• http://www.michael-martinez.com/ Michael Martinez

    I turned off Instant Search. It’s just too mind-annoying.

  • http://www.gulpmedia.com Jeff Rosen

    Danny –

    Great post. I thoroughly enjoyed it.

    Full disclosure: My knowledge of Search and SEO is just above zero Kelvin, but you hit on some very interesting points that made me take a moment and think.

    Imagine having to index the world’s telephone numbers and all you were given were the listing name, telephone number and the number of times that number has been dialed.

    You can’t verify the identity of the account name and with respect to businesses may not even be listed under the “Doing Business As” name.

    You will not be notified if someone (or a business) moves. You will not be notified if the business closes or expands.

    You will have some success using area codes and exchanges to geographically represent them, but they may have their numbers forwarded to another line in another location.

    How effective do you think a search for a name or term would be?

    This may not be a great analogy for what Google accomplishes, but in my very humble opinion, sure seems to be.

    Granted, they have the added benefit of combining that data with scores of additional terms as they crawl actual pages, but are those same terms indexed in context?

    I have experimented with varioust formulas and have concluded that in the future, sites powered by people, will offer much more effective and valuable results.

    Enjoy your weekend.

  • Stupidscript


    Nice detail of some of the Google relevancy problems. It almost made me forget the reason I clicked through to read this article. I was expecting an article with relevance to the question: “How The ‘Focus on First’ Helps Hide Google’s Relevancy Problems.”

    Seriously interesting is my behavior … I was presented with a link to an article written by a certified authority, identified by its title and a brief sales pitch; and based on the title and sales pitch, I calculated that the article was relevant to my interests, and I included reading the article in my index of relevant activities.

    And yet, the article, despite the title and the authority SEL/DS hold, was practically irrelevant to my interest … which is about how “Focus on First” helps to hide the relevancy issue, not about the relevancy issue, itself.

    It’s almost as if SEL made the same mistake as Google … intending to present relevance, but tripped up by their own cleverness.

    Fascinating … physician, heal thyself … yaddayadda. Still, a nice article, Danny, even if it did not live up to its billing.

  • http://andybeard.eu AndyBeard

    Poor Jeffrey of SEO Design Solutions
    3 years a Sphinn member
    1400+ Sphinns cast… very few submits but a large number were other people’s content.
    Outed on Search Engine Land for having a WordPress plugin that isn’t bad… has a few unique features, and from memory an option to give a credit link to the author.

    He writes some good content, has 12,000 RSS readers etc

    At least one regular writer on Search Engine Land has had similar success with WordPress plugins.

    Even worse you didn’t even mention his company name so it won’t even come up on a Google alert.

    I also wish I could see the same search you see – best to somehow have personal off, a fixed geolocation etc.

    I am not seeing Jeffrey anywhere close to first page with a slightly better query for demonstration purposes.

    I could give you an explanation into why Google doesn’t rank for search engine… the simple reason is that it ranks for other things instead.

    Matt Cutts probably gets tons of search traffic for cats and Linux so doesn’t rank quite as high for some stuff.

    Going any deeper would be delving into my weird freaky wacky stuff for which I have no conclusive test results.

    If you wanted SEL to rank for Search Engine (at least a little better), you might think to construct your internal linking such that the first text link to the home page isn’t “home”.

    There are tests out there that show that a text link takes priority over an image link alt.

  • http://searchengineland.com/ Danny Sullivan

    Andy, that set is from January. The set I see now is largely the same, except the bottom two don’t show for me. I get one geo-located SEO firm ranking, because Google makes some assumption that I really want an SEO firm located near me. I’ve also noticed that geo-located results can’t be turned off. Used to be able to, but somehow that seems to have been dropped or forgotten.

    As for the firm, it wasn’t outed on Search Engine Land for having a WordPress plugin. It was outed as a “why is this ranking” question as part of my presentation. You think that’s the 9th best page on SEO in all the world? And if we’re talking about “hey, the firm has a great blog,” then Google should rank the blog, don’t you think?

    The answer as to why it was ranking it that it seemed to have collected plenty of links from clients and other places that Google would have us all believe don’t really count. But clearly they do. Clearly they still work for other people. I thought that was worth illustrating.

    Google should rank for “search engine.” I don’t care really what the explanation is that you have, or heck even that I have, about why it doesn’t. It should. That’s the point. It should, it doesn’t, and it illustrates a flaw in Google’s search engine.

    Same thing on your suggestion that changing one single link ought to push us further toward ranking for “search engines.” Ideally, Google should do the right thing. We’re a site about search engines. That’s obvious. Google knows this.

    Heck, Andy — our name is “search engine land.” Do you know how many links, external links even — we have to us using our exact name in the anchor text? Plenty. Plenty that you’d think we’d rank for the singular “search engine” by now. Nope — page three, behind a lot of old search engines, some failed ones and older information sites.

    Bottom line, we’re competing against really old links that others in our space have. That’s an issue — not just for us, but for anyone coming in. And if Google can’t solve it, then people will keep turning just like that firm you’re concerned about to what does work, getting a lot of links however they can.

  • http://www.cmsbuffet.com/ CMSBuffet

    Great post Danny,
    I am glad Google is using stickiness as one of the parameters. it seems like this is the best way to remove my competitors’ bad content and good linking strategy from Google page 1.

    in the long run, content will get you to the top.

  • http://andybeard.eu AndyBeard

    I haven’t found a way to turn GEO off other than using the query I suggested.

    Alternative is to specificy a location


    “How, Seriously How”

    I look on that as not just a question on why a particular site is ranking, but also of relative worth.

    Website owners choose what they want to rank for and use optimized content and links to tell Google what their pages are about.
    Anchor text from other sites confirms (or not) that that is what the content is about.

    Altavista even based upon OpenSiteExplorer data has links from 60K domains, and the most popular non-brand keyword used in Anchor text is search engine.

    Google don’t want Google.com to rank first for search engine in Google – doing so from a business perspective isn’t logical.

    The most logical page to rank for Google for search engine is
    http://www.google.com/cse/ and after that maybe site search and enterprise.

    Google even block the link from the CSE page to the Google home page with the following link.


    How you link to yourself really matters – I have thousands of links to me for Niche Marketing and they are good links, every time I appear on Techmeme, Adage150, lots of top ranking lists that themselves get lots of links and authority etc.
    It could be blamed purely on a change of title tag, but the title tag still has marketing in it (diffuses Google bomb), there are lots of related synonyms on the page etc.

    The biggest changes however were when I changed sitewide linking to the home page, and when I dropped category listings (that removed the specific term form the home page)

    It isn’t a highly competitive term, but the drop is from 2nd or 3rd down to the second page of results.

  • http://searchengineland.com/ Danny Sullivan

    Andy, Google’s a leading search engine. Arguably, the leading search engine. It doesn’t matter what they do or don’t do in terms of trying to show up for “search engines.” The fact is, they list other search engines there but not themselves. If it’s relevant to list some search engines, then it’s relevant that Google should list itself. It doesn’t. It’s a flawed results set because of this.

    SEO certainly can help influence whether something will rank well for a term, but you know it’s far from perfect. And plenty of sites that to all the wrong things from an SEO perspective may still rank well, primarily because the links make up for whatever their errors are.

    For us, it doesn’t make a lot of sense to make the first non-image link on the page — “Home” — say “search engines” on the off-chance that this one factor alone might push us up. It makes sense from a human perspective that we have a home link there, for people trying to navigate and who don’t consider to click on our logo.

    We could add a navigational link called “search engines,” and maybe that would help — but it’s not like our SEM, SEO and PPC links are doing much in that regard. Those are more recent, however. Perhaps they might pay off.

    But still, in the end, we have so many external links pointing to us and saying “search engine” as part of the links, since they use our name, and we still don’t pop for that search. Very trustworthy links, too. That should count far more than internal linkage. It’s clearly not. Why is unclear, but I suspect that despite the external links, we still get done by the “old man” link factor. That, and also that Google may apply a strange mix of relevancy for different classes of searches.

  • http://purdue512 purdue512

    Danny –

    I know you were invited to make criticisms, so that was the entire point… And forgive me if this seems like undue criticism myself. Overall, I thought you raised a number of interesting points here. Specifically, the overemphasis on age and links…

    However, I also felt that your arguments are a little bit fantasylandish (a word I just made up:-). Sure, you can take anything and say how it can be better. I find that every Monday-morning quarterback has that routine down.

    “I’d like all 10 results for a query to be absolutely great.” — Seriously???

    The elephant in that room is SUBJECTIVITY… What you think is great is not what someone else thinks is great. This one drives me nuts because people talk about SPAM as if it is black-and-white… Not so. SPAM is really a sliding scale… And should really be called “QUALITY” or “COMMERCIAL INTENT” – which is the best I can figure that it really is. Or perhaps there is some “SPAM LINE” that justifies the label once it is crossed. I once listened to Matt Cutts say “I know SPAM when I see it, I can almost smell it.” And part of me agreed with him… But that’s a far cry from being able to build a computer program to reliably identify it. Are we really expecting Google to produce TRUTH? Doesn’t it take peer-reviewed journals from leading scientists to product truth? And even then, they usually don’t agree… :-)

    What really interests me are ideas that explain exactly how some complex process might be improved, with an emphasis on the practical. And in the case of search engines, we are talking about implementation at a staggering scale. I didn’t see much here to address that. In the end, Google has an extremely reliable process. The bottom line is that if you ran every other query in Bing/Yahoo vs Google, and recorded your satisfaction with the query results (which will mostly be driven by relevancy), I’ll bet you will see a serious Google advantage.

    Again – I know you were invited to make criticisms, and that’s what you’ve done… Perhaps your idea was just to dream up what *could* be better and leave it to others to translate into practical implementation. Good work if you can get it ;-)

  • http://GeeKen GeeKen

    My problem with Google’s relevancy is how searches increasingly don’t think my search terms are relevant.

    I find more and more often when using several terms to narrow down the scope of my search that Google will ignore one or more words, so that I have to force them to be considered with a + or – before the word.

    If I didn’t care if the word was to be used, I wouldn’t have typed it in to begin with. Too me that is a result of sloppiness, not just relevancy.

  • http://www.seodesignsolutions.com seodesignsolutions


    Damn??? Since when am I not invited to my own roast? And thanks Andy for at least chiming in with semblance regarding why we were pegged for a flogging.

    For the record Danny, we ranked organically for 1.5 years + (between SEO Moz and SEO Book) in Google on Page one for SEO “Before we released our WordPress plugin” in 2009.

    The links referenced in the link:seodesignsolutions.com query screen capture you illustrated was a result of those sites linking to us from the attribution link “which has an optional toggle to disable” in all fairness.

    Also, the inbound link velocity (at one point 150K links in one month) was in fact responsible for us tripping a link velocity filter and slipping (for that keyword only) from our prestigious perch.

    So, don’t make snap judgment chronological assumptions about our firm, air it as spam, then think its ok “just to poke a stick at someone for ranking for a VERY competitive keyword.

    We worked very hard for 2 years to produce that position; so, I don’t necessarily appreciate being made an “example of” particularly when the context is on the topic of someone “deserving” to rank on your personal editorial process of what you think Google’s take is on what is relevant as “SEO”.

    I have respect for you for your contribution to search and coining the term, but that was a roast on SEO Design Solutions merely because we offered a plugin and had a wide array of link diversity.

    Let’s leave the sorting to the algorithms as well as the snap shot shotgun approach of who, what, where, when and how… ok?

  • http://searchengineland.com/ Danny Sullivan

    I didn’t poke a stick at you, nor did I say you spammed Google.

    I poked a stick at Google, for saying that it’s all about getting high quality links to rank and yet, when I did a backlink look up, I got lot of stuff that didn’t feel like quality. Not spam, but not necessarily quality, either.

    If you think your company is one of the top 10 things that should be listed for “SEO” in the entire world, I get that. But I don’t know that most people would agree.

    You can’t service the entire world, for one. There’s no evidence that you are better than many other firms (and equally, no evidence that you are worse). There’s no particular reason for your company to have been there other than, as best I can tell, Google liked your links.

    That’s Google’s problem, not yours. That was the point of this article. Google would suggest that it’s great content gets great links and that leads to top rankings. In reality, it leads to a few good sites and then sometimes a crapshoot about what else goes on that page.

    That fact that your firm disappeared, and that there wasn’t a huge outcry, is a pretty good sign that Google’s searchers were in particular missing you. And similarly, if we were there — and then went away — I doubt searchers were notice. They’re interested in “good enough.” But “good enough” is not what Google tells the world that its results are all about.

    To get into the specific about the link, OK — so apparently, you had a bunch of links from your client sites. Apparently, they could toggle these off? That’s fine — but then is that the best way to decide who ranks for SEO? Because someone can go out and offer really, really low cost (and probably cruddy) SEO and do the same thing, and out you go again.

    Now apparently, you were ranking well for SEO and then decided to put out a WordPress plugin, completely with SEO as a link, because what — you weren’t feeling solid enough? And that “inbound link velocity” harmed you. So what was working wasn’t broken until you broke it?

    This was an overview of my take on search quality (and yes, it’s my take) against the stuff that Google says is suppose to matter, in an area where I feel I’m a subject expert (it’s hard for me to say, for example, what seems odd to be ranking on botany).

    I’m sorry if you feel you were made an example of. That wasn’t my intention. It’s one reason I didn’t call your firm out by name (as Andy suggests). If it were intended as a roast, I’d have done this about your firm. There’s no need for that. Your firm didn’t do anything wrong.

    What it did do, however, was stand out as a sore thumb in the results. It (and there are plenty of other firms like this that come and go) makes someone think “how” or “why.” Why is Google putting you there above all other things, if the quality of links is supposed to be so important. And what I see is that what Google says and how it acts are radically different things.

    Which I know you know! And I’m sorry you got caught in the crossfire on that, me poking at Google about Google’s quality. But maybe it’ll help you and your clients down the line.