Link building became a serious mission for most webmasters and online marketers as Google emerged as the dominant search engine. Why? Believe it or not, before Google, links were not all that important to rankings. But this was the very early days - which are long since gone - so no need to talk about history. Even in 2009, Semify was deep into the link building business. In 2010, we were talking to our SEO enhanced link building techniques as they asked for faster and faster rankings (who doesn't). So now, in 2011, everyone knows, and I mean everyone, that links to your website are critical for top rankings in the search engines. And we are seeing some big big changes from Google in the algo, making a blog post that re-visits the link building approaches for 2011 timely.

A reminder that links can never hurt you

First off, let's return to that time-tested motto that Links Can Never Hurt You. I wrote about this a long time ago here. You should watch that video from Rand - it's very informative. The key rule-of-thumb to keep in mind is this:

If a bad link could easily hurt rankings, then I could easily use it to sabotage my competitors.

This is as true today in 2011 as it was in 2007. A few bad links is not going to sink any website. Perhaps you think people would not stoop to this level. Think again. It's a competitive world out there and linking is not a legal issue (yet). Google knows this, and has accounted for it in the algo. However, in 2011 we have the Panda update to think about. This has introduced some new thinking into the mix as Google has gone after those people who have more than, cough...cough, a few bad links.

What does Panda mean for my link building strategy?

Panda means that if you were a scrapper who then monitized scrapped material using Adsense (or other advertising network) that your days of profitability are probably over. While the SEO purists have been proclaiming that Google solved this problem long ago, Matt Cutts confirmed in a recent interview that before 2011, they really had only solved the machine-generated garbage problem. This was shocking to me. You can read my full thoughts on Panda 2.2 here.

But many people are not scrapers, what about them? Well, they got hit too! If you are into really cheap link building, then you are probably spinning content to stretch your limited budget. High-quality, human-generated content as a great way to seed your links, but many can't afford to do it really well. Those guys use "spinners," which are computer programs that swap out words and sentences in an effort make the articles look unique from website to website. We experimented with spinning way back in the early days of Semify (when our budgets were really small too) and ultimately decided is was not the way to go.

Semify does not spin any content

We ultimately decided on this because we thought it was more effective. Google is just that good. They can see that you are using partially recycled material, and that diminishes the effectiveness. I don't believe it brings a penalty. But it can cause the link juice for a link in spun content to be dropped to zero. Using 100% unique content has the secondary benefit of "appearing more clean," though that was more Chad's concern than mine. As you know, I find the entire ethical debate around SEO fraught with contradiction and, therefore, difficult to put my arms around.

Here is a group interview I read that I thought was particularly interesting on the topics of site wide links, linking networks, and over-optimization. I liked the no-nonsense fashion in which the link builders addressed the pros and cons of each approach to link building. It was also interesting how little consensus there was - although not surprising. Here are some highlights:

  • Over Optimization: There is such a thing as over-optimization, especially as it relates to inbound anchor text ratios. Make sure there is a fair amount of natural junk in your anchor profile.
  • Link Networks: Some link networks are helpful, although many are not. While a link from a network is unlikely to hurt you, bad networks really won't help that much. What's critical is how the network is getting their link juice. If it's a closed-system that relies on other sites in the network for link juice, it is probably not that helpful.
  • Site wide links: These occur naturally, but not nearly at the levels of deep, single-page links. As a result, if your profile is overly weighted toward site-wides, some of these SEO's were uncomfortable. Others say that Google is just not that far along on this one.
  • Link spikes: This was probably the most ambiguous outcome of the interview, with the conclusion being that they are both GOOD AND BAD. Most SEOs understand there are naturally-occurring instances of link spikes due to news events or viral spread. However, all the SEOs in this interview were oriented toward a natural and gradual build-up in backlinks over a long period of time.

What not to do

All agree that buying over 1,000 links, using the same / very similar anchor text as a one-time shot to boost a website is a very bad idea, especially if the links are cheap and from bad neighborhoods. This type of activity really makes a website stand out from the natural link building profile, and not in a good way. Instead, you want to be on the slow-drip approach. Yes, every link counts and is helpful (at least somewhat). But I often talk about link building as an I.V. drip, slowly delivering the medicine to the patient in a very slow, consistent and deliberate fashion - slowly building the dosage as the patient can handle it. Based on our experiments, this breads the strongest website rankings in the end.