Reciprocal links

Did I miss a meeting?

[Advenlo]

I received an e-mail in response to a pretty standard link request which explained "As you may be aware, Google has recently changed their link rules. With the new Google rules, reciprocal links cancel each other out and then neither site benefits."

Is this true, have I not been paying attention of late? They are now trying to promote a link ring instead.

Thanks for stearing me in the right direction

[Advenlo]

As it took a couple of days for approval, feel entitled to one bounce up.

[ken_b]

First, welcome to WW. This place is a goldmine of good information.

Secondly, the reality is that probably no one REALLY knows Googles "rules" on this stuff. Many people have opinions. Unless they can back those up with stats or a direct quote from Google, they are just that, opinions.

That said, I haven't seen a lot of folks here saying that recipricals cancel each other out on a general 1:1 basis.

That doesn't mean they all count equally. A hundred links from the same site may not be equal to a hundred links, one each from a hundred different sites.

I'd ignore that email and move on with your link program.

[added] opps, I see you've been here a while. I got the message number and posts number mixed up. I guess I need to pay more attention.[/added]

[rfgdxm1]

I very seriously doubt Google would ever totally ignore reciprocal links. With amateur sites on a narrowly foucused topic, it is very common to find all the notable sites on that topic linking to each other. This happens solely for the benefit of the site users, and not to manipulate Google. In fact, I'd consider it suspicious if there were no reciprocal links for an amateur site. This would suggest to me the site in question is so lousy no sites on the same topic consider it worth linking to.

Having said that, it is possible that Google gives bonus points when a link is non-reciprocal.

[suggy]

I have PR5 sites that are page 1 for highly competitive search terms and have maybe tens (rather than hundreds) of exclusively reciprocal links. Whoever wrote that email, definitely had the wrong end of the stick.

It's funny how SEO's opinions differ. I went to a "web gurus" page the other day and he was explaining how I needed tens of thousands of links to get to the top!

Cheers,

Sug

[Advenlo]

Thanks guys (and gals?) glad to know my recent hard work isn't in vain.

ken_b - whilst I have posted a couple of times I'm still very much a new user and never did get welcomed on my first posting. so thanks for the thought!

[Patrick Taylor]

Purely on an arithmetic calculation of PR, when 2 pages of identical PR exchange links, I think they both lose out because of the dampening factor - or at least this used to be the case.

[webnewton]

Patrick has struck the point. However this doesn't mean that you should stop reciprocal link exchange. Besides this you should try to post your link to as many directories and other resources which would welcome your link without any reciprocation. A good way of doing is to buy a link at some site with good PR.

[suggy]

Whilst not doubting the logic of Patrick's argument - I don't really get the maths - commonsense and experience tell me two links CANNOT cancel each other out. I see too many sites like mine gaining PR5s and 6s purely on reciprocals, without cheating (ie Javascript or redirects to prevent passing of PR). How could this be?

[Patrick Taylor]

I'm not an expert on this, but it used to be the case that a page passes on a PR value of only 0.85% of its own PR, divided equally to all the pages it links to. So it follows that two reciprocal linkers not linking to anything else are wasting a total 0.30% PR between them (or something like that, I think). Someone who knows better might elaborate on the arithmetic.

[mfishy]

How can two links cancel each other out when one says "widgets" and the other "corn dogs"?

The benefit of the exchange is the desired anchor text/link pop which is the cornerstone of the Yahoo and Google scoring systems.

PR is more of a side effect and I can't think of a more time consuming way to raise PR than link exchange.

Recips are certainly becoming less valuable for a variety of reasons. A biggie is Yahoo seems to have a distaste for link directories.

I wouldn't pay much mind to what the latest rumors are with the link swapping crowd. The funniest myths become facts somehow - my favorite is that links must be from topically similar sites - that usually makes me fall off my chair laughing :)

[Patrick Taylor]

Well don't hurt yourself! By the way, I was only commenting on the dampening factor, not the pros and cons of linking.

[Hagstrom]

I'm not an expert on this, but it used to be the case that a page passes on a PR value of only 0.85% of its own PR, divided equally to all the pages it links to. So it follows that two reciprocal linkers not linking to anything else are wasting a total 0.30% PR between them (or something like that, I think).

It's true that PageRank will be damped - and that the damping factor is assumed to be 85% - but this holds true whether you chose to conserve your PageRank (linking internally) or passing it (linking externally).

If you study the formula you'll see that the average PageRank^* in the Googleverse is exactly one^** - this is a basic property of the formula, and you cannot "break" this rule and start increasing PageRank by any clever linking scheme.

--------
^* "Real PageRank" - as opposed to toolbar PR
^** In the real Google-word the average PageRank will be lower because of pages without links and because of banned sites

[dirty_marra]

Its just a thought but:

If google links to every website in the world (or near enough) and a fair few of them link back then technically these are reciprocal links. and if reciprocal links are penalised then how come google has such great PR?

Like i say....just a thought ;)

Marra.

[Patrick Taylor]

Thanks Hagstrom. And now I see I mixed decimals with percentages... never was much of a mathematician. One can, of course, increase the overall available PR of one's site by adding pages.

[edited by: Patrick_Taylor at 2:29 pm (utc) on June 30, 2004]

[Hagstrom]

If google links to every website in the world (or near enough)..

Google doesn't link to sites - it just presents them in the SERPs.

... and a fair few of them link back then technically these are reciprocal links. and if reciprocal links are penalised then how come google has such great PR?

Who said Google penalises reciprocal links?

Like i say....just a thought

Keep 'em coming :)

[dirty_marra]

Google doesn't link to sites - it just presents them in the SERPs.

fair point

Who said Google penalises reciprocal links?

thought that was what this thread was about....at least that's what the first post of this thread seemed to be questioning

Keep 'em coming :)

I shall :)

[shillard]

My site used to sit on about 14-16 for my chosen search words on G. That with a heck of a lot recip links. Then I decided to go all out and blitzed over night about 60 non recip obligated links. The next day I hit 1.

I dont know if it was a coincidence, but it certainly says to me that while G is a big fan of links generally, unreciprocated links may be the key.

[shillard]

oh and by the way. I don't believe for one moment that G penalises sites with recip links. It has always been my number one way of getting high ranks and still is. Just do a link search on whichever site is getting the top spot for your chosen search terms and you will soon realise that links, recip or otherwise don't have any negative impact and probably help out a lot.

Anyway, thats all I do, I look at who has the number one pos on my chosen search term. I check out their source code, see if theres anything special, then do a link search and methodically get the same links + at least 5% more and I can virtually guarantee that the next indexing my site will push theirs off top spot.

[Pass the Dutchie]

The core principle of how Google ranks a site is based on who and how many other sites think that your site is worth going to. G does this by counting the amount of quality and related sites that link (recommend) your site. If the logic of the dampening effect is true then G is placing less emphasis on reciprocated links compared to paid links. This makes me believe that G rewards a paid link more than a reciprocated links. However you run the risk of getting filtered for buying or selling links! In priciple all inbound links are a pro, some more then others and some are risky.....take your pick ; )

[webnewton]

oh and by the way. I don't believe for one moment that G penalises sites with recip links. It has always been my number one way of getting high ranks and still is. Just do a link search on whichever site is getting the top spot for your chosen search terms and you will soon realise that links, recip or otherwise don't have any negative impact and probably help out a lot.

Shillard no one here said that google will penalise if your do a reciprocating link exvhange. No. Go to google seo guidelines and you know that they're not a fan of reciprocal link exchange. Google's current algorithm is clearly giving prefrence to the sites which are getting non reciprocating links from other sites. Why do you think sites backed with heavy blog links are making it to the top. Because they're not reciprocating links. Why do you think people are buying links more and more these days. Simple no reciprocation needed. And result is clear cut. They're at top of SERPS.
If they don't do that do you think the official site would ever make it to the top. Hiltop algorithm identifies the expert pages on the basis of impartial one sided links not reciprocating links.

[olderscot]

A little logic should show that reciprocal links aren't a problem.

Take DMOZ as an example. They have a pagerank of 9 yet link out to virtually everybody. The implication? all links to them are reciprocal yet they still get a PR9 so Google must be counting them.

Mike

[JuniorOptimizer]

All the links to DMOZ are certainly not reciprocal.

And wouldn't banning reciprocal links be the most obvious thing for Google to do?

How does the abundance of recip links make a site stand out for quality?

[shillard]

After my last message I had a close look at my links. I reckon about 10% are unreciprocated, but of those only a handfull come from pages that are in the top 10. It is certainly true that on both Y and G if I just type in my URL it tends to only bring up the unrecip links first, and ignores a lot of the recip links.

Most recip links don't insist that they are on the first page on your site, and usually ask you where the link will be. So my question is, is it a workable solution to actually have another site, one that I don't care about the pr, on another host alltogether, which I could use to put my recip links on?

I just realised this is possibly what my main competitor is doing because they seem to bring up hundreds of links, from sites that I know demand reciprocation, but theres nowhere on their site set aside for the reciprocal link. Up until now I was assuming that the competitor was just not reciprocating, and the reciprocees (?) were being tardy in checking. But now I wonder.

This particular competitor is the bane of my existence, they have completely flooded my second favorite search term and I cant seem to find a way to knock them off. I have links from every site they do, and a lot more. The only other thing they have that I don't is a whole stack of single page sites on foreign language free web hosts that link to their main site.

I've been too afraid to mimic this assuming that G wouldn't like it. Plus, and this is even worse for me, I can't get anywhere near the top 10 on that search term because virtually every site listed in the top ten is really the same site, under a different name. They say things like "Blah blah blah Pty Ltd, also known as Blah Blah Blah Pty Ltd." and the end result on alse these pages is a link to their real page.

Why hasn't G picked up on this, theyve been on the top position now for more than a year.

Oh, and by the way, I agree now that unrecip links are better than recip links. I hope I can still be humble in my backflip.

[Pass the Dutchie]

Oh, and by the way, I agree now that unrecip links are better than recip links. I hope I can still be humble in my backflip.

we all live and learn ; )

[Larryhat]

For Google to see if a link is reciprocal or not, it has to spider BOTH sites. For each and every link on site A, it has to trudge thru every page of site B, C, D...

How likely is that? Isn't it easier to assume that MOST of the time they simply don't look? Think of the huge extra load on their already overworked robots. In special cases (link farms etc.) I can see it. Otherwise, one might argue that they probably don't waste the processor cycles and bandwidth. - LH

[webnewton]

For Google to see if a link is reciprocal or not, it has to spider BOTH sites. For each and every link on site A, it has to trudge thru every page of site B, C, D...

I think it istn't too hard for google. Why do you forget that the basics of google SERPS results rests on the link popularity of a site. Every time the bot passes through a sites it devoures all the new content changes and links on the site. It's not hard for it to see how many links are coming from sites outside and if site x is linking to site y is site y linking back or not.

Well anyways the purpose of this post was to identify if the non-reciprocating link matter more than reciprocating links. And we see that answer is YES.
A site with non reciprocating links is identified more popular or say more important by google than a site with reciprocating links.
There is anohter issue which i think should be discussed here.
Quality of links: How much does google consider this.

[elfin spectre]

Easy to test this theory - remove all of the links on your site to other sites where they have reciprocated. Do you shoot up the SERPS.

Then apologise to any sites that have noticed their links have gone

[shillard]

Regarding quality of links. I find that the links that make the biggest impact on my sites prs are those that are not necessarily of the same content to my site, but appear to have a very credible filtering process, even though they don't.

For example, I followed a link from one of my competitors a few months ago which ended up on, of all places, an astronomy site (not astrology) with links to all the official astonomy sites around the world. Since they had a link there, and it came up as the very first on a "link:" search, I figured I would see what happened if I tried. It turned out to be effectively a free for all, and my link appeared automaticaly and the next day was indexed on G.

That one link alone, it was a new site, pushed my site into 14th position on G. I took my link off, given it was only an experiment, but G still shows that link, even though my actual link isnt there.

Very strange. But for some reason G gave a lot of weight to this one link, and now can't seem to forget about it. At the same time, I've had links on hundreds of other automatic free for all's and you wouldnt know they existed.