Effectiveness of Link Popularity Decreasing?: Open for Member Discussion

[rustybrick]

To continue our moderator roundtable discussion on questions submitted by members (http://forums.seroundtable.com/showthread.php?t=115), we have post #7 (http://forums.seroundtable.com/showpost.php?p=958&postcount=7) by encanta who asks.

Does Google's apparent de-valuing of reciprocal links lead you to believe that the concept of link popularity is going to decrease in importance altogether in the near future? And, if so, how might Google compensate for the loss of this ranking factor? Perhaps they'll introduce human editorial review as a more important factor affecting SERPs? (This is probably just wishful thinking on my part )

Let the roundtable discussion begin...

[rustybrick]

Does Google's apparent de-valuing of reciprocal links lead you to believe that the concept of link popularity is going to decrease in importance altogether in the near future?
Link popularity is always going to be important in Google's eyes. The thing is they will always get better at determining what is a popular link versus an unpopular link. You see it today, but it will continue to get harder. Reciprocal links are not extinct, it is more about the type of reciprocal links you get, where they are from and so on.

And, if so, how might Google compensate for the loss of this ranking factor?
So if all your links are based on one type of link, i.e. reciprocal links, then you need to worry. You need to balance out the types of links you get. Directory links, one way links, reciprocal links, press releases, news sites, edu sites and so on. As they say, do not put all your links in one basket.

Perhaps they'll introduce human editorial review as a more important factor affecting SERPs?
I doubt it, Google is not known to, at a large scale, use human review for ranking purposes. They rather build technology to do it for them.

[dazzlindonna]

Not sure what I can add to the discussion, as I pretty much agree with everything Barry said. It is simply more important than ever to have good content that will attract good links. SEOs can no longer strictly depend upon recip links, forum sigs, etc. to make a big difference. We need to concentrate on standing out from the crowd. What makes your site different or worthy? If you can accomplish that, you'll have little need to even think about recip linking.

[pk_synths]

What I think really made an impact and made everyone question the value of a link was the devaluation of site wide links. These, in my opinion, are worthless now and changed the entire spectrum of the index. Allot of webmasters for years have been collecting these, either by reciprocal trade or by simply purchasing them. Google discounts these now and basically killed 2 birds with 1 stone. 1 bird being eliminating bought links and the other bird being removing sites that use unfair SEO tactics to rank (link farms, reciprocals, inter webbing, etc.)

This update pretty much targeted every SEOd website out there and only the authoritative sites survived.

Google will never eliminate the power of the link but they are definitely altering the way they judge them. No more is a link a link. It now matter if it's from an honest source, link history, age, varied anchor text, etc.

I agree with Rusty about an editorial review not being possible. There are simply WAY too many web pages out there and ranking one page over another is way too hard and open to favoritism. Last thing Google needs is the 1000s of daily emails screaming about bad editors, unfairness of rank, tattle tails (ex: "The site above me is using hidden text!! Do something about it") and I'm sure at least 100 new lawsuits each day from bad ranking websites.

Google established a ranking system and they will stick with it.

[rustybrick]

To expand on the site wide links topic, and it not being worth what is used to be at Google. Just recently, today and yesterday, we have two blog entries on the topic.

Yesterday, I wrote Site Wide Links; Not What They Once Were (http://www.seroundtable.com/archives/003150.html) based on a Cre8asite Thread (http://www.cre8asiteforums.com/forums/index.php?showtopic=32670). Then Patrick Gavin from Text Link Ads picked it up and wrote his blog entry today named Devaluation of Site Wide Links? (http://www.linkbuildingblog.com/2006/01/devaluation_of_.html)

[randfish]

Does Google's apparent de-valuing of reciprocal links lead you to believe that the concept of link popularity is going to decrease in importance altogether in the near future?
First off, I don't think Google is devaluing reciprocal links. I believe that's a misnomer. Google is devaluing what we in SEO call "reciprocal linking" where links point from one site's link page to another's homepage and the other site's links page points back, ad infinitum.

To suggest that all reciprocal links, like those shared between, say cnnfn.com and bankrate.com, from various pages no longer carry value would be incorrect. Just because a site links to you does not mean that by linking back to that site, in a natural fashion, you will lose any value from that link.

Google is trying to elminate manipulative strategies in order to have a more "unbiased" index. It's a logical and intelligent approach, and one that we should predict will continue for the future - if SEOs start to exploit weaknesses in search engines' algorithms, the SEs will look for ways to patch those holes.

On to the topic of link popularity in general.. Yes, I do believe that Google will put less total weight on links in the future and find ways to better analyze content quality on page, better associate specific manipulative tactics like link networks, etc. with sites and use data like search volume, traffic, searcher behavior, personalization, analytics, toolbar info, etc. to impact their results. It won't be a huge swing - links will always help to dictate relevance, reliability and reputation. But, the days when manipulating a link algorithm were easy are coming to a halt. Link spam and automated or non-editiorially-given linking will never end, but it will become harder and harder, just as cloaking, bait-and-switch and other search spam tactics have over the years.

And, if so, how might Google compensate for the loss of this ranking factor? Perhaps they'll introduce human editorial review as a more important factor affecting SERPs? (This is probably just wishful thinking on my part )
I'm with Barry a bit, though I do think there will be a lot of human reviewing in the most competitive sectors. I think Google can easily take the top 500,000-1,000,000 searched for terms and phrases and hand-check those results oiver the course of a year. If you're targeting highly competitive areas, I'd make sure that my site looked like it belonged in the top 10. Just pretend that you're standing on trial in court, forced to defend the reasons why your site is better than the thousands of clones out there - if you can present the jury with a compelling case, you've got a good shot to stay in the top 10. If you don't, you've got a good chance of having your links investigated very carefully, so make sure they can stand up to scrutiny if you're content can't.

[Phoenix]

Some really excellent responses from everyone. I hope I can add a bit to it.

I wanted to take off from Rand's explanation on reciprocal links. For sometime I have undervalued reciprocal links as a method to build links for a website. Generally it was if at last resort we would undergo linking building via reciprocal links. I do not believe they have been undervalued per se, but more that the websites that you obtain reciprocal links have little or no value. Its always a time equation for me and a lot of research undertaken in order to judge and evaluate the value of undertaking a reciprocal link with someone. There is a good deal of places that are not worth the time exchanging links with, and this has only been compounded by the lack of knowledge and reliable information webmasters have in order to wisely spend there time. Think outside and beyond the links page.

These days I often tell people to investigate paid link options, not necessarily from link brokers, but going out and approaching websites and offering to pay for link space on their homepages or subpages. Here I think lies a lot of value for the smart webmaster. There are a lot of websites out there not selling links, and there are many webmasters that would be happy to get a few dollars for a link each month. I have had some very successful relationships recently. One even sent me a Christmas present of handmade soaps, lotions, and stuff for spending just $15/month with them.

I agree with Barry in that "Link popularity is always going to be important in Google's eyes." They are just getting smarter, and so we should continue to do as well. While things will change and shift there will always be ways to exploit this and figure out new ways in order to get the maximum value and link can pass on to the website it links too.

[pk_synths]

To suggest that all reciprocal links, like those shared between, say cnnfn.com and bankrate.com, from various pages no longer carry value would be incorrect. Just because a site links to you does not mean that by linking back to that site, in a natural fashion, you will lose any value from that link.

To suggest that cnnfn.com and bankrate.com are equal to small mom and pop sites would be incorrect aswell. :) Owning an authorative website really makes you "above the law" nowadays. Owning a small, newer site is like stepping on eggshells. If your new I dont think reciprocals mean much unless you build up honestly and pay your dues. That could take years at times to do it naturally. Get greedy and you could be spending alot of time in the 500 ranks.

[dazzlindonna]

These days I often tell people to investigate paid link options, not necessarily from link brokers, but going out and approaching websites and offering to pay for link space on their homepages or subpages.

This, I think, is a great suggestion, and one that not enough people take the time to consider. It is amazing how cheaply you can get a link on a homepage from someone who doesn't know the value because they don't normally sell links.

[rustybrick]

Looks like we are past the question of is link popularity decreasing in value and on to ways to get links for our sites.

I think I will open this thread up to the public soon. :)

Outstanding replies guys.

[Darrin Ward]

I don't think reciprocal links have been made obsolete on the whole - but I do think the criteria for accepting recip links as being valid has become more strict. For example, 5 sites in the same IP C class all linking to each other are far less likely to be considered valid than a single recip link between sites hosted in different countries.

Also - links in themselves may or may not become less important, however I think how users interact between pages will play a significant role in the ranking of pages for a long time, probably forever. In this I mean that while PageRank and the actual "link text" may reduce in importance over time, the other factors that make people actually WANT TO CLICK on a link will remain of significant importance.

For example, the words included before and/or after the actual link itself (not within at <a> tag) might be more accurate at describing the target site instead of the actual link text itself, much in the same way as Google images determines the relevancy of images.

Remember Google hired a FireFox developer - well they did that for a reason. Robots/spiders are pretty dumb, and a significant way forward might be to look at how exactly a page might render in a web browser. Is there hidden text, are there css things etc...

One other thing - there are robots (including 1 major engine) out there right now that are reading and executing some relatively complex javascript on some of my client sites - so we're not too far off from from seeing advanced methods of web page rankings.

[randfish]

Darrin, your point makes me think strongly about how valuable the actual click data would be to Google. After all, PageRank was predicated on "random clicking" through the link structure of the web. If Google actually had the data on the links that get clicks, they could use that to help with weighting... Of course, I have thousands of IP addresses following links to my sites everyday, so that would put me on top of everything :D

PK - You're right, my example was shoddy. Let me make a better one. Matt Cutts linked to an SEOmoz.org post back in November. Later on, in my Beginner's Guide, I linked to a post on his blog. Just because I linked back to Matt's site doesnt' mean that the value of the link from him to us was lost... That was my point about reciprocal linking.

[Darrin Ward]

Randfish,

You are of course right. Google already have potential access to a very large portion of user tracking.

Think about this:

1. If a site uses the web version of Urchin - they basically know everything. Of course here they're given explicit rights to track all of that information - referers, keywords, pageviews etc...

2. If a site uses AdSense, they could potentially know a lot about that site too. The inclusion of a JavaScript file on a page means that a lot of things can be tracked. The primary function of the .js file in this case is to serve ads - but there's very few things that couldn't be tracked, just like with Urchin (Urchin being just a .js file too). They could potentially track referrer, page url, page title, time on page etc...

3. Google track what search results users click on. Nothing new here - we've know this for a long time. What is a sites perceived relevance for a given keyword as compared with point 4, which is:

4. If you use the AdWords conversion tracking thing, then they could potentially know a lot about a site - including what keywords resulted in sales (thus solidifying a sites relevance for those keywords). This is a very interesting one - what keywords result in sales - deduce that a site is of most significant relevance for those keywords!

[rustybrick]

Time and time again, Google's Search Team said they will never use numbers 1, 2, and 4 for ranking purposes. No one believes them. :)

[ShandyKing]

Time to implement #4 and test this out on my ecommerce sites.

[randfish]

It just doesn't make sense Barry. With all the trouble they go to to hold onto search data (like that stuff the US govt. is seeking right now), and then to say "nah - we won't use that." It just isn't credible.

It's like your best friend, who loves stamps and collects them like crazy, promising he'll throw out the ones from the postcards you send him while on vacation in Paraguay... not gonna happen - he just won't tell you where he saved it.

[gabs]

2. If a site uses AdSense, they could potentially know a lot about that site too. The inclusion of a JavaScript file on a page means that a lot of things can be tracked. The primary function of the .js file in this case is to serve ads - but there's very few things that couldn't be tracked, just like with Urchin (Urchin being just a .js file too). They could potentially track referrer, page url, page title, time on page etc...


Just an idea but onsite seo has always has had a focus on the on page position of your k/w so the position of the links of a page could be a major factor.

With the use of adsense data google must be able to credit the position of its links thus giving more credit to natural links in hot spots.. This would be taking steps towards the "how a browsers render a page"

[Darrin Ward]

Time to implement #4 and test this out on my ecommerce sites.
And please let us know.

Keep in mind that just because Google have the POTENTIAL to use this method to determine relevance, they might not actually be doing it!

There are countless other examples floating around in my head that show Google would be able to obtain statistical information to determine relevance - whether they use them (or will ever admit to it) is another thing, and is something that we as SEO's need to test and evaluate on an ongoing basis!

The last thing they're going to admit to is that they've found an innovative way to rank pages - otherwise they'd loose their competitive advantage!

[randfish]

And risk having all of us spam that method :D

[ste_dean]

Surely if Google was using this information or had plans to use it they would slip it out into the SEO arena some how...

It must be in Googles favor to get people to sign up to their services and what better way to do it than say it will improve your ranking!! I can see the stampede already.

[Darrin Ward]

Surely if Google was using this information or had plans to use it they would slip it out into the SEO arena some how...

It must be in Googles favor to get people to sign up to their services and what better way to do it than say it will improve your ranking!! I can see the stampede already.

On the face it might seem like that - but think of it in terms of business - who are likely to sign up to those services to benefit? Webmasters! Are most webmasters already signed up? Yes! Would webmasters use this info to manipulate (optimize) their rankings? Yes!

Net gain: loss of search results integrity through manipulation by webmasters, with minimal additional signups + now the competition could also use such methods to rank sites, thus loss of competitive advantage.

Note that I'm not saying they DO use this information - but they certainly COULD!

Some high profile corporate entites are very good at keeping secrets - particularly Google, they've surprised us in the past, and we still haven't a clue what their algorithm is. No - I think they'd keep such secrets very close, if they are true. Being that they are now public they have to keep their competitive advantage or shareholders would go crazy!

[ste_dean]

Agreed!

And it is certainly something I will be keeping my eye on!

[arteworks]

Hello:

New to the forum, but enjoy it.

It is my experience that reciprocal linking as an seo strategy does nothing recently. The best links are one-way, inbound links. While at one time Google considered a link to a site a "vote" for that site, in the case of reciprocal linking, they realized that such links were not so much indicative of a vote for a site than they were indicative of an agreement between two webmasters to exchange links for the purpose of artificially inflating their search rankings. Therefore, reciprocal linking met its demise.

IMHO, the best way to get one way, inbound links is not by purchasing them - you can get them for free by providing quality content to which people want to link! This is most easily done via installation of a weblog with outbound RSS and Atom feeds, with the feeds submitted to the feed directories. Also, use the article submission sites such as ezinearticles.com for content syndication. Each time you post an article to your weblog, also post it to the content syndication sites.

As people pick up your articles and feeds, you quickly obtain hundreds or thousands of one way, inbound links to your site - because of the quality content you are providing.

[JimNewsome]

This is most easily done via installation of a weblog with outbound RSS and Atom feeds, with the feeds submitted to the feed directories.

I like the way you think arteworks - I've been trying to sell RSS/Atom as a way of building relevant & genuinely useful incoming links for sometime, but it's not been easy. Do you know of any studies or testing that's been done on this - specifically on whether feed-based incoming links are being rated by the SE's and if so how much importance they are being given in comparison to standard links?

___________________________
Jim Newsome - SEO Consultant
http://www.omegadm.co.uk

[arteworks]

Hey Jim:

The only studies I have are the empirical evidence that it WORKS! Not to "toot my own horn", but I have a lot of articles about it in my weblog on my homepage. Also, do a search for something like "RSS and search optimization" and you get tons of data.

The fact of the matter is that, if done correctly (read: displayed as html and not javascript), RSS feeds create one-way, inbound links to the content provider. These are the most effective links you can have - and, in fact, the type of links the search engines most want to rank.

If people are linking to your content, that means that the content is worth linking to, and truly is a "vote" for that site, as opposed to purchased links (which google et al are in the process of figuring out how to detect and discount), and reciprocal links (which have already been discounted).

The only pitfall with RSS that I can see is that many people use javascript to display it, which the search engines can't read. That being said, if you get enough links out there, plenty are bound to be HTML and readable - so the invisible links just have to be considered as part of the cost of doing business.

Also, don't forget about content syndication - via the article submission sites - people just cut and paste that straight into weblogs and pages, with hardcoded links. No content provision is complete without content syndication, IMHO.:rolleyes:

[seojunkie]

*Much* has been talked about the topic at hand by the people above. Plus a lot of good points have been thrashed out.

Here is what I would like to add to the on-going discussion.

Does Google's apparent de-valuing of reciprocal links lead you to believe that the concept of link popularity is going to decrease in importance altogether in the near future? And, if so, how might Google compensate for the loss of this ranking factor? Perhaps they'll introduce human editorial review as a more important factor affecting SERPs?

As far as I think, Google isn’t devaluing the reciprocal links. Instead, it is getting smarter to better identify unnatural reciprocal links obtained utterly for manipulating purpose and finally tackling with them by diminishing their value in ranking factors.

This is NOT the case of reciprocal links getting devaluated at all!

There are a couple of legitimate reasons why swapping links with each other cannot be called ‘unnatural’ – either by a human being or search engine.

But, I think that Google is smart enough to better identify unnatural linking no matter it is camouflaged with reciprocal or triangular approach or any other way possible. At a certain level, of course! :P

Also, it is worth mentioning here what they state in Google Guidelines for Webmasters at http://www.google.com/webmasters/guidelines.html:

Have other relevant sites link to yours.

Submit your site to relevant directories such as the Open Directory Project and Yahoo!, as well as to other industry-specific expert sites.

And…

Don't participate in link schemes designed to increase your site's ranking or PageRank. In particular, avoid links to web spammers or "bad neighborhoods" on the web, as your own ranking may be affected adversely by those links.

Human editorial review is simply out of the question in (the near) future, because with the passage of time, their index will continue to grow uncontrollably.

To put in a nutshell…

IMHO, Google isn’t devaluing reciprocal links. It is *unnatural* reciprocal links that are being beaten by Google with each new Algorithem update. Link popularity isn’t going to decrease in the near future. It will remain as important as to date. In the end, Google will be able to develop clever tactics to sift through natural and unnatural links.

Cheers,
S

[spenno]

I have really given up on reciporical links. I only go for incoming ones now, I think they add more value from what I have read. I have to say there is a lot of conficting advice on the different forums.

There is a good backlink checking tool on iwebtool.com I do my search for my favourite key phrase, look at the sites with hogh page ranks and run their site through the tool. I visit the backlinks with the highest PR and link to them too...most are free directories, PR online or long threads on forums.

It seems to be working but time will tell on the next PR update (another 35 days?) My site is only a baby 3PR at the moment though, so I am no expert!!!! :)

Marketing Agency (http://www.grapevine-group.co.uk)

[Lorel]

Here is a quote from Matt Cutts back in May 06 re sites missing pages from Google's index and how it relates to reciprocals:

- Og.n March 13th, GoogleGuy gave a way for WebmasterWorld folks to
give example sites.
- After looking at the example sites, I could tell the issue in a few minutes.
The sites that fit “no pages in Bigdaddy” criteria were sites where our
algorithms had very low trust in the inlinks or the outlinks of that site.
Examples that might cause that include excessive reciprocal links,
linking to spammy neighborhoods on the web, or link buying/sellin

I think the key phrase here is "excessive reciprocal links" which tells me when a site has a majority of reciprocals then it's in danger of being penalized.

The problem is how does a new site, with no PR and no rank, get one way links without buying them or trading???

[fire_lady]

The problem is how does a new site, with no PR and no rank, get one way links without buying them or trading???
I agree that this would be a big challenge for webmasters of new sites. I suggest that they continue asking fellow webmasters of related site to theirs and go for directory submissions.

[masterofthe]

Not sure what I can add to the discussion, as I pretty much agree with everything Barry said. It is simply more important than ever to have good content that will attract good links. SEOs can no longer strictly depend upon recip links, forum sigs, etc. to make a big difference. We need to concentrate on standing out from the crowd. What makes your site different or worthy? If you can accomplish that, you'll have little need to even think about recip linking.

:D i believe good linking habits with regards to content is as hard to catch and get and become a linkbait itself if you do not have yet the high rankings in the Google SERPs because linkbait by providing good and juicy contents is just no use if you are not there in the top pages...it just is harder for them to link to you but of course you already have the edge at one point in the long run.

the best thing to do in my opinion is to develop first one way links and make your site/s a little bit to appear in the SERPs first before you target the linkbaiting through juicy contents. it is really hard to come up with one these days in my opinion unless you have all the time in the world for research in the web.

;) masterofthe

[ChrisWebPub]

Link popularity will always be important to all the major search engines, atleast until a new technology comes out. I cannot see the future so I cannot say for certain what such a technology would be (or I could invent it!) but without something new we'll need to rely on link popularity.

Devaluing it would be a step backwards towards the day of real search engine spam (1999).

Instead it'll merely get more and more complicated. Instead of being merely about weight there will be more and more filters and tweaks to try to improve it and weed out spam. The base concept of the link will stay the same, but what that raw data goes through will continually increase in complexity.