Correlation between PageRank and MozRank

[Dan-Petrovic]

I was always curious about this so we did a test and the results are as follows:

Over a sample of 1000 domains:

• PageRank average = 2.71
• MozRank average = 2.34

Quite close.

Does this render MozRank near-useless, especially since recent downplay of PageRank by Google? What is your favorite metric at SEOMoz? Mine is PA/DA combo.

[Dan-Petrovic]

I like that! Thanks Rand.

[randfish]

Bing's linkfromdomain: command might be useful just to get a sense of what the site links to generally. But, I'd probably only do that if you have a sense that you should be worried (e.g. there's weird sitewides or they run a spammy reciprocal links directory or sell links somewhere, etc).

[Dan-Petrovic]

OBL (from page) is not going to be a problem. OBL (from domain) could be - depending how large the site is. However to check each of 20,000 links we have in Google to see if it ranks would require some insane scraping power - not so keen on that one. Perhaps we can do it manually as we go or prioritise by most commonly used links. Ideal situation we have one unified metric which gives link builders what "PageRank" was meant to be years ago.

[randfish]

Linking Root Domains is pretty good as a raw metric. I actually did a whole whiteboard Friday on uses for link metrics specifically for link analysis - http://www.seomoz.org/blog/which-link-metrics-should-i-use-part-2-of-2-whiteboard-friday

Honestly, in addition to Moz metrics, I'd be looking at where and whether the site/page links out (if they link to crappy spots, it could be trouble) and whether the page ranks well for the keywords it targets (pages that don't rank in the top 5 for obvious title match KW searches might be trouble).

[Dan-Petrovic]

Thanks for the update Rand. The reason we did the test is because we have a very large, searchable link database which our 20 link builders regularly update and add to. Sorting by PageRank was not so great and lot of great links were missed so we decided to go with Moz metrics - primarily PA/DA.

What other signals can you recommend?

• Alexa (or some other traffic metric)
• IBL (to domain)
• IBL (to page)
• OBL (from page)
• Page topic (keywords?)

[randfish]

We've found this to be very close as well! I believe the correlation coefficient was around 0.85 last time we ran it.

In terms of the value of mozRank - our intent was always to mimic Google's PageRank algo, not to build a metric that best represents how a page/site might rank. You're totally spot on to say PA/DA are a good choice for that (and we have some tests running to make them even better in the near future).

BTW - I should mention here since it's relevant that Linkscape's estimated index update of July 11 is now moved back to July 20th due to some processing errors on the Amazon cluster (and some mistakes on our end, too). Sorry for the delay, but on the plus side, we do expect it to be quite a high quality index update (deeper crawls and better domain diversity, too).

[Hallam]

Hey there - I don't think it renders it useless. I often use mozrank and page rank in tandem. If a site has a good mozrank, is fairly new and toolbar page rank hasn't been updated for a while it can give you a good indication of whether its worth spending a little bit of time getting a link.

Combine this with knowledge of what kind of sites Google like and the sites backlinks itself - and bingo. Often I make these decisions quite quickly.