More of the same= better? (sitewide links)

[Cecil]

Hi all,

Here's a thought to consider. What is everyone's feeling towards site-wide links versus getting a singular or few links when it comes to establishing reciprocals? Of course more is always better, but my question for debate is how MUCH better.

Consider the following numbers of links with an equivalent reciprocal between two sites:
1 link
10 links
100 links
1000+ links

Does anyone have a good feel for a rule of diminishing returns? Obviously 10 is better than 1. And I'd be willing to bet that Google will still give priority to a site with 100 links. But once you get into the 1000's, do you think it makes much of a difference if you have 1,000 vs 10,000 backlinks from a particular site?

[SPARKO]

Definitely the more links you have the better like you said; since it's based on a logarithmic scale. That is until you get to be page rank 10 which seo guy seems to be on his way doing. Look at how many links this site has for a pr8; the jump to 9 though I would imagine to be impossible.

Do you mean say for example two sites with PR7, but one has 100 more backlinks than the other

[Cecil]

I was thinking in terms of four sites A,B,C,D
A & B have 10000 reciprocal links to each other within their internal pages
C & D have 1000 reciprocal links to each other within their internal pages

Off the top, A & B are probably going to do better if all other factors are the same. The thing I'm getting at is how MUCH better. If there is a law of diminishing returns, then those 9,000 extra links between A & B could maybe be balanced by just a couple more backlinks from other sites (like sites E-G) to C & D.

The idea I'm going after is "since it's based on a logarithmic scale" it seems that the dropoff in improvement would be dramatic once you get x number of links within a single site.

[Pyrrhonist]

I was looking at Adobe's (pr10) backlinks earlier today, and couldn't help but notice that they have about 123,000 backlinks... That's a lot.

I looked a little further, and there's links to them from w3.org from a pr9 page... hmm... i don't think i'm ever going to catch them.

[seo guy]

Cecil, although knowing exactly how many links and what weight are returned after the first you are right there is not much different from a PR standpoint. If I link to you from a sitewide PR 7 with 2000 pages (With 10,000 weight points on the main page) and then I link to you with a 20,000 pgaae sitewide PR 7 again with 10,000 weight points on the main page my bet is that the amount of PR contributed from each would be very close to equal. So whats the benefit? Good old advertising, the more pages your links on the more people might see it and click on it, thats about the only benefit but hey a decent one IMO. This applies as you say to many many pages and that dimished return is less and less for lower pagecounts I would say being sitewide on a site with 10 pages is 1 1/2 times as good then just being on the index page. But thats just a rough guestimate based on experience

[seo guy]

Btw cecil you get no credit at all if you dont post a site in your sig :-) You have some great questions so I'd be honored if the forum showed up in your backlinks :-)

[SPARKO]

I'm glad seo guy's answered that one; logarithmic stuff takes me back to my degree days ooh hated maths which wasnt a good thing doing a degree in electronic engineering. like the main man says post ya site, I'd like a peak 2...

[Cecil]

[QUOTE=seo guy]Btw cecil you get no credit at all if you dont post a site in your sig :-) You have some great questions so I'd be honored if the forum showed up in your backlinks :-)[/QUOTE]

Hmm, you know if it were a dynamic signature that I could retroactively remove, I would post the link. But I'm a little hesitant to have the competition see my backlinks and read through a step by step demo of our SEO strategy once the forum gets indexed! :eek:

The sig should be dynamic like the avatar is. Thanks for the complement though, it's a nice forum. I'll be sure to badger you all with more questions ;)

[vord]

[QUOTE=Cecil]
The sig should be dynamic like the avatar is. ;)[/QUOTE]

The sig is dynamic in another vbulletin site I frequent - is it not the same here? Change the signature in your profile, and the signature changes in all of your posts (but isn't added to previous posts if there was no sig there to start with).

Try it for an experiment.

:)

[Pyrrhonist]

The sig is loaded from the database every time that a user accesses a post (in the same call as the avatar, username, title, and postcount). So yes, Vord, this makes the signature dynamic.

[Cecil]

hmmm, well now that's odd. I actually tried that experiment before posting and I found that after creating a sig, it did not add a sig to previously made posts. Maybe there's a field that is empty if the post is generated without a sig but if you make a sig the field is used so it travels to the database to see what the latest sig happens to be. I'll have to try again sometime.

[Cecil]

Here's the post with a little S&M "smiley" whipping the sig. It'll be dynamic if you don't see that.

EDIT: yep, that confirms it. Sigs are either off or on. If on, they are dynamic. However, if a post is created without a sig it will not display any sigs in the future. (sorry, that little experiment was a bit off topic but I wanted to do it in same post that the issue was mentioned)

[Pyrrhonist]

Yeah, there's a flag in the post (if you look at the quick reply box, there's a check for show your sig)... that's probably set to off if you don't have a sig saved, and if you do, then to show it is the default.

[Kevin 11]

Back to the idea of diminishing returns on links...

I've also wondered about this. If you are buying sitewide links, they are a lot more costly than one link from the main page. So, if there is say a 150% benefit to sitewide links vs. 1 main link, then it doesnt make financial sense. My feeling is the effects of sitewide links diminish as the amount of total backlinks increase, but in the case of sites with less than 1000 links for example, the sitewide links may have a greater impact.

If I site has for instance 500 backlinks, and 50 of those are for individual links and the other 450 are from one domain, sitewide, then this site will likely do alot better than a site with say 200 individual links. This is what I've noticed anyways.

Now, if we are talking about a site with thousands of total backlinks, then I think the effects of sitewide links will diminish even moreso than one with fewer total links.

I havent done any conclusive testing on this, but I am very familiar with a number of sites that have about 800 backlinks and 80% are from one domain sitewide. These sites do very well against other sites that also have lots of links, mostly from independant domains, so it seems that the sitewide links are being weighted competitively against unique domain links.

Kevin 11

[douglaslampi]

I've been graphing my progress in link popularity with the great tools at

www.marketleap.com

Recently, all of the results took a big nose dive. One of the search engines used to count each link in site-wide linking strategies - and gave great numbers that were easy for webmasters to inflate by adding a link to their own site from the bottom of every page on their client's sites...

Now that one search engine reports your link popularity by domain only - just like all the others.

I'm inclined to think that many links from one IP address are of no greater value that two or three links from one VERY tighly themed pages with fantastic Ratios between incoming and outgoing links.

After managing the link campaigns for dozens of different websites with various linking criteria, I'm more convinced than ever that just a few super-high-quality links are worth more than hundreds of average links.

Quality, not quantity. It's all in your Reputation (http://www.linkpopularity.ca/link_reputation.asp)


Regards all,

Doug