New Google Rating & Webspam Document Found. What it tells us.

By September 7, 2012Google, Search Engines

Over a year ago now a documented turned up that intrigued the search marketing community, this document confirmed for the first time that Google did indeed employ people for ranking (or at least quality) factors, the document was basically a handbook for the quality team. Last night, Razvan Gavrilas from Cognitive SEO  announced that he’d discovered a new version of the Google rating document dated June 2012.

As a freelance seo consultant it’s important to understand Google mindset and these kind of little slip ups (let’s face it this shouldn’t have got into public hands), this kind of document is therefore really useful. Overall, the document is largely unchanged from last year however there are some notable amendments.

General Guidelines Version 3.27 – June 22, 2012

The document is focussed on a variety of issues which include identifying webspam, by looking at the basic elements such as keyword stuffing, redirects as well as some more qualitative elements such as authority and commercial intent.

Additionally there seems to be an emphasis on the quality of the content, with areas to review including content type, language, flags et.

On top of this there is a section that focuses on how the page relates to the search query, including the geo-location of the webpage (not where the page is physically located on a server) but rather if the page is geographically related to the search query) and also how he page matches the “user intent rating” (basically does the page give the user the result they are really looking for).

A lot of the above is very qualitative – which I guess Google needs to makes sure happens when they are dealing with a faceless algorithm.

The document is a 161 page document to be used by their human quality team, about 60 pages discusses “Rating Guidelines”, in this section the rater is talked through a number of fundemental concepts including:
  • understanding the search query
  • Understanding and classifying the user intent – what they call “Do-Know-Go”
  • Understanding the “rating scale”
  • Understanding people queries

How Google breaks down a web page?

One of the most interesting elements of the document is how it breaks down page into three different areas:

  • Main Content – This is what the page is really about, it could be a blog, news, a product, video, web tool, search box or login.
  • Supplementary Content – This is the sitewide content and additional navigational content. It could also include videos and banners pointing internally etc.
  • Advertisements Content – This is everything from the small AdSense panel at the bottom of a page through to sponsored areas such as sponsored links, sponsored results etc (I’m better it will also include guest blog posts as well – if there’s a lot of them).

When reviewing the pages based on the criteria above the user is reminded to base it on page quality – that means they are backing up their algos with user experience, which is a very interesting point compared to where we’ve been before.

How does Google work out if a page is high quality?

Search Engine Watch identifies what Google regards high quality content, clarifying it to be anything that is “very satisfying, useful or helpful for its purpose”, they then go on to list the different potential options in this content as:

  • Share objective, personal, or social information
  • Share an opinion
  • Entertain
  • Share pictures, videos, or other media
  • Sell a product or service
  • Post questions and answers
  • Provide file-sharing or downloads.

Looking at the above it’s all about the pages having a really obvious purpose, they’ve been crafted by experts in the field to produce some content that the searcher is going to find useful and possibly come back to again and again – to be regarded as high quality I expect the quality of this content is about being professionally created.

It’s interesting that content surrounding the main bulk of the content is so important as well, you’ve got to be sure the main content is the most important part of your page layout (ads etc shouldn’t overpower the page).

Perhaps most important is that reviewers will be going on to check out the reputation of the website by searching for other references to the site (this could be anything for links back to the site, awards and if they content creator is an authority in the subject).

And the most intriguing to me is that social media looks like it’s finally going to become more important in determining how trustworthy a website owner is – this is largely about blog comments so I’d suggest it would make sense to get yourself on Google Plus as this is going to be one of the first places they look.

How does Google Determine Content is Medium Quality?

Medium quality content is content that could be easily improved – it’s good, however there are lots of obvious areas that can be looked at in order to get some really quick wins.

How does Google Work Out if a Web Page is Low Quality?

Poor quality pages have factual errors, poor grammar and spelling and probably contain chunks of spun content. Also, the design is taken into account at this point – which I’ve never been convinced about because some of the best information websites out there have pretty poor designs and I wouldn’t like to see some of those sites drop out of the index.

Conclusion

It’s pretty clear that human reviewing is only getting more important in the way that Google ranks websites which has got to be a good – let’s just hope that everyone has upped their game in recent years! The key to ranking well is great content and great promotion, we can get caught up in these kind of documents but really great content and great promotions is all marketing has ever been about.

 

 

Leave a Reply