I thought I’d follow up my commentary on the “Crappy Link Building email” I had back at the end of November and see what sort of links mobile.broadbandgenie.co.uk have got now.

Using Link Assistant (which everyone should be using to at least understand the link landscape of their industry) I’m just going to take a look at the landscape of the links and give my opinion.

Obviously its going to be difficult to differentiate between the links from good relationship-built linking and the mass-produced mail-outs, but I’m sure there should be some good evidence of quality links somewhere …

First off, the tool found 1750 links (this morning) via querying the international databases of Google, Yahoo!, and MSN – that’s a pretty handy amount.

I suppose that its not too uncommon for one keyphrase to form a large part of the incoming anchor text, however, you’d usually expect that to be the name of the site, not one like “free laptop deals” – that would appear a bit un-natural. Would Google think the same? Well at the moment they’re #4 in Google for that phrase at the moment. The site is #3 for “mobile broadband” which is also not bad.

So you’re thinking that all those automated linkbuilding requests have done the job? Right? Maybe you’d be wrong to say that. Of all those 944 links they’d got with “free laptop deals” were rel=nofollow links. Now its interesting to note the PR$ on that page must have been passed on from the homepage PR6 and NOT from the direct link building efforts for the target page.

But wait, there’s more. It appears that all of those “free laptop deals” links are from one site – PC Advisor. Its doubtful that PC Advisor would co-operate in a content for links deal when they employ content writers already. Hence we can safely cross off 53.9% of the inbound links from their automated link-building efforts.

Of the next 16%, 14.8% (of all links) were from cached pages of other sites. So we can now cross off 68.7% of the links being from the automated tool they used. Actually all the other pages listed either are dead, are paid for or negotiated non-text links or site search results pages (which Google are cracking down on) apart from one (PRO) that could have been the the result of the link building effort (http://fucc.blogspot.com/2008/06/broadband-genie.html) So I’d say that the running total of links not being attributable is now 69.8%

Looking at the next load of links account the VAST majority of them are from an advertising feature such as on the right hand side of http://www.mobilecomputermag.co.uk/20080529536/hands-on-with-the-msi-wind-ultra-portable.html or the site’s designers Outplay Media – that equates to another 14% (83.8% running total). All the remaining 8 with that anchor text had PR0 – Including http://broadbandeurope.biz/2008/12/24/modern-broadband-technology/ which does appear to me to be part of the content for links type approach. Also there’s http://www.omc10anscasuffit.org/technology/an-analysis-of-mobile-broadband/, http://www.spacetruck.net/technology/3-mobile-customer-base-increases/ and http://www.w2krp.com/technology/new-prepay-mobile-from-t-mobile/ which also could be as a result of this method. I’m really struggling to see the value so far.

So far its looking like 83.8% of the links, and the high-value ones at that don’t appear to be as a result of the email-based link-building campaign. What links I believe are as a result seem to be of poor value.

So looking at the next 10.2% that came from the “mobile” internet phrase, all of them came from Webuser and their advert.

So now we’re up to 94% being non-attributable to the method in question.

Looking at the next 1.9%, 1.6% of the total were the result of a sponsorship of CrowdStorm – see the bottom. Now we’re up to 95.6%.

In this section there was a very handy link from the Hitwise blog – doubt that was an arranged link. There was a link from the CrowdStorm blog saying they’ve penned an advertising deal with Broadband Genie – another one to cross off the list. But this PR0 link http://dsiy.com/archives/2008/12/15/mobile-broadband-for-a-merry-christmas/ from the DSIY Blog (never heard of them). I’d say we’re up 95.8% non-related.

The next 0.4% I would say are related, they’re all PR0 pages such as Alternative Broadcasting, Aibtexas.org, Barovci, Darrenoronofsky.net, Janas Bear Cottage, Karla Van Huysen and Solitare007. Now I don’t want to be over-critical here, but I don’t see them being sites with a massive foot-fall or with a high-degree of relevance to the mobile internet market.

Looking at the next 4 links for there is one PR3 for Curha, back in July 2008, eventually that’ll move off the front page. The rest are low volume sites with PR0 links.

In the next 4 links they’re on relevant blogs but all are comment links with nofollows. So we’re no up to 95.8%

The next 4 appear to be as a result of the email link-building, but all have PR0 and use the domain as the anchor text.

The next load were from online PR (I believe) rather than a specific link-building effort – Ciaron Dunne did an interview about Basecamp on the 37Signals blog. This page has a healthy PR5 to boot. So its 96% now.

The next 0.2% are also comment replies – now 96.2% not being related.

Next we have the Compter Active homepage again as an advertsing link, and a couple of blogs such as Blogging Secret. 96.25% not relevant now.

And the rest? They’re virtually all from advertising, their own sites, press releases or PR0 blog posts.

So what are my thoughts? Well it depends on the price and objectives. Just looking at the evidence from one site, I’d say its not the best way to build links. As most (47.6% of the links to the site are PR0 – and I’d say not likely to improve substantially over the coming months. Overall 83.8% of the links to the site are PR2 or below. Now that may not be un-natural for a site that wasn’t actively link-building. But for a site that’s spending money, I’d say it was disappointing, especially when the high PR links are from proper, useful advertising.

And another point, the data shows that 29.6% of the links are reciprocated.

I’m trying to not be too harsh. The data may be inaccurate, my understanding of their motives may be wide of the mark, but I wouldn’t pay for a load of links from non-relevant blogs. I wouldn’t say that was “ethical”.

I welcome all parties to respond and highlight if I’m wrong in my assertions or am using erroneous or incomplete data. My motives aren’t to damage any company or website, but purely to highlight that if you’re hiring link building agencies to pay on performance – link quality is the key.

Don’t you bloody hate it when you send someone an email but forget to send the attachment? I’ve done it countless times and its a pain in the arse when its a proposal. Thankfully when I did it today Google stepped in:

Yep, I put some words in the email, which they scanned and then saw that I didn’t have an attachment. Good work.

I just get so fed up with all this privacy bull shit. People just want something to talk about, to create some “buzz” (excuse the irony) so they can get more links and visitors to their blogs.

I’m just bored with all this Google bashing. Blog about something useful to help people attract transactionary visitors please. (bugger! there’s some hypocrisy in this post!)