Link research tool data: the missing analysis
You see, there was a recent article by Rob Kerry one of the few true leading thinkers and doers of SEO in the world let alone the UK. I have first hand evidence of their work so I am in a privileged position to comment. The work in particular was comparing the diffrent link data tool vendors published on Aaron Wall’s SEO Book blog for:
- The amount of links found
- The percentage of those links being live pages
- The percentage of live pages actually containing the link
He also used the above metrics to compare the following tools:
- SEO Moz
- Ayima v2
- Majestic SEO
The winner was given to SEO Moz as Ayima v2 wasn’t public even though they had more links and more accurate data. Personally I’d much rather have AHrefs any day. You see machines, capital and technology are just a substitute for people. And organic search to me is a qualitative sport not a quantitative sport. I would trust me I’m an qualified accountant, but we all know what happened to Enron. And as Warren Buffet says in his Berkshire Hathaway papers is that finiancial statements are skewed because they give companies compensation in financial statements for making such huge purchases even though there is no cash compensation in the real world. What I’m trying to say here, is that I have no problem alowing staff to check links manually if it means we find more linkable material to work with for clients and brands. So for that reason AHref’s is the winner in my book.
The thing that I had more disagreement with was the need to exclude links that had no PR value. Again this is where Google must be laughing, because in my view, Google tends to operate a innocent until proven guilty policy when it comes to new content on the web. Sure they may sandbox new links and not transfer the full benefit but at least new content either on new sites or new pages on existing sites are given the benefit of the doubt, which is why I don’t care about buying PR.
I also couldn’t give two hoots about these PageRank emulating metrics like MozRank and AC Rank. Even if Ayima v2 is more granular as it simply means more detailed bad data. As mentioned in a previous post, they don’t know which sites have been naughty as Rob helpfully pointed out. What’s also worth mentioning is not all sites allow Moz Rank and Majestic to crawl their sites. Don’t worry I’m already developing plugins to distribute to webmasters to block SEO Moz’s and Majestic’s crawler IP address as they have no business crawling the web and then selling the information to gullible webmasters and SEO consultants.
Again I think the tools should focus on measuring Toolbar PR of sites improving over time and what the percentage increase is from one update to another. Perhaps the increase may appear as 0% for several updates in a row, but as long as the sites you’re getting links off are not losing PageRank then that is a reasonable sign that the benefits of those links are not being filtered in some way. Currently AHrefs are running a contest, so perhaps they may want to consider that.
What I also like about AHrefs is the fact they use Alexa rank as part of their metrics. Again, a good link research tool wants to be looking at the Alexa ranking over time of these sites. If the sites are getting more visitors, in theory that is a sign that the links are worth having and therefore cultivating relationships with. Which brings me to my next point.
Audience - probably the most important piece of analysis missing. Whilst I accept this may be hard to collect data on (as SEOs will probably overlook this in terms of collecting data in the first place), the missing statistic is what % of links contains the target audiences that will view the link. And, what are the target audience segments that within the audience traffic. Afterall, if we’re not trying to game Google, surely we should be measuring the value of links by the proportion of those reading or consuming the content that actually fit the profile we’re trying to reach out to.
There are more metrics for evaluating sites for getting links off which I will deal with in another post. However, as far as comparing link research data goes, the analysis conducted so far is far from complete and the only real winner to date of the candidates given thus far is AHrefs as there is a token atempt at including people which is where it’s all at - not robots.