Over 40 Questions to Ask a Search Engine Optmization Company Before You Hire Them

[Gloyns]

They may count, but if a large percentage of your links are from the same ip or range of ip's then the value would be diluted.

[deepsand]

They may count, but if a large percentage of your links are from the same ip or range of ip's then the value would be diluted.

Nope; not the case.

Not only is there no theoretical grounds for expecting that to be so, but SEOmoz.org's latest statistical data show that the correlation coefficients between rank and number of Domains of like and different C blocks are identical, with a Spearman value of 0.21 each.
.

[claybutler]

Nope; not the case.
Not only is there no theoretical grounds for expecting that to be so, but SEOmoz.org's latest statistical data show that the correlation coefficients between rank and number of Domains of like and different C blocks are identical, with a Spearman value of 0.21 each.

Thanks for digging that up Deepsand. I get tired of all the newbies repeating the same unverifiable nonsense. Not only does statistical data matter but so does experience. Experience tells you that links from the same IP count. Not to sound like a grumpy old man, but I believe there is difference in the way people think about SEO if they started out BEFORE Google dominated the world or AFTER. I remember trying to figure out why my sites weren't doing better in AltaVista and Hotbot. In those days many of us also built our own sites - real sites that reflected our business or personal passions - and our interest in SEO was ignited by our desire to help our own sites do better. The idea of flooding the world with shit to manipulate rank to sell ads was not on most of our minds. However, this seems to be the dominate ideology now. Most of these SEO "experts" don't have a single domain they actually care about. They don't built anything except machine generated junk. The internet is nothing more that an entity to extract a few pennies of wealth from at any cost. And since their experience and interest is superficial, and they don't have the benefit of time to see what's real and what isn't, they run around spewing the latest fad theory that they read on some forum. I truly believe that intent informs knowledge. Bad intent begets bad (or at least incomplete and superficial) knowledge.

So that's my rant.

[Gloyns]

Apologies if I've ruffled a few feathers here - I can only talk from my own experience and findings.

Would the above chart suggest that having links from a large number of C class IP's is just as important has having links from a large number of domains? If you had links from several hundred sites to yours would you want all of those links to be on the same IP range? Or is the argument that if these sites are quality, then this will outweigh the IP issue? Open to discussion, but perhaps in a separate thread rather than filling this one

[deepsand]

Said statistical data show that the degree of correlation between the number of unique Domains linking to the target and the rank of the target is the same regardless of either the IP Class of the the source Domain or the presence/absence of "nofollow."

I.e., given equal source Domain quality, X no. of "nofollow" IBLs from the same C Block are as efficacious as are X no. of followed IBLs from X different P blocks.

All that here counts is the total no. of unique Domains linking in.

[weegillis]

If you had links from several hundred sites to yours would you want all of those links to be on the same IP range?

You would want those links to be NATURAL, regardless of IP. Search engines can tell whether a site is a respected authority on the topic around which the referring site is linking to them for. If a site is about widgets and they are getting inbounds from travel Timbuktu and the likes, the authority goes right out the window and the link becomes 'a paid link' whether it is or not.

[MikeinFlorida]

Man Clay you sure got it right away as the first res-ponder to this post! This list would drive anybody crazy from the beginner to the old experienced pro! But I do have to say it's a good way to reflect on what questions are important to ask before you hire someone.

[LD]

Ya, for sure, this is a big list of possible questions someone might ask, but it's pretty clear one would best pick and choose the q's that are applicable that will give the most information so one could make an informed decision. If I was on the other side of the prospects desk and saw this list of q's he/she was about to go through, I'd freak. Well, I wouldn't actually freak - outwardly anyhow.

...Timbuktu and the likes...

Timbuktu /ˌtɪmbʌkˈtuː/; Koyra Chiini: Tumbutu; French: Tombouctou), formerly also spelled Timbuctoo, is a town in the West African nation of Mali situated 15 km (9.3 mi) north of the River Niger on the southern edge of the Sahara Desert.

...just for a geo-specific reference. It's Friday.

[LD]

Dang it. I was trying to post for count to get closer to 1000.

[deepsand]

Dang it. I was trying to post for count to get closer to 1000.