How important are outbound links?

[got fish?]

I recently started a directory on a domain i wasnt using.
When people add their site they are asked to link back to me but its not required. I want to keep my directory for the people who use it, i don't plan on charging for adding a site and i dont want to force people to link back to me if they dont want to.
So what im wondering is will the amount of outbound links do me favours or would it really be worth insisting on a backlink?
Comppared to backlinks, how effective are outbound links to the site that hosts them?

Also, i wrote a mod that displpays the last referrer in the footer of everypage.
Im hoping that if people have their link displayed in the footer of every page they might be more inclined to link back.

[bentong]

I guess, it would be a good idea to offer first a free listing to get your directory a good start and then you can require a recip link, IMO.

perhaps, you can have only 1 page from your main that is pointing to your outbound links page to minimize pagerank drain..

submitted also on shopping category.

[e-digicam]

If you create a specialist page then add a single outlink to a respected site for that topic - it will help.

cheers
Steve,
<url removed - use your signature for links>
Digital camera news, reviews ect.

[nemanja_nq]

I agree with e-diqicam, it will help. Its important to have at least few outbound links.

[aspen]

Outbound links do not help, its a myth.

[randfish]

Not a myth in my opinion aspen. I believe that search engines use outbound links like any other type of on-page content to help determine things like "trust", "quality" and "authority". It may not provide a large boost, but by linking to high quality sites that are relevant to your page topics, you're showing search engines (and visitors) that you take the quality of your own pages seriously and are linking with care and intent (rather than for advertising or link exchange purposes).

If I were starting a directory, my goal would be to find the BEST quality sites in each category - 2-4 each and list them myself, manually with great titles and descriptions. What's also nice about this from a directory perspective is when sites that want to be ranked for those terms look up the links to the sites that do rank they find your directory - and can get a link from you.

[aspen]

If you believe that outgoing links help, prove it.

you're showing search engines (and visitors) that you take the quality of your own pages seriously and are linking with care and intent (rather than for advertising or link exchange purposes).

Thats like saying by using meta tags you're showing that your pages are of high quality. Outgoing links are meaningless in this regard. Anyone can add them, regardless of the quality of their site, and they do not automatically add to the quality of their site.

No one has ever been able to prove that outgoing links help your site. This myth is a result of conjecture, opinion, and the illogical karmic belief that some webmasters have whereby the Internet is some utopic society where what goes around comes around.

If you're going to link out, do it to enhance the usability of your site (ie do it where its an appropriate link that leads to more content that you want to offer your visitors). Linking out with the misguided belief that it'll help your rankings is just retarded.

Although I do applaud you for correctly disclaiming your opinion as such.

[got fish?]

I don't think it will help hugely but to think it won't help at all is silly.
A site with no outbound links is a dead end, google would be more interested in sending someone to a site that links to other relevant information.
They send people to a directory named "Health and fitness" and on this page they find 100 of links to other "health and fitness" sites.
Why wouldn't google think that is better than a site that links nowhere?

It must be at leat partly responsible for PR. Though not entirely and not greatly.

[aspen]

PR is solely a measure of incoming links. Even people who think outgoing links help, like randfish, know that they don't factor into PR.

So, based on that gross mistatement, how can anyone seriously consider the rest of your post?

Now, to answer your question. Google cares about sending their visitors to authoritative sites. In general the more incoming links a site has the more of an authority it is. Google doesn't care about outgoing links because outgoing links in no way, shape, or form measure the usefulness of a site. If a site is more useful people link to it, they don't add links to it. If outgoing links actually do increase the usefulness of a specific site it will in the long run be rewarded by increased incoming links, thus Google will rank it appropriately without having to consider outgoing links in it's algorithm at all. This is why PageRank, or the measure of offpage factors like incoming links, was, and still is, so revolutionary.

[RRWH]

I disagree that PR is solely a measure of incoming links!

I have a domain that has 0 inbound links, and a single page with around 35 outbound 1 way links.

The single page was submitted to google and it was crawled

It has a PR of 3.

[got fish?]

I disagree that PR is solely a measure of incoming links!

I have a domain that has 0 inbound links, and a single page with around 35 outbound 1 way links.

The single page was submitted to google and it was crawled

It has a PR of 3.

Yea i agree, that makes sense to me.

For people to think that incoming links are the only thing that contributes to page rank underestimate how compliacted it is for google to calculate page rank.. They surely cant think that page rank™ is actually just a case of counting links. Its far more complicated than they think, and takes into consideration far more factors than just incoming links.
It has to, otherwise good page rank could be achieved just by spamming.
There would also be a guide saying "1-100 links = pr2 , 101-500 links pr3" , but this dosn't happen. Why? because there are more factors contributing to page rank than just incoming links.

[aspen]

For people to think that incoming links are the only thing that contributes to page rank underestimate how compliacted it is for google to calculate page rank.. They surely cant think that page rank™ is actually just a case of counting links. Its far more complicated than they think, and takes into consideration far more factors than just incoming links.
It has to, otherwise good page rank could be achieved just by spamming.
There would also be a guide saying "1-100 links = pr2 , 101-500 links pr3" , but this dosn't happen. Why? because there are more factors contributing to page rank than just incoming links.

PageRank is a proper noun named after Lawrence Page, the cocreator of Google. It is a system whereby pages are assigned an authority score based on the number of citations or incoming links it has.

Your page's rank is not a proper noun, it is a phrase usually referring to the position your page or site is listed in the SERPs (search engine result page).

PageRank is only a measure of the weight of your incoming links, nothing else. Your page's rank takes into account both your PageRank, other off-page features such as anchor text, and onpage features such as title and keywords.

Sometimes I wish Lawrence Page was polish. OzeranskiRank wouldn't create so much confusion.

So before accusing me of not understand or underestimating Google's algorithm, perhaps you should make sure you even know what the terms you are using mean.

I disagree that PR is solely a measure of incoming links!

I have a domain that has 0 inbound links, and a single page with around 35 outbound 1 way links.

The single page was submitted to google and it was crawled

It has a PR of 3.

This is not uncommon. Firstly Google does not list all incoming links, so chances are you have some you don't know about. Secondly all pages have a minimum PageRank of 0.15 (or 1-d where d is the dampening factor). So a site without any external incoming links can still generate PR to a factor of totalpages * 0.15.

[got fish?]

right , symatics.... a trolls best friend.

[aspen]

Its not semantics, its two very different things with similar names thanks to Lawrence Page being blessed with such a generic last name. The fact that you didn't know the difference just underlines your lack of credibility on the issue in general.

[got fish?]

We all know what page rank is, regardless of whether we know its orrigins or not.
A poster above already said he has a site with a pr 3 with no links to it at all.

[skanxalot]

With all do respect, your posts in this thread show that you in fact don't know what PageRank is, and Chris has done well to inform you. Anyone who knows the algorithm knows that you cannot get PR without links.

With respect to RRWH's post, you need to understand...

1) Google does not display all of the links it is aware of.
2) We do not know the actual PR of a site. The Toolbar is not updated frequently, and its accuracy is questionable at best.

[randfish]

aspen - I believe that outgoing links are absolutely an input into the search engines' measures of the quality and relevance of a page. If a page on the topic of spiders links to a scientific american article on cave spiders, the wikipedia entry on spiders and a page on what spiders eat, and another page on "spiders" links to Google's "spider" data, information about web crawling, etc. - you are going to see the SEs recognize the relevance of the outgoing links as making the page about one subject or another.

The subject of the pages you link to influence how the SEs view your page. Whether adding a link to one particular site or another can boost your rank is a different discussion and one that's more debatable (although I believe that to be possible as well). But, I don't think anyone should question that SEs look at who you link to in regards to your topic, subject-specific community (esp. Teoma) and the relevance of your content.

[redhits]

The cath it's that if you will offer free submision (without reciprocal link neaded) then , a lot of webmaster will add a link to your website ,telling to others webmasters/seos... "free submision"

[aspen]

aspen - I believe that outgoing links are absolutely an input into the search engines' measures of the quality and relevance of a page. If a page on the topic of spiders links to a scientific american article on cave spiders, the wikipedia entry on spiders and a page on what spiders eat, and another page on "spiders" links to Google's "spider" data, information about web crawling, etc. - you are going to see the SEs recognize the relevance of the outgoing links as making the page about one subject or another.

The subject of the pages you link to influence how the SEs view your page. Whether adding a link to one particular site or another can boost your rank is a different discussion and one that's more debatable (although I believe that to be possible as well). But, I don't think anyone should question that SEs look at who you link to in regards to your topic, subject-specific community (esp. Teoma) and the relevance of your content.

Prove it.

[ccourt23]

Outbound links do not help, its a myth.

It does not help in my opinion....but If it does it is very very very small part of the algo prob. lower than meta tags

[In10sity]

Breaking news: Outgoing links will help you with the SERP's. I might take time for the SE's to include the new find in their overall ranking algo of your site.

Example. On my web service related site I had listed keyword ranking results that I had achieved for one of my basketball sites. On this page I had displayed the Basketball related terms and had linked those important keyword phrases to each given SE so clients would easily confirm this on the SE's.

This page had no incoming links that were slightly basketball related. In fact the link anchor then had displayed "SEO Portfolio"

For about a month I was in the top rankings for those top keyword phrases until I ended this by adding the appropiate robots tag.

The site had no related incoming links.

On page I might have had about 200 characters worth of descriptive text.

Overall I had about 10 Basketball Terms linked to each SE keyword query.

To back this up: I have Awstats.

And like RandFish once said

Let me say publicly that if I ever discover ANY information on SEO answers I will publicly share it everywhere. I dont want to spoil other SEO's monopoly on their Google ranking results, but I think it's my responsibility as an SEO... to give back to the community as best I can.

And he is doing just that. I have found his feedback given to the SEO community to be quite credible and informational. I agree on his point on the worth of outgoing links.

[subnet_rx]

I control content, which obviously affects SERPs position, why wouldn't links, which I also control, have an effect? I can see why META tags would be not counted, since they aren't visible to the user.

[aspen]

I control content, which obviously affects SERPs position, why wouldn't links, which I also control, have an effect? I can see why META tags would be not counted, since they aren't visible to the user.

Why would they? By your logic everything you control should count, and thats just silly.

Really, this is a stupid argument to be having. There is empirical evidence showing outgoing links do not help your ranking on Google.

Outgoing links may be used in spam filters. Outgoing links may be used when deciding which keywords your site will be ranked on. However they are absolutely not used to rank your site.

Breaking news: Outgoing links will help you with the SERP's. I might take time for the SE's to include the new find in their overall ranking algo of your site.

Example. On my web service related site I had listed keyword ranking results that I had achieved for one of my basketball sites. On this page I had displayed the Basketball related terms and had linked those important keyword phrases to each given SE so clients would easily confirm this on the SE's.

This page had no incoming links that were slightly basketball related. In fact the link anchor then had displayed "SEO Portfolio"

For about a month I was in the top rankings for those top keyword phrases until I ended this by adding the appropiate robots tag.

The site had no related incoming links.

On page I might have had about 200 characters worth of descriptive text.

Overall I had about 10 Basketball Terms linked to each SE keyword query.

To back this up: I have Awstats.

Read this:

http://en.wikipedia.org/wiki/Control_experiment

before making claims based on nothing more than passive observation.

[projectphp]

Now, this is a REALLY old thread @ WMW, but it still is worth a look in here: http://www.webmasterworld.com/forum10003/3219-2-15.htm

Of course, folks never know when we're going to adjust our scoring. It's pretty easy to spot domains that are hoarding PageRank; that can be just another factor in scoring. If you work really hard to boost your authority-like score while trying to minimize your hub-like score, that sets your site apart from most domains. Just something to bear in mind..

OK, so its from 2002, and anything GoogleGuy says should be taken with a grain of something bad for the heart but good on Chips, but it is interesting nonetheless.

IMHO, Authorities (about something) and Hubs (link to sites about something) are two elements that have always existed as ranking elements. Can I prove it? Hell no! But if Descartes could question whether or not we exist, I think we can give some leeway for something as unimportant as SEO.

http://en.wikipedia.org/wiki/Control_experiment

Poor research methodology in SEO. You can't isolate anything, nor prove that the SE in question doesn't change during the experiment.

Scientific methodology is a very poor research methodology for many things, such as marketting and most human related endeavours (like psychology), none of which use control experiments.

In fact, studies on human behaviour often disprove themselves, and are inputs back into the thing they studie, thus creating a loop. As an example, Catherine "Kitty" Genovese was murdered in NYC in 1964. With 40,000 murders since, what still makes Kitty's stand out is that 38 people were awakened by her screams from an the attack lasted approximately half an hour and no one called the police. this is one of the most studied murders ever, as people came to grips with this chilling fact, and teh seeming heartlessness of the witnesses.

From this research, it was determined, was that if more than 4 people see something bad, all feel there is no need to help as "someone else will", and the ammount people help is inversely proportional to teh number of people that each is aware of that sees the event. This is known as the Genovese syndrome or bystander effect, and the dodgy equation is this:

Individual responsibility = Event / number of people seeing it.

Thing is, if you know this syndrome exists, you are more likely to do something. So, this effect exists until it is known, and then its existence changes human behaviour as people take on the responsibility because they kno no one else will.

SEO is very much like that. I can't prove that the SE I study today is the same one I study tomorrow, ergo I can't use a research methodology that relies upon the isolation of factors, as I can't isolate anything. I can make educated guesses, but never to teh level that would satisfy any sort of testing rigour.

So, back to the start, you need "proof", and I can't offer any . I can, however, offer up research papers that discuss the issue and, as with many issues in SEO, leave it for each person to decide whether or not they feel this is currently a factor in any algo or not.

[aspen]

Now, this is a REALLY old thread @ WMW, but it still is worth a look in here: http://www.webmasterworld.com/forum10003/3219-2-15.htm

OK, so its from 2002, and anything GoogleGuy says should be taken with a grain of something bad for the heart but good on Chips, but it is interesting nonetheless.

IMHO, Authorities (about something) and Hubs (link to sites about something) are two elements that have always existed as ranking elements. Can I prove it? Hell no! But if Descartes could question whether or not we exist, I think we can give some leeway for something as unimportant as SEO.

That quote is taken completely out of context and is completely unrelated. I remember when he posted that and I remember the confusion it caused.

It was a response to a discussion that was centered around people not linking out so as to not share their PageRank. He said that it would be possible for Google to detect such behavior. He did not say if they plan on detecting such behavior, or to what end would they do so. I think its fairly obvious that the most logical conclusion is that he was talking about a possible spam filter for sites with unnatural linking schemes rather than alluding to some sort of bonus for linking out from your site.

Poor research methodology in SEO. You can't isolate anything, nor prove that the SE in question doesn't change during the experiment.

Um, yes you can. Its not that difficult. Think about it a little bit. All you need is 2 identical* pages (better more so you know its not a fluke, but 2 is the minimum). They're really not 100% identical they differ only in outgoing links. Some pages have no links and act as a control group. Then its just a matter of checking rankings. Of course the search engines can change overnight, but you can check rankings daily if you like. I've done it, and the experiment has been live for the past few years and not once has it shown a bonus for outgoing links.