After 5 years of working at iCrossing (previously Spannerworks), I’ve finally decided to take the extremely difficult decision to move on.  Today is my final day as Head of Natural Search and as an employee of iCrossing.  The last 5 years have probably been some of the best of my life, and all I have are good things to say about iCrossing and Spannerworks.

I started iCrossing after a bad start to my career.  After graduating from Sussex Uni with a degree in Maths and Artificial Intelligence I decided to move home to parents, where I got a job working as a software developer.  I very quickly realised that neither the job, nor the culture of the company I was working for was a good fit – something that I’ve subsequently come to realise is one of the most important things when choosing a company to work for.

I decided that I wanted to move back down to Brighton, and a friend of a friend (and another Sussex AI grad) was then working for a company called Spannerworks, doing something called Search Engine Optimisation.  At the time, my knowledge of the internet was limited, but the then Head of Search, Nilhan Jayasinghe saw something in me and decided to take a punt on me.  As it turned out, I actually found SEO very interesting, and a real match to the analytical skillset that I developed during my academic career.

Since I’ve been here the company have given me some fantastic opportunities and challenges.  In 2007 I got the opportunity to visit SES San Jose, where the obvious highlight was attending the Google Dance at the Googleplex.  In the same year I summited  Mount Kilimanjaro with a group of colleagues to raise money for Amani Children’s home in Tanzania.

That’s me looking the wrong way on the back row, second from left.  The chance to travel, and work with colleagues in the US has been great, and I’ve learned a lot from the SEO team over there.  More recently, my role as Head of Natural Search has allowed me to push myself even further, speaking at both SMX London and SES London in 2009.  This achievement in itself speaks volumes in terms of how I’ve been encouraged to grow at iCrossing – 5 years ago I don’t think anyone would ever expect me to be doing that!

However, 5 years is a long time to do anything, and unfortunately I’ve reached a point where I’ve got a desire to spread my wings, and experience life on the other side of the fence.  That’s right – it’s time to move inhouse!  In October, I’ll join Review Centre as Head of Search and Social Media.  For me this job title is significant – search is no longer what it used to be.  You need to take a joined up approach that encompasses both disciplines, and I’m looking forward to getting more involved in the social side of things.

Before I join Review Centre in October I’ve a few things that I need to do – one of which is racing Tuk-Tuk’s around Sri Lanka to raise money for a children’s home.  I’ll be accompanied by a good friend of mine, David Coats.  If you’re feeling generous and want to sponsor us, you can do so over here.  Hopefully we’ll be tweeting our adventures from here.  Before I hit Sri Lanka I’ve got a short trip to India – a country I’ve always been fascinated with, so I’m really looking forward to that.  I look forward to seeing you all on the other side

Tweet This Post

1. Does QDD Give an Inherent Boost to Negative Subject Matter?
2. The 302 Hijack is Back?
3. Are the Engines Following URL Shortening Services?
4. What Was Up with The Hyves Subdomain PR Penalty Checker?
5. The Engines All Regularly Follow Many More than 100 Links Per Page?
6. Is it Really Harder to Get Rankings with a .info, .cc, or .biz Extension?
7. Does Google Employ Link Buying Moles?
8. Will Anchor Text Value Pass Through Terribly Low Quality Links?

And here are my thoughts on them:

1. I’ve certainly seen instances of this happening – negative blog posts ranking highly, but only in a way that suggests QDF is at play rather than QDD.  Explanation of the difference can be found here.
2. I’ve seen no evidence of this
3. Not sure, but will be testing
4. I really don’t know.  Being in the UK meant I missed out on seeing this in action.
5. I think Rand’s thoughts on this are pretty logical, and match what I’ve seen in other cases not related to links.  Google seems to handle certain elements differently between websites according to the trust / authority of each given website.  Therefore I wouldn’t be surprised if Google doesn’t crawl past 100 links (perhaps less) on pages on new domains, but will crawl many more links on pages on established domains.
6. Had to work with one of these domains once and it was hard work – it was a long time ago, when Google’s sandbox was in full swing, and the domain was new.  Perhaps it was sandbox related rather than TLD related.
7. Sheesh,  I dunno.  I wouldn’t have thought they’d go to those lengths, but surprisingly they do seem to be having problems dealing with this issue algorithmically
8. This is a really good question.  Page Rank not passed, but anchor text benefit is passed.  I’d say this is a definite yes from what I’ve seen, especially with certain websites on the Digital Point Ad Network doing well at the moment.  This would be a real good indicator that anchor text value on low quality links is being passed.

Tweet This Post

The other story I picked up on today is Patrick Altoft’s post on Google passing anchor text through 301 redirects.  This is really interesting, because I’ve noticed quite the opposite.  As I mentioned previously, my site has been sitting around gathering dust for a while.  Before I was Head of Natural Search at iCrossing, I used to use my website as a test bed for evaluating link techniques.  I used to be a keen photographer, and one of the best ways I found of generating good links was by uploading my photos along with some text content, and then asking websites relevant to those photos to link to me.  Within the last 6 months I decided to drop all that content, and 301 redirect everything to my homepage.  I was aware that Google were dealing with Googlebombs,  and I guess without putting to much time into thinking about it, I thought that the easiest way for Google to do this was to check that the text within external anchors matched onpage content.  If you think about the “miserable failure” example, there would’ve been hundreds of links with that anchor text, but it never actually appears within the content on www.whitehouse.gov, and so it’s clearly a Googlebomb.

With that in mind, I was quite surprised when I started noticing that the homepage of my website started ranking for terms that weren’t present in the homepage content.  One example of this can be seen here:

I wasn’t particularly happy with that showing in the SERPs, especially not the snippet text which is a relic from Dmoz, despite the use of the NOODP tag.  If you’re checking up on that result, it hopefully shouldn’t be there for much longer because I’ve now replaced the old holding page with this blog.

As Patrick says in his post, when you check the cached snapshot, Google tells you that “these terms only appear in links to this page”. However, I started noticing that I was only ranking on the homepage for certain terms as a result of redirecting old pages.  What was it that was causing my homepage to rank for some terms but not others?  After much digging around, I came up with the following theory – Google doesn’t just look at page content to determine whether a link is trying to cause a Googlebomb.  Google also looks at other onsite factors, such as internal anchor text.  If for example URL A has lots of external links that say “miserable failure” but URL A has no content to match that anchor text, then it’s clearly being Googlebombed.  However, even if URL A has no content to match the anchor text, it can satisfy the same criteria through internal anchors – perhaps an internal anchor that just says “miserable” and points at URL A is enough to tell Google that the external anchors shouldn’t be considered a Googlebomb.

How did I work out this was happening?  I had some legacy pages on my website (I still do), that have odd bits of anchor text pointing to my homepage.  Wherever these odd bits of internal anchors match external anchors, then my homepage continues to rank.  Where they don’t match, my homepage no longer ranks.  As I said previously, it’s very much a theory, but it seems to apply to all cases I’ve had to look at.

Tweet This Post