Forum Moderators: Robert Charlton & goodroi
a) Go to Google and search on [statename Realtor]
b) Go to your local real estate agent's website (check your fridge for a magnet) and look in their reciprocal link directory.
Really, give it a try. Here's what I see:
Google: a bunch of big national real estate websites, some directories of Realtors, affiliate sites, MFA sites, scraper sites, and an occasional local Realtor site. Some decent links but also a ton of crap.
On my local real estate agent's website: there is a directory of states and in each state there are links to bunch of links to Realtor's websites. Each Realtor has a description of their location and specialty. That's exactly what I wanted, and it looks like a human has made sure these websites are for real. Nice.
Repeat this this exercise with dentists, bed and breakfasts, chiropractors, restaurants, auto repair, etc. I see similar results.
So, I ask Matt (and Google): how are reciprocal links not relevant? For many search terms they seem far more relevant than Google's own search results.
Finally I offer a little tip for those at Google and the other search engines: finding and using these **human edited** reciprocal link directories as part of the ranking algorithm for search terms like these will actually help improve the relevance of your search results. May the most relevant search engine win.
[edited by: sugarrae at 3:04 pm (utc) on June 24, 2006]
[edit reason] keep it civil please [/edit]
So, I ask Matt: how are reciprocal links not relevant?
Where did he say that? I'm not saying the statment wasn't made I'd just like to know where and in what context.
Also see where this was discussed before: Matt Cutts on Reciprocal Links
[webmasterworld.com...]
If you want better results, why not pick more useful search terms? the trick is to decide what you really want to know, and then pick words or phrases that best summarise that need.
I doubt many people would need the results you sought - and even those who do want a list of realtors by state, could try alternate searches to get a better result.
Good Luck with your searching - find a search engine that thinks similarly to the way you do, and you'll get the results you want; I'd try the big three first (G, Y! M$N) before branching out to Ask! and then on ...
Also, I don't think it's entirely fair to blame Google for the sheer numbers of spammy realtor websites :)
Welcome to the World Wide Web! Isn't It Grand!
Sorry, I cannot see how you got from a search result to reciprocal linking, sorry.
Reciprocal link directories and search engines are just different ways to find websites. Reciprocal links are like referrals in the real world, while search engines are like the phone book. Both are ways to find a product or service.
Here's another example: Your kid needs braces. In the real world you can a) ask your dentist for the name of a good orthodontist or b) look one up in the phone book. The online equivalent is a) find an orthodontist in your dentist's reciprocal link directory, or b) search for [mytown orthodontist] on Google.
Here's another example: Your kid needs braces. In the real world you can a) ask your dentist for the name of a good orthodontist or b) look one up in the phone book
The one flaw in this is that you want a GOOD ortho, not just someone the dentist you are asking has some sort of "referral arrangement". Recips. are quid pro quo, and that's the difference.
One way links make sense, but quid pro quo links are just that - I'll advertise you if you advertise me. No quality necessary.
If you knew your dentist was getting something to refer you, would that change your confidence level about quality?
The one flaw in this is that you want a GOOD ortho, not just someone the dentist you are asking has some sort of "referral arrangement". Recips. are quid pro quo, and that's the difference.
I agree the quid pro quo biases the referral, but I'd argue online you still get a better end result from a reciprocal directory than from a search engine. The reciprocal listings are human edited and won't contain the MFA sites, scraper sites and other junk we see in so many search results.
If you knew your dentist was getting something to refer you, would that change your confidence level about quality?
I've been around long enough to realize that the orthodontist likely sends referrals back to the dentist too. In fact the reciprocal nature of the referral could be based more on golf than dental skill.
The first assumption: Reciprocal links aren't relevant. Reciprocal links can certainly be relevant. The problem is that Google cannot easily judge the intent of links. Google created a climate in which links were eagerly sought after and because of that climate, Google had to reassess how to weight links. Reciprocal links traded between say, a real estate site and a cruise site can be dismissed as irrelevant. But what about reciprocal links between an automobile site and say, an auto insurance site? Those links shouldn't be dismissed simply because they are reciprocal.
Next assumption: Reciprocal links are 'quid pro quo'
If you knew your dentist was getting something to refer you, would that change your confidence level about quality?
Wow, look at that leaping logic batman! Someone asking their dentist can't be sure that the dentist is or is not 'getting something' in return for referring busines other than reciprocal referrals. Nor can Google ascertain that reciprocal link partners are linking to each other for any reason other than trading referrals, unless there are some on-the-page factors that indicate a different type of relationship. Words like, 'advertise here', 'trade links', 'increase your PageRank', etc.
Before Google had such an impact on the link economy, links were traded a bit more freely, and were still often, very relevant. That's when natural link patterns could be analyzed to some benefit to determine 'clusters' of related material.
Current link pattern analysis illuminates a sharp contrast to link patterns of years past. Much more rapid link acquisition, from unrelated sites, or 'link sludge' is much more common now.
The issue becomes how well Google's algo can analyze link relationships to determine link relevance. Simply dismissing reciprocal links is not an optimal course of action as it is well-known that sites sharing a particular interest/topic develop shared link patterns.
Take this one step further, to say, a peer-reviewed paper, the basis for Google's citation algorithm. A Ph.D publishes a paper. Another Ph.D reads the papers and has some criticism, publishes his thoughts and links to the first paper. The first Ph.D reads the criticism and replies, adding a link the second Ph.D's rebuttal. Reciprocal links are generated that are very relevant. Now other's join the fray, linking to both papers and deconstructing arguments on both sides. Links are reciprocated freely. Very relevant links that should count. And all is well.
Until the first Ph.D realizes he has PR8 page and starts selling links and clutters his page with ads for Viagra and cruises...
The problem is that innocent sites will get caught in the process, and the challenge is to minimize the collateral damage.
That's just it, 'links pages' or 'resource pages' are easy to spot. They create very recognizable patterns. Which is why I assume that Google doesn't need to just dismiss all reciprocal links as irrelevant or worthless.
It is also why I caution people not to dismiss link partnerships as pointless. If the links are relevant and make sense for your end-users, by all means, link.
*As these indexing changes have rolled out, we’ve improving how we handle reciprocal link exchanges and link buying/selling.*
*If you were getting crawled more before and you’re trading a bunch of reciprocal links, don’t be surprised if the new crawler has different crawl priorities and doesn’t crawl as much*.
Matt didn't say that reciprocal links are always irrelevant, and he was talking about indexing not ranking.
Maybe you are referring to a different statement he made?
Another quote from Matt's enormous post where he references reciprocal links:
[mattcutts.com...]
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/selling
Given that reciprocal links existed prior to Google and when done appropriately they add value to the web ecosystem, I think Google should provide guidelines as to what it believes is appropriate reciprocal linking. It's not right to strike fear into webmasters with the statements Matt made without providing more guidance. What's appropriate? What's excessive?
I'm also suggesting that "good" reciprocal links provide valuable information a search engine could use to improve relevance. If they can identify the bad and excessive reciprocal links, as Matt implies, they are a step away from identifying the good reciprocal links. Just mix in a little TrustRank and voila free human editors!
They have. Don't create phantoms to fight your strawmen.
They have. Don't create phantoms to fight your strawmen.
I don't believe Google has provided guidelines, please reference.
And, no I don't believe "Don't participate in link schemes designed to increase your site's ranking or PageRank" is adequate given Matt's recent calling out of "excessive reciprocal linking."
No phantoms or strawmen here. This is a legitimate concern affecting many webmasters. Matt has scared them away from participating in a legitimate practice, reciprocal linking, that adds value to the web and existed prior to Google.
Reciprocal links traded between say, a real estate site and a cruise site can be dismissed as irrelevant.
Even a statement like this may be incorrect. Imagine a real estate site selling time shares on a Carribean Island and promoting the local "Tall Ship" cruise line that allows owners to visit other islands.
Now, a viagra site linked from a music review site might well be irrelevant.
My point though is that sites are so varied that seemingly irrelevant links might in fact be very relevant in a specific context.
Rare Bird <-> Fishing (fly tying supplies)
Covered Bridges <-> Dude Ranch (happens to fall onthe same route)
etc.
WBF
Okay, but why is the cruise line linking back to the real estate site? Your point is still taken though. Determining irrelevant recips isn't as easy as a simply looking for the absence of a semantic relationship.
if a page or site is not good enough to be indexed frequently, I doubt it is good enough to rank high.
Note the use of "excessive", using recips for ranking purposes is excessive, and as DG points out, while reciprocal citations are not uncommon in Academia, they're invariably topical to the subject matter...
if a page or site is not good enough to be indexed frequently, I doubt it is good enough to rank high.
Pages that contain "Evergreen" content that seldom if ever changes after they are first posted can easily deserve to be highly ranked and still not need to be crawled frequently.
Saying they could have guidelines that detail every possible thing that can be detailed is different than saying they have no guidelines. The guideline above is crystal clear and easy to understand. What more could you possibly need? Link to stuff that adds value to your site. Get links from places that have users who might be interested in your site. Don't particpate in schemes.
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/selling.
Did you notice the word "excessive" in front of "reciprocal links"? It is really important to consider that word.
I have a review site that links to almost every manufacturer in the industry. A huge number of them also link to the site, therefore they are reciprocal in a way. But the links are not all on one big links page on either side. There are also thousands of one way links from other sources.
He is talking about ALL the factors that they look at when deciding how much to trust the links coming into a site.
If all the links coming into your site are reciprocal AND they come from links pages AND all the sites that link to you seem to have earned their "low trust" in a similar fashion, then you are in trouble.
On the other hand, I have all those reciprocal links from all those trusted manufacturers. They don't have huge link exchange programs, so their few reciprocal links are likely to be seen as more legitimate.
It also really makes a difference if you are able to get links other than reciprocal links. You know, those good old natural one way links. If you have a lot of reciprocals, but you also have a lot of trusted one way links, your reciprocals are less likely to be discounted.
Again, you have to consider the entire context of what he is saying, and don't just concentrate on two words.
As far as my other comment: Maybe Google visits pages, even if they don't change often, stricly based on their pagerank. Matt's comment seems ti indicate so, and I know that a site of mine was indexed daily--despite no updates for years.
Until the first Ph.D realizes he has PR8 page and starts selling links and clutters his page with ads for Viagra and cruises...
Ain’t it the truth!
The problem is that innocent sites will get caught in the process, and the challenge is to minimize the collateral damage.
And therein lies the rub! In which lap does the challenge fall? Most here would no doubt claim it is Google's challenge to get it right from the get go. But how can Google do that when so many webmasters are intent on gaming the system?
Google is just a series of machines and a group of algorithms designed to produce the best and most relevant search results possible
“Google” does not have the ability to properly evaluate all the criteria available and likely never will. These machines and algorythms (as sophisticated as they may be) are inherently inferior to the human mind. They are not able to correctly identify an "innocent" from a "not so innocent" website and likely never will. But ... they are doing an "acceptable" job for the most part!
SEO types spend their time dissecting, analyzing and identifying the weaknesses of any given search engine’s algo. Their ultimate goal is to find ways to take advantage of these weaknesses for immediate and long term gain.
Those who analyze every little thing Google does may want to remain cautious when exchanging links these days. Google is not sophisticated enough to correctly identify legitimate reciprocal links from those which are swapped purely for “ranking” purposes. As a result, they have chosen to take a hard line towards "excessive" reciprocal links. Whether or not that decision is right or wrong is moot. The point is, that this is what they have done ... at least for the time being and we as webmasters need to understand that and all it implies.
If the links are relevant and make sense for your end-users, by all means, link.
I couldn’t agree more! I am of the school which says, "I will link to whomever I see fit" and let Google engineers sort out the algo problems. Belonging to this school of thought means I must accept the way Google chooses to interpret my outgoing links. Sometimes, it isn't always fair ... but it is what it is. If enough webmasters were to concentrate on doing their job and let Google and other search engines do their job, we likely wouldn't have so many "tales of woe" in the Google news forum.
The truth is that greedy gamers will always try to manipulate any algorithm Google and other search engines develop. That puts honest business people and website owners at a distinct disadvantage when trying to do business on the internet. Unless you/we are willing to pay for PPC programmes, we just have to get used to and accept the roller coaster ride the SE’s provide. Free traffic is not a “right” any of us can count on to provide our incomes. However, as consumers, we all choose to use search engines which produce the best and most relevant results. Loyalty is a fleeting quality.
In my opinion, Google is becoming far more sophisticated than other SE’s and I believe they are doing a pretty fair job of delivering relevant results. This is putting pressure on MSN, Yahoo and others to do the same if they want to remain competitive. I believe MSN is improving substantially while Yahoo is falling well behind the pack.
Given that reciprocal links existed prior to Google and when done appropriately they add value to the web ecosystem, I think Google should provide guidelines as to what it believes is appropriate reciprocal linking. It's not right to strike fear into webmasters with the statements Matt made without providing more guidance. What's appropriate? What's excessive?
What’s appropriate and what’s excessive are limits which only you, as the owner or manager of any given site can tell. You know which links are there for good reason and which benefit your readers ... and you know which are there purely as search engine fodder! But, if you honestly claim not to know the difference, shame on you!
The cure for a lack of knowledge requires a concerted effort!
I couldn’t agree more! I am of the school which says, "I will link to whomever I see fit" and let Google engineers sort out the algo problems. Belonging to this school of thought means I must accept the way Google chooses to interpret my outgoing links. Sometimes, it isn't always fair ... but it is what it is.Absolutely. Recently, several merchants from CJ have sent me emails saying they are switching their link code because Google is penalizing pages1 for too many affiliate links. The fact is, if I am not proud to feature a merchant on my site, they won't be there. Don't change your sites for Google or you will always be changing your site for Google. Additionally, it would not be out of the realm of possibility that Google will be able to easily identify those that make changes like that. After all, it is an obvious attempt to fool the search engine.
1 - I'm not saying Google is penalizing for affiliate code, only reporting what others have said and may be true..
"2. Google rewards for 'healthy' traffic patterns"
WOW... thats very interesting!
Would you be kind to elaborate more on that?
Thanks.
But keeping in mind, the more outbound links you have just for SEO, the more likely you are to have some that go to 'bad neighborhoods' or banned sites.
Did you all see the example Matt Cutts put on his blog? There was a site with a footer full of irrelevant links. Anyone who spends much time on the web runs into sites like this constantly.
There are so many sites that attempt to gain higher placement in Google by purchasing links or exchanging links. An an entire industry for buying and selling links has developed.
Google is trying to improve the results for its users. Anything it can do to bring down the paid links industry is good for the web.
Link to stuff that adds value to your site. Get links from places that have users who might be interested in your site. Don't particpate in schemes.
There are at leasdt two fundamental flaws in PageRank.
When PageRank (one page one vote) was devised it wasn't there and had no value so the playing field was flat when it was introduced and it was a valuable almost eutopian addition to a ranking algoritm.
Once it was introduced it immeditaely started a process which has lead to massive distortions.
2 Fundamental flaws.
1. Votes can be bought both by payment of money and in kind. The motivation for many reciprocal links is not a true vote for the quality of the page linked to.
2. Focus on pages and not sites. For example there is a page in the top 4 of SERPS in my niche which has a PageRank one step higher than the others that surround it. The reason for this appears to be the purchase of sitewide text links with the term targetted as anchor text on every page of the archive site of a local newspaper. The local newspaper pages are pointed to by pages that want to refer to the news story but the pages that are pointed to by the archive pages are not relevant to the story and are only there because of payment. If PageRank only counted one link from a site then this would cut out this kind of crap but is not perfect in itself either because sometimes a site may have many natural links to a page on another site.
Unless Google can find some way to identify un-natural links PageRank has virtually no value in their algorithm because at the moment we can all buy it. PageRank is no longer a secret ballot by ethical webmasters offering links for the benefit of their visitors, it is a commodity that can be bought and in many cases for less money than bidding for #4 spot on Adwords.
Sid
