The Most Dangerous Link Building Myth

(in Google)

"Incoming links can’t hurt you… the most they can do is ‘not count’ ."

In my opinion this used to be true.

In 2006, it’s absolutely, certifiably, completely FALSE.

Your neighborhood better maintain a certain level of trust — both trust IN and trust OUT — or you can kiss your rankings goodbye.

To put it another way…

1. You have six trusted links to your site, it’s ranking on page 2.
2. Then, add 10,000 untrusted links to your site. These should theoretically just "not count", and you should still rank on page 2… right?

Then why do such sites go byebye? (I’ve seen it time and time again in the last 6 months.)

I Can’t Rank for ‘Purple Widgets’… but I Rank for ‘Purple’?

So I have a new (quality) page up on a semi-sandboxed site. The page is optimized for purple widgets. As the resident link monkey, I made sure that no less than 10 out of the top 100 ranking pages (for purple widgets, on G) linked to us.

Predictably, the page isn’t currently ranking in the top 100 on a search for purple widgets. I do expect it to pop out, as, like I said, the site is only semi-sandboxed, and this particular page has all of those legit, relevant, themed links.

But here’s the kicker: the page ranks #11 now on a search for just purple! LBB readers… wtf? You got any explanation? (Before you say "OOP", please note that I wasn’t so stupid as to only use purple widgets as the anchor text — we got a lot of variation, including the odd "click here" and "Site.com" and whatnot.)

Results 1 – 100 of about 701,000 for purple widgets

Results 1 – 100 of about 2,560,000 for purple

Finally, note that neither purple widgets nor purple are competitive or "money" terms — but widgets in general is an extremely competitive affiliate space.

A little help please?

Exact Match KW Domains Rock in Google?

Well I wasn’t expecting something useful to come of the infamous v7n contest (aside from an annoying WSJ article), but I decided to do a quick n dirty analysis of the SERPs. Repeat: Quick. N. Dirty. No pretty spreadsheets or proprietary scripts like Todd or Greg could do. You know, just a glance from an "art not science" guy.

My cursory glance of the SERPs told me a couple things:

1. Links still matter in Google. (I know, I should win SEO-researcher-of-the-year)

but also

2. The "keyword in domain, exact match" factor counts for way more than you would expect.

Yeah. In the top 30 results, I found six exact keyword match* (exact = hyphens OK, but not other added characters) domains (and three were in the top ten):

1. http://www.v7ndotcomelursrebmem.com/
2. http://www.v7ndotcomelursrebmem.net/
3. http://www.v7ndotcomelursrebmem.co.uk/
4. http://www.v7ndotcom-elursrebmem.cc/ (sweet site btw)
5. http://www.v7ndotcomelursrebmem.nu/
6. http://www.v7ndotcomelursrebmem.ws/

(note that if you do the search now, you may get different results, blame everflux)

Now, I’m no statistician, but I’m gonna go ahead and propose that 6 out of 30 is disproportionately high (and certainly statistically significant). There’s a lot of monkeys participating in this thing (Google shows 2,780,000 results), and there aren’t very many TLDs (comparatively). If the exact match domain aspect were removed here, I am guessing maybe 2 or 3 of these would be ranking in the top 30.

I think the exact-match-keyword-in-domain-ranking-boost thing may be one of the things Google is doing to counteract the irrevelancy which is sometimes caused by the Sandbox (sorry, Mike).

For instance, NewSiteWithABrandName.com may be two months old, and only have 3 inbound links, one of which is somewhat trusted, the other two are junk. Is NewSiteWithABrandName.com going to rank for it’s mildly competitive money phrase, "money phrase"? Not for a year or two or three. But it should still rank on a search for "NewSiteWithABrandName" (users are going to be p.o.’d if a press release announcing the site shows up #1 instead). So, Google gives NewSiteWithABrandName.com (and NewSiteWithABrandName.net, etc.) a healthy boost on a search for "NewSiteWithABrandName". As far as I know, this wasn’t happening (at least, this strongly) four or five months ago.

Alright readers, the "comments" widget below is where you pick apart my theory.

How the FRACK is Link Building Blog still sandboxed?

I’ve posted on the sandbox before, and claimed that I know how to beat it. And I have beaten it before. Really.

But that doesn’t mean I can’t get frustrated and confused and lose a battle now and then, just like everyone else. For instance: How the FRACK is Link Building Blog still sandboxed?

What do I mean by sandboxed? Well, the site doesn’t even rank in the top TWO HUNDRED for a fairly non-competitive term
("link building"), and it’s pretty frackin’ optimized for this term. A few more facts- the site:

• is 7 months old
• has legit deep links from tons of authorities in our niche, including Search Engine Watch, Pandia, Search Engine Blog, Search Engine Roundtable, SEO Book, heck, even the current #1 ranking site for our target term, "link building" (andyhagans.com)
• hasn’t had any active link building done (e.g., directory submissions) that might raise any red flags

And I’ll admit it, all this has me mighty confused. I’ve beaten the sandbox with total crap sites that only had a few trusted links, along with a slew of actively built links from directories, article sites, etc. So how the frack does Link Building Blog get no love from Google?

My *definitive* guide to what exactly the sandbox is, how it works, and how to beat it

Now, before you read any further, I should tell you that I’m not interested in hearing any arguments. If you disagree with my opinion of what the sandbox is,

well, that’s your God given right. And you’re wrong.

First, what the sandbox isn’t:

• A dampening for all new sites in Google, which lasts for a fixed period of time

Now, what the sandbox is:

• A dampening for all new sites in Google until they acquire trust (this usually happens after a site acquires highly trusted links, or after its somewhat trusted links have aged)
• Dynamic — Google can and does change what determines the threshold of trust which a site needs to break out of the sandbox

Which brings me to the conclusion…

How to beat the sandbox

• Acquire highly trusted links, OR
• Acquire somewhat trusted links and let them age

The first option is faster, the second option is more realistic for most sites.

And that’s it.

*p.s. hat tips to Massa and Aaron – I’m not the only one who’s figured it out