I'm not sure that each of these deserve their own separate blog post, but they've been rolling around in my head long enough that at least some attention is probably warranted:
- Is NoFollow Really NoFollow?
Danny Sullivan brought up at his SMX Madrid keynote that he felt the engines were probably already "special casing" certain websites and pages that use nofollow on outbound links and "following" them for both discovery and potentially to provide ranking weight. Certainly, they'd have to lie about this in public or risk massive manipulation, but I have to admit, the quality of links on Wikipedia has dramatically improved since they instituted nofollow, and if I were running an engine, those would be links I'd probably want to at least test and probably count.
_ - Domain-Level PageRank
There's been some speculation that PageRank isn't just used on the page level (#2 on Matt's list). Internally at SEOmoz, we've often discussed that it's very likely the engines run a domain level link metric on the domain graph itself - a pretty easy process when you think about it. Just take all the links on all the pages on a domain and imagine condensing them down to one page, then run a "PageRank"-like algorithm on domains, rather than individual pages. I'm guessing this produces some very useful and usable data for search rankings.
_ - Are SEO Flags a Bad Thing?
A question that keeps popping up stems from the feeling that Google and the other engines want to find websites where "SEO" is being performed and penalize them or perhaps run them through stronger spam filters simply because they use things like nofollow on internal links, noarchive meta tags, very targeted internal anchor text, etc. I'm not sure that I've come to a full conclusion on this subject, but I tend to lean to the side that thinks it's BS. I'd estimate from experience that 40%+ of Fortune 1000 websites and 60%+ of the top 10,000 most important domains on the web do SEO of some kind. There might be flagging systems for aggressive over-optimization, just as there have always been, but penalties just for doing SEO seem fairly outlandish to me.
_ - When Will Knols Rule the Index?
I can't imagine Google not wanting some of that amazing, high quality search traffic to stay on Google properties (other than YouTube, of course), so when are Knols going to become the new Wikipedia? I can't imagine it's more than a few years before we see every broad query under the sun featuring highly ranked Knol pages - it's just too irresistible as a Google shareholder to explore that potential revenue channel.
_ - Internationalizing on Domains Back in Vogue
With the launch of Webmaster Tools' geo-targeting option, I thought we might well see a move to keeping multi-lingual and international-targeted content on a single domain (seomoz.org/de/ and seomoz.org/fr/ instead of seomoz.de and seomoz.fr), but hearing the chatter from international SEOs, it doesn't work nearly as well as you'd think to geo-target subfolders (maybe that's an error Google's working on) and people in other countries (particularly France, apparently) don't like to click on ".com" or ".org" TLDs - they prefer ".fr." This extends to branding and conversion rates, too - web users in France like to buy from websites in France that end in ".fr" and the test data, according to the venerable Andy Atkins-Kruger, is there (though not available in a publicly accessible document, sadly).
If CTR and branding are concerns (and they generally are), maybe we need a new way to tell the engines "Hey, all these domains are run by this singular entity and if you trust one of these sites and think it's important, you should pass along those metrics to these others domains." Personally, I think they already do this to some degree, which is why you don't see usually new domains registered and launched by large, important, trustworthy organizations suffer from the same effects as new sites launched by small-time webmasters.
_ - Yes, Matt Cutts Said All Links to a Domain Help Every Page on That Domain Rank Better
Danny was complaining that Matt dodged the direct question three times in rapid succession at SMX Advanced, but that's an answer in and of itself. If the answer was "not always" or "no, not really," he'd be pretty happy to share that. Reading between the lines with what Matt (or other search engineers) say isn't a fine art in this case - it's stunningly obvious. The answer is "yes" - every (good) link to a domain helps all the content on that domain perform a little better.
_ - Is There Still a Debate on Nofollow?
If there is, I can't see why. Go try setting up a nonsense domain, pointing some links no one will find at it, and test for yourself. If you have three links on a page and nofollow one of them, more ranking power and link juice will flow through the remaining two. Now, maybe you think that's only having a small impact, but if you're into "optimization," then you really shouldn't be ignoring any method that's going to help you rank better. Smart black hats did this years and years ago by pointing to the pages they didn't want to pass link juice to with external javascript re-directs or links embedded in Flash. Why would the tactic suddenly stop working now that we can use the more convenient "nofollow" tag?
If you think it's low on the totem pole of "to-do" items, that's one thing, and in many cases I'd probably agree. But to say that no one should ever do it because it doesn't work, or because they're too dumb to figure out how to do it properly, sounds like you're either in denial or don't want competition. I'm also at odds with the argument that we shouldn't be doing things just for the engines - we do robots.txt and meta tags and sitemaps and verification with webmaster tools and yes, even title tags (seriously, no one sees those things up in the corner of their browsers) JUST FOR THE ENGINES (and maybe a teensy, tiny bit for accessibility reasons, but take away engines and I think 90% of that stuff is gone). We have to do lots of stuff to make sites and pages accessible to search engines - that's why our jobs are important.
Obviously, as always, I'm sharing my personal opinions, colored by my experiences. I'd certainly love to hear community feedback.
p.s. One more thing - my amazing fiancèe, Mystery Guest, wrote a blog post for SEOmoz, her first one in many months, on the subject of sexism in social networks & on the web as a whole. We were a bit concerned that it might have a negative backlash (as it's not light material), but figured we'd ask for the community's opinion first.
p.p.s. I cleaned up almost a half dozen inaccuracies or unclear language in this post. Thanks to everyone who emailed me about these!
Shouldn't Mystery Guest just submit her post to YOUmoz? She's not an employee so I don't see why her opinion should have any bearing on the reputation of SEOmoz.
Do we need a DVD disclaimer on every post by a non-employee?
"The views expressed in this blog post do not represent the views of SEOmoz, its employees, or affiliates...even if the person writing it is Rand's fiance...she's still not an official part of SEOmoz...she's not even involved in SEO...except socially at some conferences."
@great scott!... Isn't that already implied by whether or not the "SEOMOZ STAFF" logo is appended to a person's avatar? Besides... if you have to include a disclaimer, I think it should be something like this: "The views expressed in this blog post do not represent the views of SEOmoz, its employees, or affiliates... but even if it DID... so what? If you don't like it... take your bitchass somewhere else, and stop using our bandwidth."
Which is exactly why I don't understand why we're vetting public opinion before she posts (if she's going to).
She's a big girl. If she catches heat, she can deal with it. If she doesn't want to deal with it, then she knows better than to post something.
2. Domain-Level PageRank: There's been some speculation that PageRank isn't just used on the page level (#4 on Matt's list). Internally at SEOmoz, we've often discussed that it's very likely the engines run a domain level link metric on the domain graph itself - a pretty easy process when you think about it. Just take all the links on all the pages on a domain and imagine condensing them down to one page, then run a "PageRank"-like algorithm on domains, rather than individual pages. I'm guessing this produces some very useful and usable data for search rankings. I disagree that this type of metric would be useful. First of all, Google ranks pages, not domains. The entire purpose of PageRank is to determine the relative importance of every page on the Web. Even if Google did calculate the total PageRank for a given domain, how would they incorporate that metric into the ranking of pages? Is it accurate to say that a domain with a high total PageRank is more-likely to host high-quality pages? I don't think so. There would be a trend there, but nothing strong enough to bias search results on. Take SEOmoz as an example. It has a lot of awesome SEO content, but it also has a bunch of useless crap that has nothing to do with SEO. So why would Google start making the assumption that important domains host important pages, especially when they already have importance mapped out at the page level? Personally, I have always believed that every page in Google's index is ranked individually, based on its own merits. I don't believe in domain-level metrics having an effect on page-level rankings. Every page for itself. That's the only way that makes sense. 6. Yes, Matt Cutts said All Links to a Domain Help Every Page on that Domain Rank Better: Danny was complaining that Matt dodged the direct question three times in rapid succession at SMX Advanced, but that's an answer in and of itself. If the answer was "not always" or "no, not really," he'd be pretty happy to share that. Reading between the lines with what Matt (or other search engineers) say isn't a fine art in this case - it's stunningly obvious. The answer is "yes" - every (good) link to a domain helps all the content on that domain perform a little better. Isn't this obvious? If your site is appropriately interlinked, then any page that acquires additional PageRank (through a new inbound link) will distribute that increase to the pages it links to, who will distribute it to the pages they link to, etc. etc. etc. Any increase in PageRank to a single page within a well-linked site, will also increase the PageRank of that page's outbound link partners. It sets off a chain reaction, which ultimately increases the PageRank of every page in that well-linked site. The question is... how significant is this increase? I like to think of it as a hand grenade. When someone links to one of my pages, it's like they have thrown a PageRank hand grenade at my site. The page the grenade lands on takes the most shrapnel. The pages in the immediate vicinity (i.e. the pages that the first page links to) take some, but not as much as the first page. The pages at the third tier get a tiny bit, etc. So in theory... the explosion of this grenade would produce a violent burst of energy that would travel outward indefinitely. But if you're half a mile from the explosion... are you going to feel that energy? The moral of this story is... inbound links are going to have the most significant effect on the page they land on. If you want to turn this information into an actionable item, then try this: 1.) Get a list of the pages on your site that have the highest PageRank. (Doing a site: command will give you an approximate list.) 2.) For each page, review the crawlable links (i.e. what pages does this page link to?). 3.) Remove links that don't go to important pages. In other words... only the highest-quality content gets to feel the shrapnel. Some of you may be thinking, "Pffffft! What does THIS guy know about grenades?" Well let me tell you something, buddy... I passed the U.S. Army's grenade course with a perfect score, when I went through basic training. I'm a certified EXPERT at throwing hand grenades. So THERE!
Love this:
SEO 101
Note to self:
When dropping a link (i.e. grenade), vicinity to the target is critical to achieving desired effect. Okay, I think I got that concept down. Can we move onto some more advanced SEO concepts now?
Photo edited to incorporate small amount of brain matter. I new I forgot something. Sorry folks.
Consider me permanently scarred from that photo. Thanks a lot, Sean.
So Lisa, are you saying you don't like my Darren Splatten?
Edit:
Lisa, I'm glad I had the opportunity to come back and edit this comment before it was locked in. I had forgotten you were dating Darren and I apologize for coming across so callously.
I realize how difficult it must have been for you to look at that photo and I certainly don't want to be remembered as the person who seared that image into your mind.
I went ahead and covered up the offensive part. I hope this helps you remember Darren in a more dignified manner.
It was a typo, Sean. By "scarred" she meant "aroused."
I new I forgot something. Was it the letter k?
knice.
wow interesting picture to see at midnight
Hmm.. Grenades & SEO - i like it :)
As to domains - makes sense - you get the kill zone of 5 metres with the Page - a casualty zone of 15 metres with the category that page is in - and then the fact that frags can fly as far as 200metres which'd be the domain!
Personally i didnt rate throwing grenades - never that fun in a trench with ear protectors etc. - shooting was more fun :)
And before it gets to macho I can tell you for a fact that the British army officers are the biggest bunch of nudist crossdressers on the planet! Thats how they balance the whole macho thing out..
Maybe an SEO fancy dress party?!
I never knew anyone in basic training who didn't get "expert" in the "grenade course". If you can throw a grenade in the back of a large pickup truck you too can get "expert" everyone - that's about the skill level. I think it must be an artificial confidence booster - I admit it made me feel good to get "expert" at grenades when I couldn't manage that on the firing range.
What nostalgia...I'm glad to be out of that corporation.
Nice bunch of thoughts! I agree with you on everything but 6. I thought Matt ducked the question - and I think that could be either because, as you say he wanted us to work it out for ourselves (or couldn't duck it better) or because the answer was complicated / obscure.
I believe that domain weight passes to other pages on that domain as long as they are linked to from existing pages on that domain (i.e. as long as the domain-level link graph is connected) but that is a natural consequence of pagerank and doesn't need a separate domain level pagerank... I don't think I've thought about it as much as you though ;)
Regarding the PS: You and Geraldine both know my thoughts on her posts - big thumbs up. I've been waiting for the next Mystery Guest post for ages. All the more so now we have actually met!
Re: Mystery Guest post
It's about damn time. I didn't dress up that way for three days for nothing, you know!
I am with everyone here... we've been waiting too long for Mystery Guest's posts :)
Good post Rand - I agree with almost all of that and the info on the .fr CTR/conversion rate is fascinating.
As for mystery guest's post - my vote is YES. Though it does sound like the kind of thing that might get negative feedback. Personally I don't think that should stop you posting it. Just my $0.02.
Are we allowed to add our own scribbles? One thing that's been bouncing around in my head recently is this - are we going to see more and more SEO conferences/events/blogging done either offline or behind closed doors to stop the search engines getting in? Historically it's been all about algorithm this and algorithm that but recently they're saying things like "normally we wouldn't have done anything about it, but since it fell into our lap we might as well do something about it" (ref: money.co.uk and the lyndon thing by Matt Cutts at SMX). I can see this kind of action driving more SEOs underground.
Just a scribble :-)
That is a good point there from Tom regarding the negative feedback. I want to see the post (see below) but there might be negative feedback from some quarters - do factor that in to the decision...
Is NoFollow Really NoFollow?
Heh... I've always thought that this has been part of the "Reverse Psychology Algo". Anybody who will pay big bucks for a nofollowed link has a site that is really worth something - and Google gives them 3x the linkjuice... hmmm... maybe it's 5x?
EGOL,
it's been awhile since I've seen a comment from you, but that might also be that my time here is much more limited and sporadic.
But as always, you come out swinging and hit a homer!
Hello, my name is David Martín and i'm SEO.
First i would like to express my congratulations because i saw you at SMX in Madrid.
In reference to point five: Internationalizing on Domains Back in Vogue. I would like to comment something:
I think that Google Webmaster Central should permit to connect a domain with a language [like English, Spanish or Chinese] and not only to a country.
Are you agree?. Im waiting for your answer.
Thank you very much for all, and please excuse my english.
Att. David Martin.
Spacebom.com
Velneo.com
Some interesting points here Rand (as always); I do wonder about Knol, especially as many of Google's actions recently seem concerned with keeping people with them for a bit longer.
I actually wrote a post about it a while back, on how I thought they could do this and it's pretty simple; set-up some filter to start blocking out the Wikipedia results.
Whether they would do this is one thing, the other thing I wonder is whether Knol will simply wither slightly, as so many Google projects/purchases have (Jaiku anyone?)
And yes, MG should definitely post. If anyone gave her any trouble she's kick their arse anyway...
Yes, Google is holding on to the visitor and using things like site search for authority sites as another way to plug a few more PPC ads before sending the visitor on their way.
What, you mean you let her out of the kitchen to write blog posts?! :)
Post sounds very interesting; look forward to it!
I think SamIAm's point is - isn't it kind of sexist to ask if you should post MG's blog post?
I gotta tell you - I'm a bit taken back by this question.
I don't know Geraldine personally, but from everything I do know of her she seems like a big girl capable of taking care of herself and doesn't seem like the type that needs protection from her big strapping SEO fiancèe.
And she's one of the most well-received contributors on the whole site.
If she's worried about another Factpile incident, I can think of only one current contributor who would be so harsh - and he seems to have mellowed over the last week or so.
And as others have pointed out it is topical to Ciaran's question.
Just post the damn thing already.
I'll post some proper thoughts below, but have to add that when I referred to machismo, I wasn't talking about misogyny (although it is very common as well) so much as just a general level of aggressive behaviour (from people of both sexes).
[waves to Darren] ;)
[waves to Darren] ;) Put your hand down, baby. The men are trying to talk.
If I would cyber-slap you for that comment I could. :)
I'd thumb you down but I know that's what you want. So instead I'll thumb you up. Take that, jerk.
I think SamIAm's point is - isn't it kind of sexist to ask if you should post MG's blog post? I gotta tell you - I'm a bit taken back by this question. I don't know Geraldine personally, but from everything I do know of her she seems like a big girl capable of taking care of herself and doesn't seem like the type that needs protection from her big strapping SEO fiancèe. @Vingold... I don't think your response is fair to Rand OR Mystery Guest. Perhaps you should reread what Rand's exact words were: We were a bit concerned that it might have a negative backlash (as it's not light material), but figured we'd ask for the community's opinion first. Sounds to me like they discussed this with each other and are making decisions together. I don't know how you managed to twist that into Rand = sexist. Have you met Rand? I can't imagine him ever being sexist... he virtually IS a woman! [Just kidding, Rand. Please don't exile me from the SEO industry. =( ] If she's worried about another Factpile incident, I can think of only one current contributor who would be so harsh - and he seems to have mellowed over the last week or so. Mystery Guest has total immunity. Anyone who is physically attractive is automatically exempt from my harshness. That's the only reason why I've tolerated your gibberish, Vinny... cause you're just so damn pretty. In contrast, there's this guy...
I happen to know for a fact that that is inaccurate information.
I happen to know for a fact that that is inaccurate information. Well... to be more specific... my harshness variable is dampened by a person's hotness variable. It's all part of a complex algorithm I use for ranking humans. Let me put it this way, babe... if you weren't so hot, I'd have killed you by now. *kiss*
Get a room, you two!
I can't. I'm still trying to figure out if I was just insulted by Darren. Again.
"Anyone who is physically attractive is automatically exempt from my harshness."
So that's why you never make fun of me :)
So that's why you never make fun of me :) Let me rephrase that: Anyone who is physically attractive--or who looks like they were raised by grizzly bears--is automatically exempt from my harshness.
cute, very cute.
I happen to be both.
I happen to be both. Yeah, I know. When I first met you... I didn't know whether to ask for your number or play dead!
I tie my food up in a tree whenever Pat's around.
I haven't figured out how to get to it either. But dammit I will...
Darren - we actually had a conversation just yesterday about how high my estrogen levels probably are. I think Rebecca responded that she's balancing the equation with an abudance of testoterone - no wonder we work so well as a team :)
Darren - we actually had a conversation just yesterday about how high my estrogen levels probably are. I think Rebecca responded that she's balancing the equation with an abudance of testoterone - no wonder we work so well as a team :) Um... there are way too many ambiguous pronouns in that paragraph. This is how I interpret it: Darren - [you and I] actually had a conversation just yesterday about how high my estrogen levels probably are. I think Rebecca responded that [Rebecca's] balancing the equation with an abundance of testosterone - no wonder [you and I] work so well as a team :) I'm confused, Rand. First of all... I don't recall this alleged discussion you and I had yesterday, regarding your estrogen levels. Second of all... I don't understand why Rebecca's genetic predisposition for a beard and chest hair would have any correlation to the success of you and I working well together. I'm sorry, but this is your worst theory yet, in my opinion.
Between my lack of estrogen and his abundance of it, we balance each other out...when we're not yelling at each other, that is. ;P
Between my lack of estrogen and his abundance of it, we balance each other out...when we're not yelling at each other, that is. ;P I think you might be confusing "estrogen" with "human compassion." ZING!
Nofollow is not, in any way "nofollow" I have ample examples of this.
I think that as far as relevency goes, nofollows have weight, but for what we think of as "linkjuice" it has noweight.
Linkjuice: No weight
Relevancyjuice: Weight.
My vote on Mystery Guest is let it rip, I would love to read her opinion on that, the heavier the better. I love heavy.
@feedthebot... Are you saying a nofollowed link from page A to page B will help page B's rankings for the term in the anchor text?
It is to long to go into for the moment but what I am saying, and have proved to myself through some experimentation is this... (oversimplified)
If 10 (or however many) quality, relevant pages nofollow link to page B, then page B will show up for relevant searches more than it would if it had no links whatsoever.
As for your question of anchor text, I do not think anchor text is the main factor in these instantces, I believe other relevecy factors are in play in addition (or perhaps instead of) anchor text.
I'd love to see MG's post, but if you're concerned, why not put it out on YOUmoz and see what people think?
I think that's a good idea. Post to YOUmoz and see how well it fares.
I think that's a good idea. Post to YOUmoz and see how well it fares. I don't know... I've seen the people on YOUmoz reject some really good posts. I don't think the opinions of YOUmoz members should be trusted... or even considered. They're animals at best.
Oh, and Rand and Rebecca:
Stop trying to rank for "estrogen"
Hi,
I definately think that your fiancee should publish her post. She is an excellent, witty, insightful writer. However, she is very sensitive to people's comments. I'm sure there will be comments that agree with her and disagree with her, which is fine, but there may also people who question her place on this blog and who may be insulting. They will not be responding to your question here on whether her post is welcomed.
I really hope that this very very small minority does not dissuade her now and in the future.
Geradine writes with heatfelt emotion and I hope to see her post.
Risa
#6 - completely agree - I have competitors that rank very well for cerain terms, but their domain does. Their links for their homepage are for other industries, not mine. And when you check the links for that specific product page, they turn horrible like everyone else in the industry. You can't tell me that links to the domain don't help the site as a whole.
Mystery Guest's Post: Bring it on. I am really interested to see what she has to say and willing to fight anyone that messes with her. She saved my life the other night (had a nail file) so I owe her.
Also Rand ... like my new name?
I would argue that title tags are for people due to their browser tab usage. I don't know many people that don't use more than one tab/browser window at a time and I think they are necessary for organization on a human level.
Thanks for the insights!
I definitely agree with you Tim that without title tags I'd be lost in firefox tabs. Also - aren't they kind of critical to helping your CTR from the search results page?
But of course Rand is right, we'll always be doing stuff just for the engines.
On #3 - that we might be penalised for doing SEO; isn't it less that we're peanalised, and more that Google continues to better learn what really makes a useful webpage?
Eg: Back in the day it was deemed relevant to have the keyword on the page, so we started keyword stuffing, so the keyword density had its influence reduced as a factor in the algorithm. Google may start to find that 'targeted internal anchor text' doesn't actualy make the site more relevant/useful to people, and start to scale back the importance of that.
Definitely +1 for MG's post.
Yes, but I would consider that "for engines" as much as for people . . . point being that if SE's didn't exist we wouldn't need the titles to populate the results.
#4 Domain level PR - it makes perfect sense that a domain level PR could and would be calculated, and I'm not sure that it totally equates to the homepage PR, but as you think, a combined and weighted score.
Just like looking at IBL distribution...do all links go to the homepage, or is there a decent proportion that goes deeper into the site? Point being, if IBL's and PR are too heavily skewed to the homepage, then the site may be providing less content value than one would hope.
Some sites are very authoritative, but shallow...people link to their homepages because they are authoritative. Other sites are super-authoritative, gaining links not only to their homepages, but to deeper content as well..."hey look at this information b/c it is really good." I could be wrong, but I'd think that the engines would like to identify those super-authorities, and looking at IBL and PR distribution across the site and some overall domain score that represents that would be important.
#6 All Links - would be nice to get a clear answer, and I agree with Will, that one could still read this different ways; but ultimately I would agree that there is a cumulative effect here. Even if it isn't so, building good, strong links to a site is a win no matter what...and if it carries exponential value, all the better.
#7 Nofollow debate - you know this won't end any time soon ;)
MG - bring it on. Will it spark debate? I could only hope so. And nothing better to fire up some debate and get neck hairs standing.
With social networking sites becoming more and more relavent to SEO, I don't see why Mystery Guest's post shouldn't be posted. After all, isn't this blog supposed to be about discussions and stimulate thoughts, ideas and interaction?
Also, one of the things I really like about SeoMoz is that people aren't afraid to speak their mind or be honest. Why break the trend?
I also vote yes for mistery guest´s post.
As for the mini-questions you mentioned:
1. I think a lot of SEO had that in their minds, the last couple months, and you just spoke for a lot of them (including me).
5. IMO if you have a defined geotargeted market, you need a specific TLD to perform as best as you can there. But a .com can give you ranking across similar language search engines - like Portuguese and Brasilian Portuguese ones. In my experience, if you have a .pt TLD you won't rank very well in Google.com or Google.com.br, but if you have a .com TLD you have better chances to rank well in both.
The strongest arguments against large implementations of nofollow internally are that it's a) imprecise and b) not standardized. Sure the engines announced they're working towards standardizing REP and link attributes including nofollow, but I'll believe it when I see it. At this point rel=nofollow is a suggestion, not a command, and it's not treated equally across the engines.
Here are 8 Arguments Against Sculpting PageRank with Nofollow:
https://www.audettemedia.com/blog/arguments-against-nofollow
It's an interesting debate. I still use it for overhead pages, but am scaling it back. Would love to hear some other perspectives.
@Adam Audette... I enjoyed your presentation on this in Seattle, BTW. I have always been a strong supporter of PageRank sculpting, and the nofollow attribute is the obvious choice for that. However, I agree with a lot of what you're saying, which is: don't trust completely. With so many unknowns involved, I admit that relying on it completely is something I try to avoid whenever possible. For a client's site... sure, I'll just slap a nofollow on links that have no value, but for my own sites (which don't actually exist), I have been using an external JavaScript (which is Disallowed in robots.txt) to insert iframe tags (thus allowing XHTML strict compliance) that reference a src file that contains a Meta robots tag set to noindex, follow. =P
Wow fancy stuff.... or you could just use good site architecture in the first place :)
thanks Darren, appreciate the response - sorry to have missed you at SMX - let's meet at the next show.
I guess... but a good site architecture tends to bias your PageRank distribution towards landing pages that actually have supporting content pages. This becomes an issue when my clients want to rank well for keywords that they don't provide valuable content for. ;) As for SMX... we didn't officially meet, but I'm sure you saw me there. Maybe you just assumed I was The Lisa's bodyguard or something. =P
Thumbs down for that craptastic picture.
Rand in point 5 you mention,"which is why you don't see new domains registered and launched by large, important, trustworthy organizations suffer from the same effects as new sites launched by small-time webmaster".
I'd have to disagree to a certain degree. I work at a fortune 1000 and we split main site into brand sites and suffered some of the issues that are faced by smaller sites.
It took about six to seven months to fully recover.
So while the issues may not be as severe, when launching a new site or even rebranding large organizations do face some of the same issues as smaller sites.
This is a highly informative post with a lot of new interesting points I never knew about. I liked all of the 7 tips about what SEO company’s jobs and why they can be so effective. You used great informational data to prove your point which gives more creditability to you about what you are talking about. Also I would love to hear a post from your mystery guest because of the seriousness of the topic she wants to talk about. This is a very nicely written blog and I look forward to reading more posts!
Could part of the reason for the popularity of having seomoz.fr or seomoz.it be that surfers in those countries prefer their native extension?
Perhaps its not entirely about SEO, but also about brand trust and giving the impression that the localized site is not just an afterthought.
I'm not a world traveller by any means, but in my time in Italy I noticed that most outdoor display ads featured sites with the .it TLD. A .com would have seemed out of place.
I miss mystery guest's posts since the last time she posted and felt like she got run off the stage. She is an excellent writer and if people don't want to read her post for whatever reason, they don't have to. It's the web. It's not like our fingers are incapable of clicking away if we don't feel like reading. Personally, I'm looking forward to reading some more of her great writing no matter what the subject is.
As ususal great stuff. But I only comment regarding:
p.s. One more thing - my amazing fiancèe, Mystery Guest, wrote a blog post for SEOmoz, her first one in many months, on the subject of sexism in social networks & on the web as a whole. We were a bit concerned that it might have a negative backlash (as it's not light material), but figured we'd ask for the community's opinion first.
Umm - bring it ON! I used to love her posts.
I like what you said about the nofollow hype....it still reduces weight of the links, so it is still useful (sorry if I've mis-summarized).
Rand,
About #2:
I always considered the concept of Trust Rank as a Page Rank for a domain -- not only as black-list thing (ensuring which domains are bad), but also as a white-list thing ("oh, this is such a good domain, so I'll give extra relevance to this page, even if it has low quality")
Honestly, I'm kinda surprised because it seems plain obvious.
Another example which shows that Google sees PR on domain level is on Google Webmaster Tools, where it shows your amount of pages between each grade of Page Rank. To me it's obvious that box reads as an domain quality score.
My experience here in France is that most companies creating an online store want a .com not a .fr
.org is largely misunderstood but is starting to be used
A lot of French sites are .com and Google.fr doesn't seem to show much preference for .fr domains ; at least not as much as Google.co.uk does for .uk domains.
So the theory that French buyers prefer .fr domains suprises me. I can believe that French users will prefer the .fr domain of an international brand but for a French brand where .com is understood for its orignal meaning : commerce (which translates as commerce or shop in French)
Some examples of .com domains for major French sites :
https://www.fnac.com - Online store of France's leading high street book and music store has
https://www.cdiscount.com - France's leading e-commerce pure player
https://www.darty.com/ - Online store for France's leading High Street electronic stores
Is NoFollow Really NoFollow?I think that wikipedia represents a higher trusted authority, it could be an exception, and we should not forget that google is a partner in many open source projects. today i hit the link from wikipedia, tomoro another web site adopts the link because it is in wikipedia, later too many trusty web site do the same... so it starts from wikipedia but this is the root/source not the cause_
Domain-Level PageRank
I think domain level represents a weight just when backlinking others not for on page optimization
Regarding #5, I wonder what the effect (if any) of adding a link tag to the appropriate language / locale would be. e.g. <link rel="alternate" href="https://www.mydomain.fr" lang="fr" />
It would seem to be a good way to associate a group of sites into a common 'cluster'.
Thanks for the insights Rand!
Leadegroot- I think Danny's question was what was Matt's thoughts on Page Sculpting. Does it have any impact?
+1 for Mystery Guests post
Link to FR data?
The jury may be still out on geotargeting, but I've seen some pretty impressive results from it :)
Personally I think that the issues may be more with geotargeting in general (and in particular UK), rather than the tool. I also think that even Google may not fully understand the implications of the tool, and that it could be considered Beta until we see more results.
Following Ciarans observations yes I think its about time someone wrote about it - so let mystery guest post! And negative feedback.. well screw 'em.
As to internationalization and tld's... I'm thinking of a youmoz post on this because it doesnt seem like its as clear cut as people hope.
In point three, Rand mentioned 'over-optimization.' Was that a black hat reference? If not, could you give me an example?
To further solidify some of her assertions or give cause to rethink some insights, it would be a good idea for MG to place her post on other blogs perhaps one with a guy's name and avatar and one with a girl's and feminine picture.
It would be interesting to see the reactions people have according to the supposed 'sex' of the author... just a thought.
Of course, she should post her thoughts. If people do not like reading other's thoughts which inspire their own, then they are lacking awareness.
Or she could adopt the Scarface approach to them...
I've been musing upon the paid links thing again recently - I read a great post by Rae (Sugarrae) Hoffman re paid links and like her it really troubles me that of course Google couldn't possible know that a link was paid for (as they don't have access to that kind of thing), what they do is guess.
I also have to question why an authority site would (or indeed should) link for free, when they could charge and still keep editorial control over who they are linking to. Everyone has bills to pay...
I think it's tricky because having said all of this, I do think that links are a very good way of judging the authority of a site - gah am going round in circles.
Anyway, re Mystery Guest's post - I vote yes too :)
I'd love to hear it!
I would love to buy the Wikipedia nofollow argument as I love Wikipedia but I will have to disagree with you on Google creating a special case for certain sites' nofollow. This would create an overall nightmare as I assure you SEOs would already be all over it by now and the spam would get a lot worse. Professional SEOs still use Wikipedia even for referral traffic. I am pretty sure the cat would be out of the hat right now. But again, I truly believe I am wrong and you are right for all our clients' sake (that wish by itself is proof enough for me to know that Google will not go for it).
MG All the way. Any data on Social Media marketing is good data.
I for one would love to read her blog post - I'm interested to see which way people perceive the sexism to exist.
Please post it Rand.
What exactly was Danny's question? I get the general idea but no one seems to be saying just what he asked Matt!
And I'd love to read MG's post :)
I'm all for Mystery Guests post.
good post Rand.
As a Frenchman, and working in SEO, I do agree that French people tend to click more on a .fr rather than .com link. Could it be that the French arrogance is showing up on the web, while choosing between 2 links ? But one thing is sure, is that French people tend to trust more those .fr sites
Regarding the Google Weblmaster tool and the geolocalisation for a website : I just hope that this tool works fine, and will give even more options in the future, especially for major sites that are using for instance English in many countries.
I think we Brits also prefer a .co.uk address particularly for online transactions - it's a trust thing
I forgot the mystery guest's post : Yes, she should do it, so we could see what will be the reactions for that kind of post.
Domain-Level PageRank
I was chatting with Rebecca about this the other day... PR0, brand new page.. can rank #1 out of like 200,000,000+ listings.. why?The experiment returned only one conclusion..Domain PR.... _
Are SEO Flags a Bad Thing?I'd estimate from experience that 40%+ of Fortune 1000 websites and 60%+ of the top 10,000 most important domains on the web do SEO of some kind.
(you would be SO surprised.. met with two fortune 100s last week.. neither was doing SEO other than PPC and were NOT running metrics on organics.. notice i said "were" lol..)
=)
<2 cents>will go look for mystery post.. but if you are the kind of man i think/feel/perceive/assume you are... based on electronic musings from DMOZ to here.. people slamming mystery guest might not be good for Rand's being perceived as "cool and calm"... trust me.. i'm the nicest guy ever.. or a total ArseHo!3, depending on the situation..
me defending a female friend would be one of those being precieved as an "arseh0!3 moments" .</2 cents>
That PR0 page - did it have 0 internal links? In which case that could be evidence of domain rank if its external link profile didn't justify the rankings.
I had the pleasure of meeting Geraldine and she has plenty of spunk to defend herself (probably better than even Rebecca can defend herself).
Post it . . . if she gets heat for it, then she's a very intelligent woman and can defend her points with the best of them.
Brent
Seriously? That's pretty sucky.
Damn taken literally I guess. This response by Rebecca proves solidly that I was wrong. Rebecca has more spunk for sure!
BTW, I was speaking metaphorically but . . . lesson learned. ;-)
Rebecca . . . you know I love you (and I don't mean that in a sexist, offensive, literal way). Then how the hell do you mean it Payne? Who knows . . . I don't think that hard.
Payne
lets see the mystery post from mystery guest.