|Image via Wikipedia|
Patrick Altoft of Branded3
Patrick argues that not enough attention is given to link building (really?) and that once your on-site SEO work is done the vast bulk of your monthly activity should be the hunt for links.
It’s important to conduct a link analysis before you begin any work – without the analysis you won’t know which is the right strategy to follow.
One form of link analysis is to study the anchor text distribution of your client in contrast to their competitors. This will tell you whether you need more brand anchor text or generic. He points out that the voucher code space is one where many of the major players are lucky enough to have keyword domain names. This means they can build brand and generic links at once.
Another form of link analysis is to us SEOMoz data to study domain authority and link quantity as a graph. Domain authority is the x-axis and link quantity is the y-axis. Sites should aspire to be on the right and high up. This reflects their site having more high quality links than their competitors.
Patrick suggests there aren’t that many types of links as you can boil them down into less than a dozen categories.
Syndicated links are gained when content distributes around the internet and should be used with caution. It’s too easy to get too many poor quality links in this way. Make sure you have plenty of good quality links before you try and push syndicated links too hard.
Patrick explains that link exchanges have evolved and dismisses the idea that link exchanges no longer work. He says that link pages no longer work and reveals a more sophisticated model where sites producing news or blog content agree to reference each other in content posts.
When it comes to social signals Patrick has seen examples where they clearly seem to work but also cases when these signals appear to have had no impact. He suggests that the biggest impact in social may be from blogs which syndicate tweets that they’ve shared or which mention them.
Patrick explains that Branded3’s view on paid links is that they are only risky if they look like a paid link. A free link in among what’s clearly a paid-for blogroll is dangerous, he argues, but a paid for link hidden in the content of another otherwise well behaved blog is safe. The rule of thumb is whether you could show the link to Matt Cutts of Google and confidently lie about it.
The easiest links to get are often from the people you already know so create a prospect list and work through it.
Linkbait is hard, Patrick claims, and you can’t control it.
To wrap up, Patrick reminds the audience that bad links can hurt sites more than not having any links. He suggests that the two biggest “link building agencies” in the UK have their link networks rumbled by Google and that if you hire them, and end up on their exposed network then a penalty may be likely.
Lisa Myers of Verve Search
Lisa begins by pointing you that you need brains for link building. Tools are helpful because they enable you to work more efficiently – but you still need brains.
Lisa’s tried most tools but her slides mention Linkdex, Raven, SEOmoz and Majestic SEO.
A strong tip from Lisa is to know your audience. Go beyond knowing their age, gender and location and put yourself in their shoes.
Identify the social influencers in your subject area. Be prepared to reach out to them and engage.
There’s (professional) disagreement between the two presentations when Lisa starts to highlight how linkbait can be useful. She’s been able to get highly rated links from the likes of Apple.com. Verve Search discovered a connection between their client and Apple and following that up by reaching out to Apple.
Lisa’s zooms through over 50 slides in about 20 minutes (the presentation may become available later) and concludes by agreeing with Patrick in that bad links are worse than no links.
Key Link Building Strategies Questions & Answers
Question: Can having the same anchor text too frequently be harmful?
Answers: Patrick says that it can be. Lisa agrees but suggests you need a lot of this before it’s a problem.
Question: Can social signals impact rankings?
Answer: Lisa says they can and cites SEOMoz’s research in their area that seems to provide some statistical evidence that (some types of) social signals do influence rankings. Patrick points out that these tests are very hard to run cleanly.
Question: How has Panda impacted link building and link profiling?
Answer: Patrick suggests there’s been no impact – except that some sites might now need to work harder in order to maintain the same listings. Lisa disagrees and suggests that Panda could devalue the worth of some sites and therefore it could, in theory, could have had an impact on the link building landscape.
These are not my views. These are the summarised points of two basic SEO presentation given under the spotlight to a mixed audience. By the time you read this post it may no longer accurately reflect the views of today’s presentations. As a reminder; undisclosed reviews or undisclosed ‘advertorials’ are a breach of the Unfair Trading act in the UK.