Mike Levin SEO

Future-proof your technology-skills with Linux, Python, vim & git... and me!

Checking PageRank

by Mike Levin SEO & Datamaster, 08/25/2005

Now that I’ve formally joined the ranks of professional SEO’s, I’m filling in any missing holes in my methodology. Over the years, paying attention to PageRank has fallen in and out of favor. And almost every tool I use, I have built. This allows the extreme level of control I require for hyper-competitive search engine optimization. After I run a report, for example, I want the data to reside in a real database that I can run SQL against and extract data unimaginable from canned analytics. So, when I realized that I had to track the PageRank of my pages, competitor’s pages, and pages linking to both my pages and competitor’s pages, you can imagine my surprise in what I encountered.

Google PageRank is a tough piece of data to grab. At first, I looked at the Google API naturally assuming it would be an afternoon’s work. Neither the AdWords or the search API had any mention of PageRank. So, I did a little Googling. I found Proogle, which when you order by relevance shows a similar bar as the Google Toolbar. I also found the PageRank Lookup tool in the SEOChat site (now, the same as Proogle). With further inspection, I even found a kooky script that read the graphic off of the Proogle page and did a binary compare against a lookup table. The lengths people will go! I found an encrypted script that people want payment for, and I found people talking about stored page requests for toolbarqueries.google.com in Temporary Internet Files on PCs, which might indeed work, but was labor intensive.

Nowhere did I find anyone speaking candidly and openly about a decent method of determining PageRank in a systematic fashion. I learned that SEOStudio is able to do this, and sounds like a great deal for the Enterprise Edition at $199. We will probably buy it for double-blind testing. For anyone who doesn’t consider him or herself a very technical control freak, I would recommended a commercial package such as that. But for the rest of us control freaks, stay tuned. I’m going to walk you through the procedure of systematically checking PageRank. It can be automated, but not without violating the Terms of Service. So, because I want this tutorial to be white hat, linked-to by the adoring SEO white hat public, I will show you how to do it only in a non-automated fashion.

This project will tie together a number of disparate technologies that are already sitting on your desktop, waiting to be used. I’ll walk you through fetching an XML feed that shows you your competitor’s referrers. Then, we’ll use XSLT to transform it into a surfable web page. Then we will set up a network monitoring tool that will snag information off of the network. Then, we will then manually visit each of the competitor’s referrers, simplified by the fact that you are just clicking on links on a web page. At this point, you could just look at the Google toolbar and take note of the PR of each site you visit, and the project may end there for many. But for the real geeks out there, the next step will be to use Regular Expression Matching (RegEx) to turn the captured network data into a list of URLs which lead to XML files that reveal PageRank. These XML file requests can be made through the search API. What a wonderful project!