In one of its first posts about the original Panda update, Google’s Official Blog stated, “This update is designed to reduce rankings for low-quality sites – sites which are low-value add for users, copy content from other websites or sites that are just not very useful.”
It goes on to talk about rewarding sites with quality content. Those two statements alone beg the question: how does Google judge quality?
Several months after Panda launched, Google provided a 23-point checklist with a bit of insight… questions they’ve been asking to determine which sites offer quality content/copy and which don’t.
Words like “trust,” “authority,” “value,” “share,” “expert” and “comfortable” appear numerous times on the list in relation to quality sites. On the other hand, we find repeated mentions of terms including “redundant,” “errors,” “mass-produced” and “excessive” when talking about low-quality sites.
Being a search engine, Google uses robots (bits of software) to travel around the ‘Net and gather information about billions of websites every day. While there is a human quality team at Google, they are only spot-checking sites – not evaluating every single one.
How Do You Train a Robot to Assess Quality?
By relying (at least partially) on quality signals from site visitors.
Wordtracker’s own Mark Nunney has done an excellent job with his Panda Update Survival Guide where (among other things) he points out several metrics Google may watch to determine the quality of a site’s content. Among these are:
- duplicate content in high quantities
- low amounts of original content
- page content not matching the Google snippet in the SERPs
- unnatural language (over-optimization) of copy/content (also a Penguin issue)
- boilerplate content
And these signals are communicated to Google in the form of:
- high bounce rates
- low time on page
- low time on site
- how soon (if ever) you return.
“It’s really hard to say for sure, but my feeling is that somehow Google is able to tell if a user gets to a page from search and then can’t easily find exactly what it was they were looking for (based on their Google search query) that it’s a negative signal to Google. They possibly can tell when a user clicks around a site a number of times and then comes back to the Google search results and tries another site. With the sheer number of sites using Google Analytics these days and Google’s Webmaster Tools, it seems that they can pretty easily gather this information.”
What else do you do, as a site visitor, when you find content that hits the nail on the head? Frequently, you might share it with others. This leads to social signals that Google could use to determine quality.
Video shares from YouTube, indicators from Facebook, Twitter, LinkedIn, Google+ and more show up in your Google Analytics account. This offers proof to Google that your content was well thought of and shared with others.
Take a look through your site’s copy and the content of your blog. Are they up to par? Measure your pages against the criteria in Google’s Panda checklist and Wordtracker’s Panda Survival Guide. Do you see any of the problems I mentioned above? Make a list of action steps you need to take in order to improve questionable areas.
The Panda update is a permanent part of Google for the foreseeable future. Taking some time now to fix issues can save you lots of headaches and rank reductions in the future.
Karon Thackston in SPN
How Your Visitors (Unknowingly) Send Quality Signals to Google
