor one can use to take a dinky little site with near-zero DA to outrank big competitors? Probably not! But it's probably a mildly positive factor, just like having a lot of rich content seems gives you a boost.
Michael - great advice there... I'm in the middle of putting together a new content strategy for a couple of sites and one of the mid-level things I'm looking at is linking out to high authority, relevant sources.
Hi Rand! Happy New Year! Well, as always, I learned something very valuable to me--I often link to more than one page on my own site so, I've got some tweaking to do. "Multiple web pages will help you, but if they're from the same domain, that's not nearly as valuable as if they're from different domains."
For instance, I will write a post on golf course communities around my area and link to the page for each of them that I've written on my site. I always thought those posts were helpful to pull EACH of the individual pages up in the rankings but, it's good to know that I don't need to be writing these collective posts with tons of links.
Great info and thanks for the tips!
Hi Rand, what are your opinions with W3C's recommendation for linking?
https://www.w3.org/QA/Tips/noClickHere
This is great for the beginner SEO. Rand sums everything up within 10 minutes regarding anchor text. I have been using this exat link building methods for two years. One thing though, a link from a PR 1 home page is not worth as much as a CNN PR1. At least not to me.
First I would like to say you offer very helpfull stuff. Love all of it. My question is that if you are sending the lonk to your hompage does it matter how you type out the url? For example which one of the below is right or wrong or it doesnt matter.
https://www.sample.com
https://www.sample.com/
www.sample.com
www.sample.com/
Thanks
I always use: https://www.sample.com
Excellent post which shows how important are ‘Anchor text’ ..!
I believe keywords concentrated through Anchor text,while doing off page for a site will tremendously boost the value of the website which has been interlinked also I believe this is one of the main factor to get website rank higher in SERP’s.
Thanks for the wbf. I hadn't thought about the anchor text at the top of the page being what counts. Good to know.
What if you link from a website that another website that you own with the anchortext you want to rank on while both websites are hosted on the same ip-address?
Does that work? Or does Google ignore this because you are essentially linking to your own sites?
This would essentially be where a number of other factors come in to play. They certainly won't be ignored (unless the sites are flagged as spam, in a bad neighbourhood or have conflicting/duplicate content), the value that they pass will depend on a number of other signals but will ultimately come down to the authority / relevancy of the linking page - if the link serves little purpose, then don't expect much out of it. If there is a semantic connection between the two pages, and the method in which the linkage has been conducted ticks all the boxes, then expect weight to be passed.
Good information as always. I picked up on a few new things.
#5 and #6 were new to me. Thanks for the video Rand. Will you be at SMX West in San Jose?
Happy 2012 SEOmoz!!! Great write up Rand! An excellent overview of anchor text and some a basic link building startegy.
Hi,
A few months ago I started a paid image/banner advert on a niche related website. For about 5 years I have been ranked 1-2 for a key phrase and mid December my site dropped off of page 1 and now crept back to 10.
I have been searching for causes for this. Unlikely that 8-9 other websites have suddenly got better seo so I think a problem with my site or how Google sees it...
I have noticed in Google Analytics that the linking text for mysite now shows the alt text from the advert which is "paid advertising" and this is my second highest linked text and second only to my url.
Could this be the reason why I have had a 'penalty'?
My website has NOTHING to do with advertising....
Any thoughts on what to do? Can I block that anchor text from my GA?
I can try to get the alt text changed... or ask for a no-follow?
Or do you think this anchor text is unlikely to be the cause to my drop in rankings - not just for one key phrase but for many.
Hi there,
Paid advertisements and links should always have the nofollow attribute attached to them. I would start there. :) If the link is nofollow, I wouldn't worry much about the alt.
Great WBF, Rand. You neglected to include anchor text in external links to related content in your list. Do you think this is used as an anchor text signal, or just considered part of the page content?
Good point! I'd say it's probably both. Anchor text on links IS a part of the page's content, and the sites/pages pointed to also likely figure into the equation.
Nice work, Rand and crew.
What are your thoughts on how title attributes fit into the mix? Specifically, I assume search engines associate title attributes with the page being linked to, but is there any evidence that supports the page doing the linking also benefits from target terms woven into the attribute (within the natural flow of the sentence)? So if you can pull it off smoothly, would it be smart to include a target term about the page you're on (let's say, 'international recipes') as well as a term for the linked page ('Portuguese cooks'): "For great international recipes, trust these staples from our favorite Portuguese Cooks." ... If so, that could help entice links from sites, because you'd be directly benefitting them too.
Finally, do you have a recommended character count limit for both title and alt attributes? This may very well be covered in the forum, but if so I've not been able to find the right discussion thread.
Cheers to you and the SEOmoz team for providing an endless flow of insight.
Joe
I seriously want to get more data on the "sourranding text" Rand. I have done a bounch of tests and that was never true!
Also I wish you indicated the anchor text trust. If you get exact match anchor text from trusted sources then you are good. Matt Cutts himself has said that a few times "if we trust the anchor text..."
Thanks
Insight from this post, provide a vital information about anchor text. I really enjoyed read this. External link is very helpful for those website that are not your own linking to you. Mostly the engines are using the anchor text to help determine the topic of a page.This link are the mainly important source of ranking power. Thanks for informative post.
adsfasdfadsa