Welcome to WebmasterWorld Guest from 126.96.36.199
Forum Moderators: martinibuster
It's been a matter of debate whether or not Google discounted reciprocal links. Today it is safe to say that Google is not fond of reciprocal link exchange schemes. It doesn't get more explicit than this.
Within a recent Google Blog post [googlewebmastercentral.blogspot.com] about bad linking practices, the author lumps reciprocal link exchange tactics alongside paid links. The author does not qualify the statement against recips by singling out non-relevant or aggressive exchanges. Neither does it make a distinction between relevant link exchange or free for all link exchange. The blog simply references the exchange of links.
To sum up, even though improved algorithms have promoted a transition away from paid or exchanged links towards earned organic links, there still seems to be some confusion within the market about what the most effective link strategy is.
It's fairly clear that Google's algorithms are targeting reciprocal link exchanges.
Has the day finally arrived when we can toss off Reciprocal Link Exchange tactics the way we ditched Netscape 4.7 and the "Web Safe" color palette?
Today there is no ambiguity: Reciprocal link strategies are viewed by Google as link manipulation on the same order of buying links. There is no ambiguity about how Google views reciprocal link strategies
...reciprocal links have been around forever, and Google
doesn't frown on engaging in reciprocal linking in moderation.
The key here is, indeed, moderation :). If, say, 90% of your backlinks are reciprocal, that's probably not going to improve how our algorithms view your site. Or worse, if 90% of your backlinks are reciprocal and not likely to be of interest to your user.
But exchanging links here and there -- *especially* when
done with clear editorial judgement (e.g., you're not just
accepting dozens of link exchanges willy-nilly) -- that's
not the sort of thing Google looks down upon.
Hope that helps clear things up a bit!
[edited by: Marcia at 10:13 pm (utc) on Dec. 17, 2006]
Another useful test is to ask, "Does this help my users? Would I do this if search engines didn't exist?"
Intelligent, on-topic linking such as DigitalGhost <added: and Sugarrae> described will always have value with or without the search engines.
If you've been giving and getting links because they'll make good sense for users, I don't see much need to run around taking down links for fear of displeasing Google. The algos won't always get it right but the mutual linking DG described is exactly the sort of citation they should be trying to treat positively.
On the other hand, the sort of whoring that trades links with absolutely anyone who will reciprocate is a different matter. If that has been your main approach to link development, your web presence will likely be sending different signals to the spiders than a site that has used more editorial discrimination. In that case, some trimming might be in order. Start with those who have taken down their links to you!
If you remove a misfit link that you posted because of a trade, it would be decent to let the other site know you're backing out of the deal. You don't need a big explanation, just say that your editorial policies have changed and you won't be linking to them anymore. A couple of sentences would suffice.
Why wouldn't the number and percentage of unreciprocated outbound links from a site also be factored into the equation for figuring the balance?
again we come back to definition of reciprocal linking. I run quite a big site and have many pages with many sections in it. Each seaction gets a cool links page with links to nice sites covering the area of the section. i do not contact these sites' owners, just put links to them.
what if they link to me because they find my site a cool place on the Web and i know nothing about it?
will we both get banned?
IMO, google are smart enough to figure out what a regular links page looks like.
they definitely can figure out what links point at the site and what links leave the site.
but they cannot ban sites for linking as described above, YET. and they know it.
so what they say now is they now know how to 'safely' not count links located on long links pages when calculating ranking of page in their search results.
You plug a bunch of variables into a classifier. Humans classify the web pages on a scale for various criteria. The classifier learns patterns of inputs that correlate with ratings, and continuously fits these into ongoing ratings.
I'm sure this frustrates those trying to devine the formula. But I'm sure that frustration pleases Google.
That said, I don't know the extent to which Google uses classifiers. Nor am I especially an advocate of them. I worked for a company that is a leader in this field a few years ago. I was not impressed with accuracy. It's scary to think that this type of technology is used to produce credit ratings, yet (at least at the time I was working with it) wasn't good enough to be practical to help people with online form-filling. It missed often enough that users had to check the forms very carefully. Saved some typing time, but sure didn't make for a good user experience and didn't instill confidence in the tool.
It's just that these three way link exchanges involve you linking directly to a commercial site and them sending you a link from a crappy links directory - I've discounted those from their inception. It was obvious that this wasn't going to be a strategy that was going to gain you anything in the long term. (and remember that in SEO you have to be looking at the long term - if you aren't then you are a spammer in the eyes of the search engines - explanations of that logic are available elsewhere.)
I wouldn't look at a policy of rejecting all link exchanges on the basis of the Google blog post, because some link exchages are relevant, are helpful to users and can benefit your site in the long term (Adam Lasnik's post bears this out)
Effectively people need to be implementing a policy of looking at the value of each link exchange request and acting accordingly - effectively have an editorial policy of your link exchanges the same as you would on any other links on your site.
From my own personal point of view the question of link baiting is much more of a concern, as there are also different qualities of link bait in the same way that there are different qualities of reciprocal link exchanges.
At this point in time I don't see Google taking steps to identify and weed out low quality tabloidesque link baiting strategies in the same way as it is weeding out low quality directories and reciprocal link exchanges. (and before you point it out - yes I know that this is a different level of problem)
However to give you an example - take cold fusion (not the prograqmming language by the way). If a scientist claims to have achieved cold fusion - an article to this effect will probably achieve significant numbers of links, mainly from people who aren't experts in the field, because everyone hopes that cold fusion will be achieved. On the other hand a well thought out highly technical article on cold fusion is likely one to achieve links from other people in the same line of research, who number significantly less than those who hope that cold fusion can be achieved.
Which of those articles has the most value?
IF true the first obviously is the most important (this is a very very big IF as we have seen these claims numerous times in the past so the chances of this being true have to be considered as very low.) On the other hand in the overall scheme of things the second type of article is probably of much greater long term value value but will have garnered much fewer links.
Therefore is we look at the way Google is currently considering link bait as the great answer to your link building strategy, all you need to do is keep writing sensationalist twaddle. As if your first article loses all it links after a short while, the next sensationalist story will, of course, keep any quality articles nicely buried.
[edited by: IanTurner at 12:53 am (utc) on Dec. 18, 2006]
1. writing dozens of articles and submitting them to dozens of article factories.
2. Buying hundreds or thousands or tens of thousands of advertising links.
3. By engaging in massive link exchanges.
4. By operating networks of similar sites that interlink.
None of these methods are exempletive of google's ideal of links (freely given by choice). Yet these are the methods of the day and they are effective, i.e. rewarded by google.
My own belief is that google has VERY VERY VERY limited capability when it comes to discerning "ideal links" versus links obtained in one of these ways.
Should you be able to get a site to the top using such methods? Ideally, no. Nor should you be penalized for doing so. But these statements from google employees are a little ridiculous. Do they really think anyone believes this stuff?
Google, sadly enough, is incapable of discerning anything but the most outrageous and clumsy link scheme.
Discounting them, okay... they can discount them, it won't matter much, such links are about visitors, PEOPLE. I won't link to someone who i don't think is of interest to the people on my site. But penalizing for them would be insane, i think this is a point everyone in here has agreed on. If the official stance would be "reciprocals are evil", they'd be stone cold PARANOID by now ( might be ), and think that most reciprocals are done by SEOs, spammers and evil viral marketers aiming for PR and relevancy they don't deserve. ...Yeah faking tons of publications at communities people don't belong to is much better.
The only things i see right now from this message is...
Trying to create a network of sites related by relevance might get harder.
Older sites would lose the last reason to link to new ones... even if they're not competition. Why would they bother? Do YOU bother?
Linking back would not be seen as a valid exchange.
Yet one way links would be even more valuable.
What do you get in the end?
On one side, you'll get the flourishing of web communities/blogs that are visited with or without G as the medium, and which will pick up the new information. These sites will be of real value to any webmaster. Owners will or will not like the traffic and the ever-updated news and content. Users and moderators most likely won't.
This is how DMOZ became so overly introverted. This is why MSN has blogs all over it. This is why the other thread discusses whether Wiki passes PR. This is... basically one step short of spam.
On the other side, there will be MUCH more one-way links appearing on-topic out of nowhere... compensated in some other way than a link back. And through a channel G wouldn't be able to track anymore.
How does that sound. From the bought links are evil point of view. From the crack down the spam point of view.
If interpreted in this way, this statement downright pushes people to buy links.
Not necessarilly for money, perhaps for a steaming plate of pasta, some rare to get phone card, a prom date, whatever. Perhaps not mentioned on the sites like "pay to be included" just a little classified in the local niche magazine. :P
It also calls out for you to create yet another blog on blogger to spam your site into the index. Gateway pages, subdomain spamming, free hosted pages would be swarming with "one-way links" until the patterns are figured out. But then new patterns would arise. Sounds familiar to me.
Reciprocals ... well except for FFA... are a great quality filter.
Webmasters look harder at who they link to or who they're linked from that Google will EVER be capable of!
Unless this is an over-generalisation of a single possible way of linking... from the most... pessimistic, most paranoid point of view... i don't think i know where to put this. And while it may scare people who need to be scared too... if the rethorics are flawed, they'll be seen as big brother trying to dictate what the web needs to be. Running around in cirlces.
Or more like in a downward spiral :)
Mind you i don't think this was an official statement on reciprocal links being ignored or penalized... for the reasons above that i'm sure G is aware of all the same. I think they meant to say...
"Don't build yet another directory just to promote yourself, please. You're bribing people to link to you to increase your PR in a way that's not good for them OR the visitors. Your directory page has 3 unique hits a day while your home page - that the poor people link to - has AdSense all over it bringing you a fortune. They don't get traffic from you. They don't get nothing from you, yet if they don't link to you from a PR 3 page you won't let them in? You're gonna get it!" ... which is none the less interesting in a democracy but i can at least relate to the idea :P
Make that i endorse this idea. I like it!
Go G Go. :D
sorry. again. long post
I get several reciprocal requests to trade links per day--most of which are totally irrelevant to my content so I ignore about 99% of them.
When I write a new article I go searching for items I may not have covered in my article and link to sites with quality content my readers will be interested in, never requesting the owner link back. If we all did that we wouldn't have to worry about what Google thinks of our links.
They don't know what is legit and what is not.
We may not give credit where it is due. I'd have to ask all those who have been part of the fallout when Google decides what is legit and what is not. Unfortunately there is collateral damage in the process.
Ever use Google Earth? Imagine that same interface in a Google Engineers office but with a map of all links on the web. Apply Google's algo and what do you think they see? I would think it is not that difficult to detect the bulk of what is legit and what is not. ;)
As a matter of fact, it's all those RECIPROCAL LINKS that allowed Google to index the web in the first place and figure out what the network neighborhoods were that originally existed.
Sorry, but I won't give up the power of link exchanging just because one large company suddenly doesn't like it because some people are gaming the system, too damn bad, as this has been the way the web has worked since the beginning.
[edited by: incrediBILL at 4:11 am (utc) on Dec. 18, 2006]
This is a lot of speculation about reciprocal linking in response to an official blog entry, when there's not even one mention of "reciprocal" on the entire page ;-). Take a step back, look at the bigger picture, take a deep breath!
people not just bots click on links.
Yes! If it's good for your users, link to it; if, by chance, the link is not given the *full* weight of a "vote," by Google or MSN or Yahoo or Ask or whatever... that shouldn't be a huge deal. These things tend to work themselves out in the aggregate.
i bet we can still fool them by doing recips on a bit different basis.
No. Hint: Anything attempted to "fool" or "deceive" is not a smart long term strategy in search stuff. Current blackhats may disagree with me. The long list of former blackhats, however, may be ruefully nodding their heads (er, to clarify, that was a collective reference: one head *per* blackhat; Zaphod Beeblebrox, to my knowledge, was not even into SEO. But I digress.)
well for a start, everyone can change links.htm resources.htm file names that would claw something back.
Seems more a swipe at links schemes in general
Yes yes yes! THAT was the intent of the entry.
Discounting reciprocal links isn't the same as penalizing such links
If a Webmaster is engaging in reciprocal linking in a way that clearly indicates to us that he or she is doing so to garner PageRank, not out of a genuine interest for that other site... well, that's the sort of linking scheme we don't see as very user-friendly. Are we apt to ban that Webmaster's site? I highly doubt it. Are we likely to value those links less? Quite possibly.
It certainly is not official. I'm also surprised that anyone is taking any content published on Googlee's blog as gospel.
What part of "Official" in the title didn't resonate with you? :). The people who write on our Webmaster blog are either engineers or product managers -- or those who work directly with them -- in Search Quality and Webmaster Tools. While there may be nuances here and there, I do think it'd be wise to view stuff posted on that blog as accurate and important to Webmasters.
all reciprocal links aren't the same.
Not all reciprocal links are equal.
Right and right. Patterns are what matter.
People are going to read your post and think that Google is going to penalize their site because they have a few reciprocal links.
I hope not. That isn't reality. Our aim isn't to penalize sites, it's to deftly determine when and to what extent a link is indeed a "vote" for a site.
Link bait is quite visible nowadays... it's the trend de jour. That doesn't mean that it's accounting for even 1% of all link popularity, nor is it -- by a long shot -- the only way to get links. With that said, I agree that perhaps the choice of words ("bait") at the end of the blog entry might have been unfortunate.
Anyway, I hope this has put some fears to rest. I link to friends who link to me; we like each others' sites, we think that folks who visit our sites might like them, too. And that's fine! And also, as Sugarrae pointed out, it's only natural that someone may want to link to an article that links to them. Reciprocal linking happens, and it's very often done in a natural, innocent way.
Over time and with lots and lots of data (and very handy tools for crunching it :-), it becomes more clear to us at Google what is "natural" (or organic) on the Web and what is not. We aim to reward the former, discount the latter. Take that as a broader SEO strategy statement if you will... it's not just about links, and it's DEFINITELY not all about reciprocal linking.
I remember the pre Adwords days, only a few years ago. You had a "normal" smart indexing powerhouse named Google. No lost sites, no messy incoming link reports or other wired hourly glitches and a web full of opportunities. And what have we today? a freaking greedy, utterly confused, arrogant and hypocritical web policing entity with a clear intention of swallowing everything in it's path and penalizing to death whatever is left.
Yea… go penalize link exchanges Google. A really great move.
So, explain to me - you start a brand new website about "something"; what do you have to do to get "organic" links? Do you have to be "connected" so people link to your website?
Or you start creating content and then what? Who would know about this content - nobody. If you dont get LINKS even Google wont visit.
You have 2 options:
a.) pay for a link
b.) exchange links
That's it. No other choices, period. And Google considers a reciprocal link as page A linking to page B and page B linking to page A. It doesnt matter where you place the links, they will still be reciprocals.
Are one way links better - yes, but recips are also good. Maybe 1 organic link equals 3-4 recips, but it has been like that for a long time now, so nothing new.
If you have a website with 500 recips - of course, you are telling the bot and your visitors that there are at least 500 websites, that you know of, that are at least as good as yours - that's common sence.
Nothing has changed, and whoever made the post on Google blog was, yes offence, an idiot.
And on another note - I see nothing wrong with adult websites - I dont have one, but I can imagine that the competition is big. I dont think that at a webmaster forum we should have lower-class websites based on industry.
I myself have been known to indulge in adult entertainment every now and then, and if you havent - go back to your game of Dungeons and Dragons and let the grownups talk ;)
Check out our link dev library. There are plenty more options. I outlined some of them at PubCon, too.
Good post Adam and I appreciate your coming by on a Sunday evening to chip in on the discussion. :)
[edited by: martinibuster at 7:28 am (utc) on Dec. 18, 2006]
No, there are not.
a.) buying the link (in any form - press release, directory, bribe)
b.) exchange links - A->B->A
c.) bait/spam - digg, free news release, blog post, three way, doggy style etc.
I didnt find anything in the library that does not fall into any of the 3 cats above.