Mark peers into his tacklebox, looking for that perfect lure, knowing in advance that one particular fish on DP just cannot resist himself from posting on every LSI topic on this forum, telling everyone, what a load of bumkum the whole concept is in relationship to the search engines, hence this thread, to throw down the gauntlet to say to this poster and a few others like him, post up your concrete evidence that the major search engines do not and have never used, LSI. I think you might find that you are completely and unequivacably wrong about this particular issue, but here, be my guest, let's see your final concrete evidence that what you state is true, that LSI has no place whatsoever and is not used by the major search engines. If you cannot provide the concrete evidence to back up your claim, perhaps we can put this issue finally to rest. Rather than telling everyone that posts on the subject, or brings this subject up, that they do not know what they are talking about, either because they are 'newbies' here or whatever, let's see exactly what you've got in good old fashioned, down to earth plain English, with no technical jargon or gobbledegook to back up your statement that it's a load of hogwash, the asociation between the two. Mark has fixed his bait and with a clean sweep, his rod arcs heavenwards across the expanse of the forum, watching the bait plop gently and patiently watches his float in expectation of a good nibble, settling back to watch the show unfurl comfortably.
If you're going to sit back and demand concrete proof that lsi is not currently used, why are you not providing some kind of proof that lsi is currently used yourself? Just because something can't be proven, it does not mean that the opposite must be true. Obviously you can't prove it's use either.
That article just says that Google uses LSI to help recommend related searches: "The new technology will allow Google's search engine to identify associations and concepts related to a query, improving the list of related search terms Google displays along with its results". They go on to say that they want to use LSI in their ranking algorithm in the future. In other words, it doesn't sound like they actually use LSI for search engine rankings. It is a good concept, but it will take several years to properly implement this in their algorithm.
I'm just asking to see some concrete proof, that is all. If anyone wants to start up asking the opposite question, which is not the purpose of this particular thread, of course you are more than welcome to start up your own discussion on the subject and posing there, that subject question. However, it is patently obvious that this thread is not asking that question, it is asking however, to see the proof that the major search engines do not use LSI. Rather than derailing the subject matter therefore, taking this off into a subject area that here on this thread has no relevance, perhaps we could just stick to the question posed. Thank you. Let's see that evidence, that the major search engines are not using LSI. For some to state this supposed fact so categorically, you must surely have some evidence to back up your claim/s, or not, as the case may be. Well, do you or do you not, have that evidence to back up your claims? If you do have that evidence, let's see it. Then we can put this matter to rest and not have to endure upon this forum being called a bunch of idiot's anymore, so called without an inkling of understanding for the subject matter in question, as has been indicated in no uncertain terms. It rather irks me you see when people state so firmly that one does not know what one is talking about, so I thought to pose a simple question which should by rights be fairly easy to supply one with a fairly straightforward answer, please furnish me then, with said proof.
Start two blogs: find a keyword niche that can get you around 1,000 searches a month. Blog 1: put 5 well written articles SEO'd for the keyphrase. Blog 2: put 5 completely random text articles, SEO for the keyphrase. Build 200 links each. See which ranks better and put an end to this debate. (hint, #1 will rank better).
My friend, that is not the question. Maybe you mistook this thread for another one, I don't know, but how does that prove anything especially with furnishing this concrete proof? In a nutshell it does nothing to aid this thread whatsoever in the absolute slightest. I'm not looking for a discussion on this topic, which would only end in further drivel being posted, all I want to see is that concrete proof, nothing more and certainly, nothing less.
...not sure how it could be a more relevent use case. I thought LSI was the 'use of synonyms and keyword density to influence search engine rankings' From this standpoint, it seems easy. Two sites, one with "garbage" text that will not relate to the keyphrase you are targeting for, and one with "relevent" or "LSI Friendly" content. Do the /exact same/ seo techniques on each (keyword density in h1 tags, h2 tags, title, meta desc/keyword, and links) After all, without a controlled experiment, there can be no concrete proof EDIT: not entirely sure why this is off topic, or why you gave me negative feedback.... especially with this quote from your original post: "Rather than telling everyone that posts on the subject, or brings this subject up, that they do not know what they are talking about,"
this is like saying "lets see your concrete proof God does (or does not) exist" speculation, rumours and "faith"
Not exactly no, the two questions are as different as chalk and cheese. Whereas one can speculate about the existence of God or the non-existence of Him all day long without getting anywhere, unless He posts on a blog somewhere, this factor regarding LSI and it's non use in major search engines should be very well documented online, although interestingly enough, no-one has popped forth yet to provide a link to any website or factual document that can be viewed stating this position categorically. It's ok, I've got all the time in the world and if I'm proven wrong, that'll be fair enough. I'd just like to see a proper answer one way or another. Simple.
The above link by the way does not answer the question, rather it seems to back up my pov, so whatever you are trying here with the above reply, I have not the foggiest clue, no matter, your'e out of the discussion anyway. Personally I think you were getting a little mixed up on what the question is asking, but there you go. Another time, may I suggest you pay more attention to the words in italics.
I'm sure you won't get any useful responses after trying your hardest to be an ass. Who is going to try to help someone who is so blatantly rude to people who try to contribute to the conversation? I know I would never try to help such an ignorant prick. Good luck finding your concrete proof that you yearn so much. Learn some manners and people might want to help. For those about to reply to this thread: BEWARE THE OP WILL GIVE -REP TO ANY ANSWERS HE DOESN'T LIKE. So you are best to avoid this thread completely.
Which is which statement that you made? Concrete proof? Can't do it. Doesn't exist. What you are basically asking for is internal documentation from a search engine about their algorithms. The only other way to do it is by black box A B testing it yourself. I would love to be proven wrong though. If you, or anyone else, can provide a document where google/yahoo/aol that states "here is the math behind how X affects our rankings" please PLEASE let me know.
No, what I'm asking for is the proof that they don't do it and don't use it, that is LSI. That's all. PS Thank you mods for removing Vanderstam's few posts above, I appreciate that, I don't want to see anymore of his rudeness / personal threats by pm etc and that action was appreciated. My apologies to everyone else if the thread doesn't seem now to make too much sense in places due to the posts that have been removed. Let's get back to that original question if we may.
ok how about this? the man looks like he knows what he's talking about to me, and Stompernet do "ok" http://www.stomperblog.com/warning-advanced-seo-technique-does-not-work/
BWAhahahahahahahaha !! really? oh NO!! ...neg rep, what am I going to do? he neg repped me dear oh dear. chuckle chuckle...
The neg rep was for reply #10 where it looked as though you were going off at a completely different tangent in your first reply. Having viewed the video, he's hardly coming from an unbiased opinion since it's just a lead up to his subscribers for his own next big thing, his own product, it's not exactly neutral information is it? Certainly, there's nothing concrete as regards evidence presented there. Even the way he presents LSI isn't brilliant, it's presented negatively from the very start i.e. slam another concept, to give credence to your own product, engineer a vacuum and fill it with info that your subscribers will love and pat you on the back for. If his own product was that good, he wouldn't need to stoop to such a method to promote his own wares. He would present his own wares purely on their own merit, if they were that good. It certainly wouldn't need, any building up of suspense marketing strategies. His argument is flawed. Literally.
I understand it's a sales pitch and definitely has something to prove from the outset, but what about the indisputable facts regarding: * the patents - no actual mention of it. or since. * the SERPs - dont show LSI style results ie no plural / singular alignment * same results for different tenses? nope. * word pairing (synonyms) nope. * LSI doesnt scale past 1 million documents, Google handle 50 billion pages. * LSI is still actually only a theory, Google is real. and that Google go way above and beyond LSI processing, which would in effect hold them back. ? I dont profess to know much very much about it, is the description of it inaccurate then? the guy seems to know what he's talking about, and Stompernet are very successful.
I agree that Stompernet are very successful yes, there is no disputing of that fact. Personally I use long tail keywords with LSI to excellent effect, that is two words or more, the use in the video by way of an example, the mentions of simply adding an 's' on the end of any given single keyword, within the context of his argument against LSI, this point I failed to get entirely where he wascoming from. Again, within the body of his complete argument against LSI per se, I just thought that he was using really rather insignificant points which looked like they purported to back up his pov, when in fact for LSI to be successful it's not really designed, at least from an end users position, for words such as those illustrated. In a nutshell it was bending the truth quite a bit to make it appear that he is correct when in fact this was far from the case. If he had approached it without the introduction of his own 'replacement' product, it would give more credibility perhaps to some of his points in a very, very basic sense, i.e. he should have presented his case against LSI truly factually from a completely neutral perspective. All I know is that it does work and works well. The tool which I use doesn't cost $300 per month, it's about half that and a one time cost too, with free upgrades for life. LSI is of course not the be all and end all of good SEO, it's just another part of the process. As for your point about the first million documents, really from an end users perspective, I'm frankly not really interested in the websites on page 2 of Google or any major search engine, I'm interested in the first page keyword themes only and chances are that usually at most, for any given long tail keyword, the entire content of that first page, all the pages within each of those first ten websites, does not amount to more than several hundred or perhaps a few thousand individual pages at most and most often, a lot, lot less. Those are the sites I am competing with to displace them from Google One and replace with one of my own websites - if I can get number one position, I'm doubly pleased. At the end of the day, the Google algorithms will always remain a bit of a mystery. If you find that something works and is whitehat, use it, if it doesn't work for you, there are always other options, it's all good, as long as one is not trying to trick Google or any other major search engine.