I’ve been meaning to write this post for a while now and I’m sure the title will grab the attention of any remaining PageRank lovers out there. This post isn’t intended to provoke a reaction from anyone (although I’d love to hear your opinions), this post is meant to save people time on worrying about something that you can’t accurately measure.
I want to slip in a quick caveat before people jump straight to the comments and shoot me down. I think that PageRank can be a useful metric to keep an eye on, I’ve been the same as everyone else and when talk of a PageRank update surface, I’ll go and check some of my sites. I’ll also use it sometimes when doing some top level analysis of a sites structure and architecture. But I certainly do not rely on it as a metric to influence my decisions for SEO campaigns and tactics. I can probably sum it up by saying the following –
I use PageRank as an indicator of what Google thinks of page, I don’t use it as the truth
The prompt to write this post came about when some of my team at Distilled were discussing running a test on the flow of PageRank. I love hearing them talk about testing stuff, I’ve done my own PageRank tests in the past. But I had to voice my opinion which was that any test that involves the use of PageRank as a key metric is ultimately flawed. Why?
Because you do not know what the PageRank of any given page is at the current (or when the test finishes) point in time
But what about this:
That is NOT the Google PageRank of the page you are looking at. It is a snapshot of what the Google PageRank was at the time of Google pushing the data out to their Toolbar. It has also been rounded off – real PageRank is in fact the output of a complex calculation that does not nicely equal a whole number.
Part of me is actually hoping that someone out there can prove me wrong and show me that PageRank can be reliably and accurately measured. I’d love to see an example.
PageRank is also a fluid metric, it changes constantly and the only people who see this are Google themselves – and probably Matt Cutt’s cat 🙂
Think about it this way. If, shortly after I publish this post, I get a link from the homepage of the BBC and a link from the homepage of NASA, what will my PageRank be? According to the Google Toolbar, it will probably be N/A or 0. But both of these pages are PR9. But whenever Google next crawl the links to this blog post, they could give it an immediate ranking boost by flowing the PageRank. But I’d have no idea what my PageRank actually is until the next Toolbar update which could be months away.
So when you cannot reliably measure a metric, you cannot use it as the basis for testing. Yes you can use it as an indicator, but in my opinion, you can’t use it to backup your claims on certain SEO tactics.
Something you CAN measure which is almost the same and more useful
Crawl stats. Matt Cutts confirmed that your PageRank can influence how often and how deeply Google crawl your site:
“There is also not a hard limit on our crawl. The best way to think about it is that the number of pages that we crawl is roughly proportional to your PageRank. So if you have a lot of incoming links on your root page, we’ll definitely crawl that. Then your root page may link to other pages, and those will get PageRank and we’ll crawl those as well. As you get deeper and deeper in your site, however, PageRank tends to decline.”
Why not use your own crawl stats as a measure of which pages Google see as most important? It makes sense that Google crawl more important pages more often because of the sheer links pointing at them. You can then take this data and cross reference it with the pages that are getting organic traffic from Google.
Even better, you can see which pages Google are NOT crawling as much and see why. Perhaps they are not linked to well internally, perhaps they have no external links. Either way, you can take reliable data from a test such as this and be pretty confident in the results and your chosen action.
I’d love to hear everyones opinions on this and if someone can prove they can measure PageRank accurately and reliably enough to make it the basis of an SEO test, please let me know!