Does Facebook help SEO?

[hoangvu]

Will having a Facebook account and fan page and getting lots of shares and likes to website. Do those links count as backlinks or not and also the URL in Facebook page help site's SEO?

 

[savidge4]

Ultimately no.. all the likes and shares and all that does nothing for backlinks. The most important aspect of social platforms is actually the profile page. Properly developing the profile, and then using say schema tagging to connect your site with that profile.. this starts developing a chain of authorship - and that can be a good thing.

 

[HCFGrizzly]

To answer your question let`s see the definition of SEO:

Search engine optimization (SEO) is the process of affecting the visibility of a website or a web page in a search engine's unpaid results—often referred to as "natural," "organic," or "earned" results.

The problem with the links on social media is that the links are marked as nofollow, and that meant that the links don`t pass PageRank.
If PageRank is not passed => the visibility of a website in SERPs is not affected => SEO is not affected.

In short: NO, Facebook doesn`t help with SEO

 

[SenseiSteve]

There is a great article that addresses this here. In the article, it quotes Matt Cutts from Google discussing ranking signals from Facebook and Twitter. On the surface, the answer is no.

 

[ericplotz1]

ericplotz1

That article was from 2014 and long before Google began indexing a small percentage of tweets. Now how does a tweet being indexed have no SEO value? A good profile has a solid mix of nofollow and dofollow.

There's a good case study on another forum where a guy is ranking sites with only social signals. I haven't checked back on the thread in a while, but he was on the first and second page for multiple keywords last time I checked. You could just Google social signals case study or something like that. Look at the dates and find recent threads.

I ranked a site as a case study, before the rankings plummeted due to being hacked and locked out of my site (freakin' yoast plugin vulnerability).

Social signals are becoming more valuable.

Think about it, when you go to a blog post there are share and like buttons. Not "link to my post contextually" buttons.

 

[PTTed]

PTTed

I Googled and searched high and low for a valid study that proved social signals will improve rankings. Can't find one. Love a link to one if you have one.

I found some studies, all of them were inconclusive and/or murky at best.

See here is the thing. You can create one dofollow link from a page that you know has some PageRank. Let's say you have a page that you are guessing is equivalent to a PR1 value back when Google still published PageRank values. You can achieve a PR1 value easily just by creating a 10 page blog where every page on the blog links back to the homepage while having zero other outbound links to anywhere else. Once you can see that all 10 of those pages are indexed in Google, then the amount of PageRank of the homepage of your blog will be approximately PR1 give or take.

Now create one outbound link from that homepage and point it at some page on some other site that you want to rank higher in the search results. As long as the competition isn't fierce, you will actually see that page move higher in the search results meaningfully. And the higher ranking of that page will stick.

If you pick a low competition keyword, you might move your page from ranking number 25 all the way to ranking in the top 3 with that one link. And it will stay there unless you do something to screw it up. It will stay there for years without you changing squat as long as the competition doesn't radically increase.

When you run a test like that, the results are perfectly clear. You can see that your one single stinking link meaningfully affected rankings. And you know that it was that one link that did it. And there is no question in your mind that the link did that.

The problems with most studies, especially these studies with social signals is that the people who are seeing what they think are good results are not isolating the social signals in their tests. They allow too many other variables to come into play.

If social signals meaningfully affected rankings and were actually worthwhile focusing on for SEO, then you should be able to conduct the same experiment I just mentioned. Except instead of creating one dofollow link, just go to Fiverr.com and buy some tweets for your page or blast it with Tweets from your favorite social drop feed service. And see if that page moves higher in ranking and stays there. If it does, then you can surmise that Tweets help.

Every time I have tried this or anything like it where I isolated the social signals from everything else, the result was the same. Social signals didn't do jack.

Every once in a while I will see someone authoritative in the SEO field run another test to see if things have changed and social signals are now having a measurable effect on rankings, and every time one of these tests comes out, the answer is the same. The test showed they didn't move rankings like you would expect if they truly counted.

There are all these rumors out there that social media boosts rankings, but I can't find one single study that proves it does. And it is really easy to conduct a clean test. It should be really easy to provide concrete evidence if it actually worked. Where is this evidence?

 

[samimnoorzaitgc]

Facebook doesn't affect a site's SEO. It'll just help to drive some of your friends to your website. If you want FB clicks from huge crowds of people, you'd better try advertising then.

 

[jeffhowell]

Some say it does but I would say it doesn't ... as it doesn't render any direct links to the pages that you link with - at least I haven't seen it...

 

[ericplotz1]

I found some studies, all of them were inconclusive and/or murky at best.



When you run a test like that, the results are perfectly clear. You can see that your one single stinking link meaningfully affected rankings. And you know that it was that one link that did it. And there is no question in your mind that the link did that.

Thanks for providing that value and people will learn a ton from you pointing out linking to the homepage.

You are speaking about the present because of the way the algorithms evolved in Google, but not the way they are evolving. I spoke of how they are evolving. Does that mean we ignore the present strategies? No. But that does not make the future irrelevant either. Observing the human behavior across the web is SEO, not understanding what those have already done. Thats like SEO 101. SEO is intuitive ingenuity. This comes after learning all the crap thats been blogged about 10 times already.

i'm not sure how your "social media study" was conducted, but there is a science to everything.

Social signals included. Can you rank with only a blast of fake social signals? HECK NO. I use fake social signals and a site called:

socialnetworksignals.com

Not real but as real as fake gets, and they work.

The problems with most studies, especially these studies with social signals is that the people who are seeing what they think are good results are not isolating the social signals in their tests. They allow too many other variables to come into play.

How did we not isolate the social signals? We cannot say that one tweet or one pin or whatever can affect a site, no. But what studies have concluded is that multiple social signals tiered up with likes and shares (I throw in some RSS stuff too) can boost the site dramatically. We've tested this on brand new sites with newly registered domains.

We cant control if someone sees a tweet or post and links to it from their site. Which hardly EVER happens. I've had pins from one of my Home and Garden sites get embedded by mommy bloggers, but not much direct linking to a post. And definitely no contextual, natural links.

So I'm not sure what you mean by "allow too many other variables".

I will say that I have never seen the SS alone take me to the front page, but surely 2nd and 3rd. Now mix that with your backlink test and what have you got?

 

[PTTed]

PTTed

In order for you to conduct a test to prove something with a high degree of certainty, you have to remove as many other variables as possible that could affect your results. That is the thing that is missing when people do these tests. In almost every case, there are other variables that could have altered the results thereby leaving you still questioning what caused what to happen.

That is what I mean by that. I can't find any studies that have done that and proved that social media links directly impacted rankings. I challenge anyone to find one.

Another thing, when you do a ton of SEO work over a period of many years, certain tendencies begin to reveal themselves. For example, you realize certain things like how much it helps to have a keyword in the URL when you are trying to rank for something. Because you have seen something like that help so many times, pretty much every single time you try it, then you know with confidence that adding a keyword to the URL when you create the page will help that page rank higher. You can say the same thing about internal links. You learn from experience that when you point more internal links at a certain page on your website, it helps that page rank higher. And you know it with confidence because it works pretty much every time you try it. And you know that if you get a stand alone contextual link from a juicy page, you know with conviction that this link is most likely going to help you rank higher. And you feel that way because it works that way nearly every time you try it.

But with social media promotions, where is this conviction? It isn't there at all. There is zero conviction about it.

Do you notice how there is nobody trustworthy out there claiming that getting juicy links to your website doesn't help rankings? There is nobody out there saying that because everyone who has ever tested it knows and sees with their own eyes that links work for boosting rankings. The same can be said for every single other major ranking factor in Google's algorithm. Things like keywords in the page title and keywords in the URL and keywords on the page. Nobody argues against these things because it is obvious they work pretty much every time you try them.

Yet social signals don't. They don't work like that at all. If they did, then every person who belonged to some goofy thing like TribePro or any other social media link service (pick one) could easily dominate the search results using their competitive advantage. Maybe they wouldn't dominate, but they would be ranking high, wouldn't they? If they were, everybody would be doing it. And you know as well as I do that it just doesn't work like that.

The only thing you will see experienced SEO's saying about social media promotions is that "doing real social media promotions can lead to acquiring natural inbound links that will ultimately affect rankings". In other words, the social media links themselves don't do jack. (at least not for Google) The only benefit you get is a "possible" secondary benefit of getting your page seen by someone who might then choose to link to it. And that natural link is what would ultimately boost the rankings.

If you have a social media source that is allowing you to create dofollow links and the pages those links are on are able to be crawled by Google, then they are probably helping to some extent. But the vast majority of social media "likes" "links" "shares" or whatever, don't create those kinds of links that Google counts at all. A lot of them can't be crawled by Google at all. And most of the ones that can be crawled are nofollowed. And they are also spammed to death which would prevent Google from ever wanting to use them as a source of web trust anyway.

If you honestly think that creating these (fake, syndicated or whatever) social media links are helping you rank higher, I challenge you to try all of the same tactics you use other than social signals and just quit the social signals for a few months. You will discover that your pages do just as well without them as they did with them. And you will begin to realize that just maybe - those fake social media promotions weren't doing jack squat for me all this time.

Here is an excellent study (by an extremely trustworthy individual) that basically proved Google Plus shares don't meaningfully affect rankings. That study was conducted after the announcement by Moz that Google Plus shares correlated well with high rankings. That Moz correlation study is an example of why people think that social media links work for SEO. But when you test it, you realize they don't work. And the Google Plus accounts that were used in the test were mostly authoritative accounts by people who are picky about what they share.

 

[savidge4]

savidge4

I personally would say that even this study is flawed... the initial release of both articles include 6 G+ shares... How can you be testing the effects of G+ vs no G+ if both the test and base are using G+? makes no sense really.

Here is the thing... I have found that using feedburner that is attached to G+ to be a very effective tool. There are pretty instant results - IE your G+ post of new content on your site is listed usually page one for a number of hours. Very identifiable... very testable.. release a new article without the use of feedburner and then share to G+ no link to your G+ content.. use feedburner and have it attached to G+ and low and behold there is a ranked listing ( for a short amount of time ) to your G+ copy of the content.

Google is real keen on Authorship.. I know I know Google Authorship got shut down a while back..but any time you can directly link content to its source Google actually prefers this. There are a few ways you can do this. In todays SEO environment the use of Schema tagging does this very well. you can lay claim to all sorts of content that you have out in the e-world. Its simply understanding how to do this.

So back to the test... IF you were to use feedburner to instant post on your attached G+ account. Google KNOWS where that content is coming from. ( you have just set authorship in place the easy way ) In mid moderate to low competition terms you will see a serp listing in the top 10 linking to your new G+ post - just short of instantly - we are talking minutes to a few hours. Sharing helps this process to keep a listing there.. and here is where my "test" would be flawed. Its not about the shares so much as it is about the clicking that takes place in those hours that your new content serp listing is in place.

Clicking on any link you have in the serps is a good thing. If you are looking for the placement you have on a search and you find your page and dont click on it.. well your an idiot. OR better yet you click on it and then hit the back button... Serp listing clicks is the unspoken element of SEO. We all know when you first list a page it bounces.. page 1, page 4, not listed, page 13, page 1, etc I believe this is Google looking for its click potential.. and more importantly checking your content for User Experience related variables.

The issue with Social only testing and how it effects SEO is that there is no way to keep the sharing and linking purely within the social platforms. the moment a page is shared on someones blog.. the test is out the door.. a "traditional" back link has been developed.

 

[PTTed]

Here is the thing... I have found that using feedburner that is attached to G+ to be a very effective tool. There are pretty instant results - IE your G+ post of new content on your site is listed usually page one for a number of hours. Very identifiable... very testable.. release a new article without the use of feedburner and then share to G+ no link to your G+ content.. use feedburner and have it attached to G+ and low and behold there is a ranked listing ( for a short amount of time ) to your G+ copy of the content.

Google is real keen on Authorship.. I know I know Google Authorship got shut down a while back..but any time you can directly link content to its source Google actually prefers this. There are a few ways you can do this. In todays SEO environment the use of Schema tagging does this very well. you can lay claim to all sorts of content that you have out in the e-world. Its simply understanding how to do this.

So back to the test... IF you were to use feedburner to instant post on your attached G+ account. Google KNOWS where that content is coming from. ( you have just set authorship in place the easy way ) In mid moderate to low competition terms you will see a serp listing in the top 10 linking to your new G+ post - just short of instantly - we are talking minutes to a few hours. Sharing helps this process to keep a listing there.. and here is where my "test" would be flawed. Its not about the shares so much as it is about the clicking that takes place in those hours that your new content serp listing is in place.

Your point is valid but off topic. Getting a parasite page (Google Plus page, Facebook profile page, Twitter profile page, or whatever) ranking high in the search results is a valid approach to getting traffic. However, it has nothing to do with getting your existing page ranked higher which is the premise being suggested by the people claiming "social signals have a positive impact on search rankings for pages that get shares, likes, or whatever".

The issue with Social only testing and how it effects SEO is that there is no way to keep the sharing and linking purely within the social platforms. the moment a page is shared on someones blog.. the test is out the door.. a "traditional" back link has been developed.

Ah, but there is. You can create an HTML page using a text editor like Notepad. You can upload the page to your domain using an FTP program. If you don't link to the page and you don't tell Google that this page exists, Google will never find the page and it will never get indexed.

Then the only way Google will find the page is if you build a link to it or if you submit it to Google for discovery. These pages are called "orphan" pages. They have no inbound links whatsoever and are not part of any sitemaps and basically don't exist on the web until you force Google to discover them. You can use pages like this to conduct a wide variety of tests including things like:

1) Does a link pass PageRank or not
2) Is a certain tactic effective at getting a page indexed or not
3) How much PageRank gets passed by a certain link
4) How much PageRank does one link pass compared to another link
5) How much does one on-page SEO tactic influence rankings compared to another one

And by using multiple orphan pages published on multiple sites you can conduct analysis on things such as - does domain authority actually exist and how much of a ranking factor is it compared to other things. You can also determine how much one ranking factor counts in relation to another one. Which one is more powerful?

The thing is, almost nobody bothers to take the time to do these tests. If you did, you would find out the real truth about SEO and what works and what doesn't. It takes very little money to conduct these tests. All it really takes is the time and willingness to do the tests and a little bit of creativity to figure out how to eliminate other variables that affect rankings so that you can isolate which variable is causing the impact.

 

[savidge4]

Ah, but there is. You can create an HTML page using a text editor like Notepad. You can upload the page to your domain using an FTP program. If you don't link to the page and you don't tell Google that this page exists, Google will never find the page and it will never get indexed.

Then the only way Google will find the page is if you build a link to it or if you submit it to Google for discovery. These pages are called "orphan" pages. They have no inbound links whatsoever and are not part of any sitemaps and basically don't exist on the web until you force Google to discover them. You can use pages like this to conduct a wide variety of tests including things like:

You have an orphan page no links no indexing.. Im with you so far.. if the idea of the test is to tell how a page gets indexed and ranked you obviously cant have a no-index command on the page. So what happens is you share content on Social, and then someone takes that content and shares on their blog... you now have been indexed and have a backlink. Concept of study has now been blown. THIS is where these studies fail.

The other aspect of this is on-page SEO. If you are isolating a page as we are discussing here..you are working at a SEO disadvantage... All of the standard signals that we all know to help in page rank would have to be removed.. well if they are removed.. how are you supposed to rank a page?

Your point is valid but off topic. Getting a parasite page (Google Plus page, Facebook profile page, Twitter profile page, or whatever) ranking high in the search results is a valid approach to getting traffic. However, it has nothing to do with getting your existing page ranked higher which is the premise being suggested by the people claiming "social signals have a positive impact on search rankings for pages that get shares, likes, or whatever"

this is where the feedburner G+ connection gets interesting. the initial G+ listing does not stay ranked for very long. hours to maybe a day at the most - IF you really start looking you will be hard pressed to find G+ content ranked in the serps. Using this method, you are dictating the origin of the content.. the post itself and NOT the social property re-posting. Google will ALWAYS ( well tries to ) include the original content in the serps no? The only variable in this is IF the G+ content is the original source.

Authorship plays a HUGE factor in this whole thing. Google wants to display original content from its original source. if you are will nilly posting content without efforts to show the chain of authorship you are fighting a losing battle. If you want to win with Google, you have to play by Googles rules... Authorship is an extremely under utilized variable... there is plenty of instances where google has made it clearly not so clear how to do this. Schema IS todays version of Google Authorship. Utilizing this is how you rank page content by sharing socially. its not the social that will rank, it is the original content on your site that does.

IF I were to run a test... I would index 2 pieces of content and get them in the serps in a steady position... its at this point that i would introduce the social sharing to one piece of content and see what the difference may be. This is honestly a test I will never run, because I see the advantages of instant indexing using the methods I do. there is a clear distinction between using feedburner and G+ vs not doing so. Does it effect the overall ranking.. I am not so sure I could say yes to that.. but does it get my content in the serps and stablized faster so I can make the needed SEO corrections to increase the pages serp rank.. heck yeah it does.

 

[PTTed]

You have an orphan page no links no indexing.. Im with you so far.. if the idea of the test is to tell how a page gets indexed and ranked you obviously cant have a no-index command on the page.

I would not put a no-index command on the page. You want the page to get indexed when you announce it to Google. The point I am making is that the page won't get indexed until you tell Google it exists by either submitting it to them or by adding the page into a sitemap or by creating a link to the page. That is the whole point. If the page is not indexed at all, then there are no other factors influencing rankings of the page at all until you tell Google the page exists. Get me?

The whole point is to eliminate every variable you possibly can so that when you send a ranking signal of some kind, you can measure (with more accuracy) the effect of that thing you did. When an orphan page first gets indexed, the only things affecting that page's ranking are going to be the domain authority of the domain and whatever SEO signal you chose to test. You can eliminate on-page SEO factors by not using them. You can eliminate domain authority by conducting this test on a brand new domain that you just registered. That would set the domain authority at its lowest possible level.

If you wanted to test social signals, you would take an unindexed orphan page and point whatever social signal at it you wanted to test. You can buy Tweets for it if you want or create them yourself. If the page is not indexed and you Tweet it you will see if the Tweet even gets the page indexed (hint - it won't). And you can send 10 Tweets at it if you want. And you can send 100 Tweets at it if you want. or whatever. Or you can Google Plus it or you can Facebook share it or whatever and test to see what happens.

So what happens is you share content on Social, and then someone takes that content and shares on their blog... you now have been indexed and have a backlink. Concept of study has now been blown. THIS is where these studies fail.

If your followers are so link happy where they automatically link to your material from their blogs out of their generosity or because your content is so awesome, then I congratulate you. But that ain't reality. When the average person Tweets or shares one of their blog posts, the only inbound links they typically get are from the social media sites they announced the page on. Very, very rarely does a person start getting any natural links at all. That only happens in the right industries. Most of the topics people write about and share in social media get absolutely zero natural inbound links. That is why link building is such a pain in the ass to begin with. If it was as easy as you are suggesting, everyone would be pure white hats all the time. It doesn't work that way. Its not even close. If it works that way for you, the I sincerely congratulate you. It doesn't work that way for 99% of people.

And for the purpose of testing, you could just create a new account at some social media site and buy fake followers for the account. Or if you don't want to spam, then just create a separate legit account and test it. It's up to you. Conduct the test however you want to.

The other aspect of this is on-page SEO. If you are isolating a page as we are discussing here..you are working at a SEO disadvantage... All of the standard signals that we all know to help in page rank would have to be removed.. well if they are removed.. how are you supposed to rank a page?

Again, the point of the test is to eliminate every other variable so you can see what is causing rankings. So you want to intentionally eliminate (and/or) control every variable possible. Otherwise you can't get a clear picture on what is causing what to happen. Get me?

You can eliminate on-page SEO by creating a brand new word that is not in the dictionary right now. Include that word on the orphan page. When that page gets indexed, it will rank for that new "imaginary word" that you made up. You can take that very same page and publish it on 3 different domains (each having the exact same amount of domain authority) as an orphan page on each domain. Then you can change one thing you are doing to one page and measure the results.

Or you could publish the same article on the same domain 3 times all as orphan pages and just change one letter in the page URL. Call one page - dog.html one page fog.html and one page log.html

Add in your SEO signal and see how Google ranks those pages. If you Tweet the dog.html page 100 times and it doesn't outrank the other pages, it is safe to assume those Tweets were completely worthless for SEO.

By doing a test like that you can easily see whether a suspected SEO tactic is legit or not. You can prove to yourself what is truth and what is not. It sounds complicated, but really its not. You just have to think about how to remove every variable possible. Once you conduct a couple of these tests you will see how incredibly helpful and clarifying they are. You will know with certainty whether something is an SEO signal or not and you will get a really good feel for how much of a difference maker it is.

 

[SEOPub]

I'm with Ted. I have found zero credible evidence or a single case study isolating social signals, or even attempting to isolate them, as a ranking factor.

The only time I have seen anyone publish anything saying that social signals are a ranking factor are when they are pushing something social media related. For example, Moz... notice their Twitter tools they are trying to sell you? They have an invested interest in people believing that social signals matter. Whenever I have found a "case study" on a forum, it has been published by someone selling some sort of social signals in their signature.

As I have said many times, from a business perspective it would make zero sense for Google to use social signals in their algorithm because there is no guarantee that they will continue to have access to that data in the future. There is a lot of it they cannot crawl now. They could also be blocked from it at any moment in the future.

For awhile, Twitter was charging Google for access to their content. That agreement expired, and for awhile Google stopped crawling Twitter to any significant degree. Google cannot crawl Twitter regularly because it will cause performance problems for Twitter. They have to make a special arrangement to feed Google the content?

So what happens when that agreement expires again?

What about Facebook? Facebook and Google have anything but a friendly relationship. If Facebook ever decides to really move into search, what are the chances they block Google from crawling any of their data? I would say pretty high. Facebook doesn't rely on Google for traffic, so there is nothing to lose there on their end.

 

[savidge4]

End / Beginning of the month.. been busy sorry...

Ok so you develop 2 Orphan pages... one just sits there ( in theory this will never be indexed or ranked ) and the other gets 100 twitter likes... I cant say i have ever done this.. would be worth trying I guess.. but I would think that there is the possibility at the very least the page gets indexed and maybe to some degree ranked. because the orphan page would be stripped of many SEO factor we know to help in the ranking process.. I would think that it would not rank well.

But lets throw in a third page. As I keep describing.... Publish a page.. feedburner passes onto G+ we already know the page will be indexed.. and Yes I understand it has nothing to do with social. but shortly after we can see that the G+ content gets ranked. ( its not hard to set up and you can try this for yourself ) Like I have been saying.. the G+ listing only lasts for a short time hours to a day generally. After that point the page is ranked - to clarify again.. the G+ content listing falls away, and the page content is basically taking its place.. maybe not its position, but indexed and ranked none the less.

IF you are to remove the G+ portion of the above scenario things are more drawn out... and as stated in the 2 orphan example because of the lack of SEO indicators on the page ( for the sake of determining Socials worth ) It may or may not get indexed.. and it may or may not get ranked.

It IS the G+ social aspect in this that helps in fulfilling both requirements; Indexed and Ranked. - This plays within the same line of thinking as SEOPub states... why would Google use 3rd party data that they may or may not have access to tomorrow... He is correct, makes NO amount of sense. But G+ is a whole other animal... they WILL have access today tomorrow and for as long as they decide it is worth their time and money. G+ is Google.. so there is no 3rd party issues.

 

[PTTed]

Ok so you develop 2 Orphan pages... one just sits there ( in theory this will never be indexed or ranked ) and the other gets 100 twitter likes... I cant say i have ever done this.. would be worth trying I guess.. but I would think that there is the possibility at the very least the page gets indexed and maybe to some degree ranked.

If you were creating 2 orphan pages for a test, you wouldn't let one of the pages just sit there unindexed. You are confusing things.

If I want to test whether tweets will get a page indexed, I would just create one orphan page and tweet it. Or you could create 2 or 3 or whatever and tweet them all. It depends on how many times you want to repeat the test to convince yourself of what is the truth. It's totally up to you. Because I have done these tests before and I know how to do them eliminating every variable that matters, I would only create one page in that instance.

The reason to create 2 or more orphan pages is to test ranking effects of some SEO technique. When I test such things I create 3 orphan pages so there is more data and the answer is clearer.

If a Tweet won't get a page indexed, then it is kind of pointless to test whether it affects rankings or not. Common sense and logic tells you - it doesn't. But, I suppose in theory it is remotely possible. So, if you want to test whether Tweets affect rankings, then I would create 3 orphan pages. I would submit them to Google so that Google indexes all 3. When all 3 were indexed I would write down the order in which they rank. I would wait a week. Then I would check rankings again to make sure nothing has changed. (it shouldn't change. If it did, then I would wait another week before tweeting anything) Once the rankings were stable, which should be right away, then I would Tweet the lowest ranking page of the 3. I would check monitor the rankings by checking on them manually once per day over the course of a week. Then once per week for a few weeks. If the page that was ranked the lowest moves into the top position and stays there, then you could assume that the Tweet(s) caused it. If you don't believe your test, just repeat it with 3 new orphan pages using a different made up word.

because the orphan page would be stripped of many SEO factor we know to help in the ranking process.. I would think that it would not rank well.

The key to the test is to use a made up word that is not in the dictionary or make up a fake name for a person that isn't a real name like Porganways or something made up that doesn't exist when you Google it. Make up a new word every time you want to conduct the test.

The reason you make up the new word is because then - only the pages you created will be relevant at all for that word. And because of that, you will see only your pages in the search results in Google. Google will show you all 3 of your orphan pages because they are relevant for that keyword. And it will rank them in a certain order. Then you can apply your SEO tactic and measure the effect. This test has worked every time I tried it and I have done numerous tests like this.

Google is happy to show you your three pages even though you have removed every single other SEO factor except the existence of the word on the page. Because you created the page this way, there is effectively NO SEO on the page whatsoever. You have to make sure you make every page identical including the positioning of the made up word. Because (keyword position within a document is an on-page ranking factor, albeit a smaller one). So you have to make them all identical so that Google has to choose which page to rank first without having any clear SEO benefit. That way your test will be as sensitive as possible and measure even the smallest SEO impact.

You want to eliminate every variable you can possibly think of including things like domain name length and URL length and anything you can possibly think of so the test is as accurate as possible. I know it sounds complicated, but its not. When you go to run the test you begin to think of all these little things that could impact the results even slightly. And you figure out how to remove them.

By doing these tests you learn an exceptional amount of unbiased truth about SEO. You even learn small tidbits about SEO and how Google works that nobody talks about on the internet because they don't even know about them. Well, I am sure a very, very small percentage of other SEO people know about some of those things, but they don't talk about those things publicly on the internet.

 

[savidge4]

savidge4

I am going to say thatI am doing the exact opposite.. my test theory is way cleaner and would deliver the desired result based data. Take a ste back for a moment.... what would we be testing? Is it the ability of a Tweet to effect SEO? Is it the ability of a tweet to index a page? I think we need to get even more basic than that and start with a bench mark that can be CLEARLY tested. Does Google in any way shape or form use tweets as an SEO variable.

Again if you create 2 Orphan pages... one by its lonesome self ( the benchmark or test base ) and then a page that is also orphan and the only connection to the outside world is twitter. IF the twitter connected page does NOT get indexed..there is no sense in further testing if there is SEO effect. The test is done finished and over. Simplicity with minimal / NO exterior variables.

The reality is we really dont even need to run this test.. we can simply look at serps and determine the answer to these questions... On a positive note.. lets look at images first. Google indexes and displays images correct? Looking at image listings we see images from Pinterest, from Instagram, from Flicker etc.. when was the last time you saw a Facebook meme? ( actually linked to facebook ) the answer is you dont. This in un itself is a glaring indication of the communication between Google and Facebook.

When you do see Facebook property listings.. what exactly are you seeing? you are seeing USER profiles. You are NOT seeing timeline posts.. but again ONLY user profiles. It could then be suggested that some time needs to be spent in optimizing your Facebook profile no? It is listed ( indexed ) in Google search ( it is delivered as a result in Serps )

This same logic and analysis can be done over and over Digg, LinkedIn, Tumblr, FourSquare, and the list goes on and on....

Simply put your test logic is flawed.. it is introducing way to many variables that could possibly skew the results. Those variables are not need to determine if there is a communication / web / bot / spider link between 2 web properties. If Google in whatever way has a connection or access or what have you, the content on the 2nd party site would be indexed.. or the property that the 2nd party site links to would be indexed.

Will tweets index a page? its from this point you could dig deeper and test the theory of if a tweet could effect SEO outcomes.

Lets look at Twitter.. seen any "tweets" in a Serp listing lately... either have I. Now lets look at G+ Have you seen a G+ post in the Serps.. I know I have...

 

[PTTed]

PTTed

Dude....All I can say is OMG.

I am sorry I wasted so much of my time in this thread. I am sorry if you can't grasp what I tried to explain to you there. I am also sorry if you can't follow the logic. I don't think you understood any of it.

Hint - Just so you know, you can't figure out most of what Google is using in its search ranking algorithm by just looking at SERP's. If you think you can, I promise you that you are wrong. You can get hints and clues and some obvious things jump out at you, like the importance of page titles and keywords in the URL and the apparent importance of things like domain authority and inbound links. But you can't tell many of the other things, like which links Google is counting (and how much they affect rankings), unless you actually test it using some method similar to what I explained to you repeatedly (as clearly and detailed as I was able to). If you just study the SERP's without testing and you try to make logical conclusions based on looks alone, many of your conclusions will be dead wrong. That is why so many people believe things that aren't true about ranking. They assume correlation = causation and it doesn't. That is why people believe things like "nofollow links help rankings" or "old domains get a ranking boost over new domains" or "most social media links boost rankings in Google" or "you need a mix of nofollow and dofollow links to have a natural looking link profile" or pick one of the many other totally wrong interpretations of the search results. I have even seen some people dream up other completely unfounded nonsense like only pages that rank high can help other pages rank high.

Whether or not you see Tweets ranked in the search results is completely irrelevant to whether or not Google counts Tweets to pages as part of their ranking algorithm. Whether or not you see Google Plus pages ranked in the search results is also completely irrelevant as to whether or not they count those links as part of their search algorithm. Completely and totally irrelevant. If you don't understand that basic premise, then I am wasting my time in this discussion completely.

It is apparent you simply don't understand what I was trying to explain to you. And perhaps we are at two totally different levels of understanding of how Google works. I don't know how much experience you have or your background and you don't know how much experience I have or my background, so maybe that is why I can't get through to you. Maybe I am just really, really bad at explaining things. That must be the case.

Actually, achieving clarity is one of the things I find most frustrating about written forums. It is so hard to explain things in words alone without writing a ten thousand word essay. And if I am going to write that much, then I am going to sell it to you. I sure as hell am not going to give that knowledge away any more than I have already done. I should have never taken my explanation as deep as I did on a public forum. Foolish me. I give up.

 

[EpicGlobalWeb]

Will having a Facebook account and fan page and getting lots of shares and likes to website. Do those links count as backlinks or not and also the URL in Facebook page help site's SEO?

Facebook links are no follow. You get nothing from Google by them. They may have an impact on your "authority." The page itself will be indexed so keep that in mind when choosing a page name. Facebook traffic should be largely regarded as independent traffic. About a fifth of the world's population uses it, so it's internal ranking factors are certainly measurable. Ads on it are very cheap and the demographics you set can be extremely precise.

 

[vishwa]

Facebook does help you to drive traffic to your site. however it has zero effect on SEO ranking factors. I am also receiving good amount of traffic from my Facebook page, But doesn't find any evidence of getting better ranking with it.

 

[savidge4]

In the link that you posted: [URLnofo]https://www.stonetemple.com/measuring-google-plus-impact-on-search-rankings/[/URLnofo] as you read through the test you get to the indexing portion of the text. in that portion there is this paragraph "At least one person dropped one of the pages into a Flipboard magazine, so we did see visits from a Flipboard crawler from time to time to some of the content. Based on the checking I have done so far, I do not see evidence that these Flipboard pages are indexed, so I dont think that Google would have used those links to drive indexing, but I welcome corrective feedback from Google on this point."

Considering the post was published on 9/17/2013 one would assume that Google is keeping their mouth shut OR the statement is true. Through my tears of experience and testing.. i am partial to this being a true statement... No index of content - and Google has probably laid the link to rest. - I am in no way saying Google is what you see is what you get.. BUT there are clearly some indicators to be looked at.. maybe not in what is there, but often times what is not.

Because I am new to the forum.. let me share some of my history with you. Over the past 20 years I have spent much of the time as a SEO specialist. Its what i do.. and whether you believe it or not, I am pretty good at it. About 10 yrs ago I started into site development, with the idea of building sites from the ground up with proper structure etc - instead of coming in after the fact and re-working an entire site. I still do this today. About 4 years ago, I went more towards the conversion side of site development. I was building sites and developing traffic, now I do that as well as make them convert as well.

My current daily routine is 80% CRO ( Conversion Rate Optimization ). Because this is what I do.. It means I pretty much am running tests all day long. I am a testing fool. When testing, the one thing that bothers me the most is that little moment of doubt in regards to your testing model. In the study linked to above that moment is actually shared " What are the possibilities of corruption of this part of the test? " and then followed up with " There is a microscopically tiny possibility that someone implemented a link during that 6 minutes " and then previous to that, this little tidbit " It is possible that links were implemented to the pages that did not show up in our monitoring tools. This is not an insignificant potential problem, as by my estimate the cumulative links found by Open Site Explorer, Majestic SEO, Webmaster Tools, and Ahrefs is probably at best 50% of the total links to a site, and it may be as low as 30% "

I really dont like the idea of the possibility for 70% room for error. I work in a world where my testing goals are 85 to 90% significance. I will always start testing from the very simplest of terms. As I have explained above... An orphan sitting unto itself, "should" clearly not be indexed... and the key word here is "should" but that as I see it is a very clear baseline. With my baseline I am not having to worry about any amount of room for error. sure I say "Should" ( I have actually had a page in a test similar to this get indexed once ) Might be correct and 100% correct are 2 differnt things.

Indexing both pages.. again you are opening up your test for the possibility for error. The same in the linked test... If you are trying to determine if G+ effects Ranking.. then why in the heck would you submit the Test and the Base page to G+? that makes absolutely no sense. The only thing accomplished by the test was a pretty decent write up on the difference between 6 and 40 shares on G+. Kinda silly really.

In regards to sharing our knowledge... I have read your stuff here... I can tell you only scratch the surface of what it is you "know" and what you share. I personally... I share pretty freely.. the stuff we are even discussing here.. its to much like work and people in general wouldnt follow through with it. They would continue searching for someone to just give them an answer vs finding out on their own. much more beyond the level of SEO we are discussing here and I think a few of us could start sounding like hard core weirdo conspiracy theorist. SEO is freakin technical... its not all flowers and just write for the people crap.. it really is science

 

[PTTed]

In the link that you posted: [URLnofo]https://www.stonetemple.com/measuring-google-plus-impact-on-search-rankings/[/URLnofo] as you read through the test you get to the indexing portion of the text.

I didn't conduct that test. I can't defend it or explain it. The only point in linking to it was because I asked other posters to link to their evidence of social links boosting rankings. So I thought I would throw a bone by linking to a test that said these social links didn't boost rankings. That is the only purpose of the link. I don't speak for the guy who ran the test. Discussing it doesn't honestly interest me one bit. Believe it or don't believe it. Love it or hate it. Your choice. I'm fine with it either way.

An orphan sitting unto itself, "should" clearly not be indexed... and the key word here is "should" but that as I see it is a very clear baseline. With my baseline I am not having to worry about any amount of room for error. sure I say "Should" ( I have actually had a page in a test similar to this get indexed once ) Might be correct and 100% correct are 2 differnt things.

I don't even know what you are talking about here. We are not on the same page. It is clear to me that you don't understand the reason and purpose for the orphan pages or the role they play in the test. They are your primary tool for removing variables from your test.

Indexing both pages.. again you are opening up your test for the possibility for error.

If you get an orphan page indexed by manually submitting it to Google, you are not influencing the ranking of the page at all.

If you build links to the page, then yes you are influencing rankings. When you conduct a test to test an SEO tactic, you submit the page to Google manually first. Once it is indexed, it will rank based on its on-page SEO and the domain authority of the domain it is published on. As I explained previously, you can control both of those factors and nullify them at the start of the test. I explained how.

So you have 3 pages indexed in Google that are essentially (for all intents and purposes) exactly equal in SEO scores for ranking for that "made up word". And because their SEO scores are equal, you can measure (with very good precision) exactly how much your SEO tactic affects rankings. And then you take the knowledge gained from that experiment and you build on that knowledge by adding in other SEO factors. Your Google rankings for that "made up word" will fluctuate according to the power of the SEO tactic you are applying to your test page. You only change one thing at a time and measure its effect.

This type of test has an extremely high level of precision. If you think about the tactic you want to test and you prepare your test carefully, you will learn what works and how much it matters and you will have a very, very strong conviction of your findings because it is irrefutable.

I cannot explain it any clearer than that without laying out an exact example. And like I said, if I was going to do that, it would take me 10,000 words and I would charge big money for it. And the average SEO schlep won't pay a reasonable price for what that knowledge is worth. They would rather just read what other people say and keep guessing what might work and what they think should work instead of just finding out the answer with certainty.

SEO is freakin technical... its not all flowers and just write for the people crap.. it really is science

In the grand scheme of things (all things SEO) there are really only a handful of things that matter. All of the other things are so minor they are almost negligible and only really applicable in low competition situations (where they are extremely useful). There are a handful of powerful SEO factors and only a handful. It has always been that way, and continues to be that way even after every single one of these Google updates, major or minor. And social media links isn't one of those major factors. In fact, it is completely and totally negligible according to my experience. Social links for SEO are generally a waste of time.

Pages on social media sites and links from those pages, as far as I can tell, are treated absolutely no differently than any other page on any other site. If the link is nofollow, it doesn't count, period. (most are nofollow) If the page is crawlable by Google (many social site pages aren't crawlable) and the link is dofollow, then the link will probably count. The extent to which it counts depends on how much PageRank is passing through that one particular link. And I'm not talking about that original PageRank algorithm that everyone can read about on Wikipedia. I know for a fact (well with a high degree of certainty at least) that Google isn't distributing PageRank the same way that original algorithm did. I know that with conviction because the very tests I described to you here in this thread proved it to me.

 

[savidge4]

I don't even know what you are talking about here. We are not on the same page. It is clear to me that you don't understand the reason and purpose for the orphan pages or the role they play in the test. They are your primary tool for removing variables from your test.

Because we are not communicating in 10,000 word shots to explain ourselves, I think we are both feeling misunderstood. I do understand the use of the orphan pages. As I have tried to explain, I use them in testing as well.

If you get an orphan page indexed by manually submitting it to Google, you are not influencing the ranking of the page at all.

If you build links to the page, then yes you are influencing rankings. When you conduct a test to test an SEO tactic, you submit the page to Google manually first. Once it is indexed, it will rank based on its on-page SEO and the domain authority of the domain it is published on. As I explained previously, you can control both of those factors and nullify them at the start of the test. I explained how.

It is this that you and I may agree / disagree. I understand your indexing the page. you are specifically looking at the SEO effect of in this example how twitter may alter page rank. My hmmm looking for words.. greatest "concern" here would be Click Rate. 2 3 10 pages indexed for an off the wall term, and 1 of those pages gets a serps click. I, a bit above "believe" that single click will alter the outcome of the test. As I see it, this would be an induced variable that you would have little to no control over.

Click rate is something that I personally have tested quit a bit, and yes there is benefit from getting clicks in serps vs not getting clicks. Are people looking at your page or are they not, which do you think Google wants to be listed? Even with this, there is a host of other variables that come into play which further complicates things.

As I have tested these things, I am simply looking for the ability for 3rd party web properties ability to index my orphan page. I believe that this is a clear indication that there is the possibility of a positive SEO benefit. I do go on to further test this, but come out of a simple test, into something a bit more complex. I get a number of pages indexed and ranked. I let the position stabilize for a period of time. From the initial set of pages I then select the closest matching of the bunch and run testing from there.

All of the other things are so minor they are almost negligible and only really applicable in low competition situations (where they are extremely useful).

I am a HUGE low comp fan. Greatest bang for the buck if you have the understanding and time to seek out this type of specific term. I would rather have 90% of 10 searches a month over 1% of 1000 searches a month. Then throw in the consideration of conversion percentages, and I think this becomes a no brainer.

 

[PTTed]

PTTed

How in the world are you ever going to get any clicks on any of those pages?

The only thing any of those pages will ever rank for is that "made up word". And you are the only person on earth who is searching for that "made up word".

You can randomly pick a sentence out of your article and Google it. None of your pages will rank for them. Maybe they could rank for an exact match search if you grabbed a couple of consecutive sentences and did an exact search on Google using quotes. But nobody is going to do that. These newly indexed orphan pages are among the lowest ranking pages of all the pages in Google's index. Absolutely zero visitors are going to accidentally find them. Friggin zero dude. These pages have so little Google PageRank that they are barely hanging on to indexation as it is. In fact, if you don't send any PageRank to those pages, Google will eventually drop 2 out of 3 of them from the index. That is how weak those pages are. They rank for one keyword and one keyword only - the made up word. Other than that they won't rank for jack squat, ever, period.

I can't believe I even have to explain this. How is it possible you would not know this already based on your experience?

The newly indexed orphan pages will get zero, nada, zip, zilch SERP clicks unless you click them yourself. They won't even show up in a SERP high enough to get a natural click. There are millions of pages in Google's index that never get any SERP clicks. And these newly indexed orphan pages rank even lower than many of those.

And even if you do click them, as I have done at times, clicking on one of your articles a handful of times changes absolutely nothing about the way they are ranked. It doesn't affect the rankings at all.

The only people who know these pages even exist are you and Google. And Google thinks your pages suck from an SEO standpoint. They aren't highly relevant for any search phrases at all. The only reason they rank for the "made up word" is because they are the only 3 pages Google knows about in existence period, that even have that word on the page.

Your test pages never get indexed in any of the other major search engines unless you point links at them or submit them to those search engines. And you are running these tests over a period of a couple months. These pages aren't even alive long enough for anything natural to magically appear and impact them.

Okay, so I have told you that SERP click rate is a non-factor in these tests. What other SEO flaws are there in the experiment?

 

[savidge4]

savidge4

And how in the world are you going to see SEO effect on a page that has some "Made up word" that has 3 pages listed in the serps as a target.. because I would sure hope you are ranked #1 for that term the moment its indexed... Oh thats right.. you are sand bagging the page so it lists last, so you can test if there is an increase.

I would have to wonder how many tests you have run to this point that are returning false negative because the test pool is so shallow a piece of paper can touch bottom and still be dry on the top side. For that matter I have to wonder how many are returning false positive.

And then you throw in the "And even if you do click them, as I have done at times, clicking on one of your articles a handful of times changes absolutely nothing about the way they are ranked. It doesn't affect the rankings at all." oh ok.. yeah that has no effect - my bad ( insert sarcasm ) I believe you. There are way to many published case studies on this one.. the only negative one I can find is by Bartosz Góralewicz and he is so focused on beating out Moz that he totally bypasses the gains in his own findings

Like I said you and I would either agree or disagree on that one.. you dont agree and thats fine.... Im all good with that. Our testing methodologies are different thats all.

 

[PTTed]

And how in the world are you going to see SEO effect on a page that has some "Made up word" that has 3 pages listed in the serps as a target.. because I would sure hope you are ranked #1 for that term the moment its indexed... Oh thats right.. you are sand bagging the page so it lists last, so you can test if there is an increase.

Because you said that, I am going to flat out question your knowledge of Google SEO period. You don't even know what you don't know. You obviously don't understand how Google works even on a basic level if you ask that question. I suggest you start here.

Watch that video a couple dozen times until the basic stuff starts to sink in for you.

The reason Google only returns 3 pages in the search results for that test is exactly because there are only 3 pages (on the entire freaking web) that are relevant at all (even in the slightest) for that made up word. That is why Google only return 3 results. And that is exactly why the test isn't biased at all. My friend, that is as basic as SEO gets. Learning that is like learning how to do addition and subtraction before you try to learn algebra.

There are only 3 results returned. So they rank 1, 2 and 3. Get it? You can actually conduct searches on Google that have less than 10 search results. I just showed you how. I let Google decide what order it wants to rank the newly indexed orphan pages in. I try to remove all SEO from the page, therefore I don't know what order the test pages are going to rank in until they actually show up indexed in Google.

Then I apply my SEO test to the lowest ranked page. Why? Because if that page moves above the other 2 pages, then you can assume your SEO tactic caused it. If it only moved up 1 position, then your SEO tactic is not effective at all. (I have never seen it do that anyway). In order for the tactic to be proven effective at all, the lowest page would have to move into the number one position and stay there. That tells you that it worked.

Remember when you are first starting out, the SEO score of each of those 3 pages is identical (or as identical as you can make it based on everything you know). If the scores are all identical (or as close to identical as you can logically make them) then your test will be highly sensitive to even the smallest SEO advantage such as adding the keyword on the page a second time.

You don't have to remove all SEO from all the pages if you don't want to for some reason. The goal is to make them all exactly equal in score. For example, you could add the keyword into the page title on all 3 pages in the exact same position (keep the pages identical). You would still be able to conduct your tests with the same high level of accuracy. I prefer to remove all SEO other than the keyword on the page. The whole purpose of the made up keyword is to limit your experiment to the 3 pages you have control over.

You could not conduct the experiment with that much accuracy if you used a common word that many pages rank for. The reason you couldn't do it accurately is because you don't know the SEO scores of all those other pages you are competing against. You could only guess what their scores might be. And there is no way in hell you could guess accurately. You couldn't conduct a really clean experiment. This is the flaw with most people's tests. Most people conduct a test with variables like this - A + B + C + D + X + Y = Z They change A and Z changes so they say A caused Z result. But they don't know if B, C, D, X, or Y changed at all and impacted anything.

By controlling all 3 pages (every page Google would possibly rank for that term) you control all the variables. Thus, the whole point. Get it?

You only change one variable at a time while eliminating everything else possible to eliminate.

And because of that, you can actually measure how much of an impact something has. If you don't control all the variables then you can't get an accurate measurement. Instead you would have to conduct your test a whole bunch of times and say that your pattern shows that "this result is probably the case". But you would still be second guessing yourself.

For that matter I have to wonder how many are returning false positive.

Exactly zero false positives so far in my testing using this method. At least as far as I have been able to tell. Unless I just overlooked something which is possible no doubt. As I pointed out, as far as I can tell, there are no other variables other than the variable you introduce into the equation. Therefore the tests are quite clean, incredibly revealing and incredibly accurate, at least as far as I can tell. You watch the search results change directly as a result of your SEO tactics. It's wicked cool to watch. And it leaves you asking other questions about what else might work and how can you test that other thing. If you tried it you would see it with your own eyes. And you would get it.

And then you throw in the "And even if you do click them, as I have done at times, clicking on one of your articles a handful of times changes absolutely nothing about the way they are ranked. It doesn't affect the rankings at all." oh ok.. yeah that has no effect - my bad ( insert sarcasm ) I believe you. There are way to many published case studies on this one.

Notice how I said clicking on the result a handful of times (and that is over a period of a couple months as I also said) (like maybe one click a week - not even that much) has no influence on the rankings (at least not in my testing). I didn't say it wasn't possible to influence rankings by clicking on the search results. I believe it is possible - just not over a two month period with one click once in a great while. You just heard what you wanted to hear instead of reading what I wrote there.

Go conduct the test and see for yourself. I am not making this stuff up dude. Don't believe me if you don't want to. I don't care. I already know the truth. You are the one guessing.

The difference between you and me is that I am telling you what I saw in my own experiments. You are just telling me things you read about somewhere else over the years.

I would tell you to just go do the experiment and see for yourself, but I don't think you fully understand what I tried to explain to you. So how could you possibly conduct the experiment properly?

I'm sorry that these tests (as I have explained them) are not in your realm of understanding. And that's okay. Use your own methods. That's cool. I get it. Sometimes I think something is ridiculously easy to understand and can't understand how someone else struggles to understand it. The reverse situation has happened to me before too. It happened to me a lot back in college.

 

[savidge4]

savidge4

Here would be my issue with such a shallow pool... You and I both know there are positive and negative effects to actions taken in regard to SEO. I personally spend a lot of time testing the negative side. just how far you can push a variable.. and at what point it is a positive action and then when does it start sliding downward, and how far gets you sandboxed. I do this type of testing for a couple of reasons. #1 generally speaking, if you can get a variable to "Fail" that means the opposite should be true. That variable should have the ability to provide a positive effect. #2 Often it is harder to get a positive result from a single variable. but having the ability to get your listing to drop 50 points, then making the correction and getting the 50 back can give you a good indicator of what could be done to inch that variable along for a better overall positive response.

I would say because you have tightened up your test structure in such a manor that you are only going to see 66% of the total possible effect of every action, you end up missing half the data. In search there are 3 effects. Positive change, no change, and negative change. Using CTR as a great example here... Not having a description tag.. you get good rank, but your CTR is a bit lack luster. You then add a description tag and you realize that you are not losing rank ( or gaining for that matter - no change ) by doing so.... at first your CTR goes even lower ( boo hiss ) so you make a correction, and it gets better. You can then make further changes and optimize your CTR for that listing. for the most part in this process you have not altered the rank placement of the page. However towards the end of the process you might see an increase in serp rank. So a conclusion can be made... Description tags unto themselves do not effect rank, but an optimized description tag that increases CTR can.

What gets really interesting with this topic in particular.. ( Does facebook ( social ) help SEO? ) We are really looking at what I call a dimentional variable. What i mean is Likes, Shares, and even CTR in cases can be time ( event ) oriented. You place a piece of content on a social site.. in the next few days you get a few likes or shares. As Rand Fishkin tested, you can ask your social followers to type in a search term and click his link - rank was effected, but in time fell back to roughly its pre action state. Because the action peaked and wained.. so did the effect. Again seeing the positive and negative effect of any action in terms of testing SEO I believe to be important.

Again.. even in the last communication I got an idea that your test period is a bit longer than I suspected ( a good thing ) The issue you and I are having is the amount of depth we want to throw at the topic in such an open space. Neither of us are going to write 10,000 reports... even tho I think we are both getting close at this point. LOL We do things differently... If it works for you, what can I say?

And for the record.... As I shared in a previous post.. I test for a living.. I run tests all daylong.. I also read a ton of case studies.. I try to use case studies when discussing things like this to take the expertise off of what i think and know to be true, and share what others find. I have case studies I could pint you to, but would that increase the merit of my words?

 

[wmguides]

Social media and SEO are two different sources of traffic and the links from Facebook is not counted by search engines. But If the number of shares and likes are increasing day by day then it is a very good signal of a huge source of traffic in near future. Now a days when people visit a new site they also first check the social media pages of that site to see how many likes it has received and if a site has received a lot of likes then it will make a good reputation in the eyes of the first time visitors and then they come back again.