#1
  1. Not An Expert
    SEO Chat Hero (2000 - 2499 posts)

    Join Date
    Jan 2015
    Posts
    2,172
    Rep Power
    2033

    Google's Full Search Quality Rating Guidelines


    Somewhat old news, and I'm sure many of you have heard this by now (I think ThomasHarvey might have mentioned it in a post somewhere!)...

    But Google released all 160 sweet pages of its Search Quality Raters guideline and handbook.

    The big, fat ol' handbook can be found in all of its .pdf glory here.

    What's getting the most press right now, I think, is the big "Part 3" of the guidelines which is titled "Needs Met Rating Guideline." Part 2 is also about a very popular topic: "Understanding Mobile User Needs."

    Important to note is that this guidebook is used (as I understand it) by Search quality evaluators. The intro reads:

    As a Search Quality evaluator, you will work on many different types of rating projects. The General Guidelines primarily cover Page Quality (PQ) rating and Needs Met (NM) rating; however, the concepts are also important for many other types of rating tasks.

    [...]

    0.2 Raters Must Represent the User

    It is very important for you to represent users in the locale you evaluate. You must be very familiar with the task language and location in order to represent the experience of users in your locale...
    When I have time, I will try to skim through these pages for interesting snippets. If anyone else is bold enough to join me, please post your own snippets for discussion too!

    Or post your general thoughts, criticisms, and comments. Just don't spam ;)

    I'll make this a sticky thread for now, since I think it's a big news item. It may be unstickied at a later date.

    Comments on this post

    • msaeedonline agrees : 5000
  2. #2
  3. My views are my own ;)
    SEO Chat Good Citizen (1000 - 1499 posts)

    Join Date
    Jun 2012
    Location
    Clifton Park, New York
    Posts
    1,217
    Rep Power
    1739
    Actually a very good move from Google this time: They made it public! I don't say this often but "Thanks, Google!"
    Everything will be ok in the end

    If it's not ok, it's not the end
  4. #3
  5. Contributing User
    SEO Chat Adventurer (500 - 999 posts)

    Join Date
    Apr 2011
    Posts
    970
    Rep Power
    1550

    Google Releases Search Quality Evaluation Guidelines


    Just got this and need to read it properly later:
    http://static.googleusercontent.com/...guidelines.pdf

    However some interesting and odd things here:
    Section 5.3 on page 27 - "For some topics .... popularity, user engagement and user reviews can be considered evidence of reputation"

    Page 107 on the search about [tom cruise] it makes it sound like someone at Google is looking at websites page by page and making decisions???? The page given in the example is not considered 'Needs Fully Met' because there are lots of other pages ... which seems weird?

    Big section on Needs Met ratings for porn!!! Wonder who defines that!
  6. #4
  7. Not An Expert
    SEO Chat Hero (2000 - 2499 posts)

    Join Date
    Jan 2015
    Posts
    2,172
    Rep Power
    2033
    Ha! I hadn't seen the section on porn yet. Sounds like Google might take a more rigorous approach than the US Supreme Court: "You know it when you see it."

    Must be a really weird office setting.

    "Hey Dave, what are you working on today?"
    *Dave sits at his computer, idly clicking through different porn sites*
    "Oh, you know...same old boring stuff."

    I was also interested in the section on reputation:

    Reputation research is important when giving Highest ratings. Very positive reputation is often based on prestigious awards or other recommendations from known experts or professional societies on the topic of the website. Wikipedia and other information sources can be a good starting point for reputation research.

    For some topics, such as humor or recipes, less formal expertise is OK. For these topics, popularity, user engagement, and user reviews can be considered evidence of reputation. For topics which need less formal expertise, websites can be considered to have a positive reputation if they are highly popular and well-loved for their topic or content type, and are focused on helping users.
    I wonder where SEO Chat and other forums sit along that line...did someone have to manually flip a switch to place us somewhere in the middle? I know we all hold ourselves to a high standard here...does Google hold us to that same standard?

    Not all forums are about formal expertise though, so maybe they individually evaluate that for each site? Can a site be considered less formal over time if the tone of the forum changes? What an incredible amount of work that must be to keep up with it all.

    -Also, I think I'm gonna merge this thread with a sticky we've got for this topic: Google's Full Search Quality Rating Guidelines
  8. #5
  9. I love SEOChat.com
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    Jan 2016
    Posts
    31
    Rep Power
    30
    Originally Posted by Doodled
    Just got this and need to read it properly later:
    http://static.googleusercontent.com/...guidelines.pdf

    However some interesting and odd things here:
    Section 5.3 on page 27 - "For some topics .... popularity, user engagement and user reviews can be considered evidence of reputation"

    Page 107 on the search about [tom cruise] it makes it sound like someone at Google is looking at websites page by page and making decisions???? The page given in the example is not considered 'Needs Fully Met' because there are lots of other pages ... which seems weird?

    Big section on Needs Met ratings for porn!!! Wonder who defines that!
    > However some interesting and odd things here:
    > Section 5.3 on page 27 - "For some topics .... popularity, user engagement and user reviews can be considered evidence of reputation"

    Google loves this most.. it can go bonkers when it sees developing around you on social pages..

    Use Engagement is the key IMO
  10. #6
  11. No Profile Picture
    Contributing User
    SEO Chat Discoverer (100 - 499 posts)

    Join Date
    Mar 2017
    Location
    Sydney, Australia
    Posts
    316
    Rep Power
    639
    Folk need to be clear that these important quality guidelines are only coincidentally relevant to G's ranking algorithm.

    You would not expect to find a business directory or shopping cart category listed in G's results if you took these quality guidelines to be algo ranking factors.

    Another non-ranking factor is "user engagement".

    "User engagement" as per click-through rates, pogo-sticking, likes on social media, etc. are not ranking signals according to G.

    G's given reasons:

    • We can't track it
    • We don't have the capability to track it
    • It is a poor quality signal for many search queries
    • It is too easy to scam
  12. #7
  13. No Profile Picture
    Newbie
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    Nov 2017
    Posts
    1
    Rep Power
    0
    Can you please elaborate how the user engagement can be a scam?
  14. #8
  15. No Profile Picture
    Moderator
    SEO Chat Good Citizen (1000 - 1499 posts)

    Join Date
    Sep 2016
    Location
    USA
    Posts
    1,413
    Rep Power
    2532
    Originally Posted by prabhash
    Can you please elaborate how the user engagement can be a scam?
    No one said that "user engagement is a scam".

    Originally Posted by JohnAimit
    G's given reasons:

    We can't track it
    We don't have the capability to track it
    It is a poor quality signal for many search queries
    It is too easy to scam
    It can be scammed.

    Just for instance, someone could possibly write a bot to fake user experience / engagement.

    Simple enough to do. If you put the site together, then you know which pages create the funnel. Then you create a bot, do the funnel clicks, Google records the clicks because you have Analytics installed. Google see's a 100% happy user that went to the checkout page.

    Scammed. Why, because your bot doesn't pay for anything, it just leaves, so Google is fooled.

    Good thing they don't use that for ranking huh ?
  16. #9
  17. No Profile Picture
    Contributing User
    SEO Chat Discoverer (100 - 499 posts)

    Join Date
    Mar 2017
    Location
    Sydney, Australia
    Posts
    316
    Rep Power
    639
    Hi Prabhash,
    I believe I raised the issue of potential "scamming" with engagement signals.

    1. The easiest scam to debunk is CTR manipulation.

    The scam involves selling click-throughs to supposedly increase a site's ranking for specific search queries.

    Here is the clicker recruiting offer from one CTR selling site.

    "Get Paid To Visit Websites
    Earn $0.05 to $0.15 per click
    Earn 10% referral commission"

    The same site claims their click buyers pay $0.15-$0.25 per click.

    If you hunt around in black-hat SEO forums you will find they've been discussing, buying and testing CTR services since 2014 or earlier.

    To cite a black-hat forum response involving a monthly single search query click-through budget...

    "Let me save your $47, CTR manipulation has more or less no impact on rankings"

    Why would anyone buy clicks from non-potential clients that don't work when they could at least put the same budget into Adwords or real SEO services that are targeted at real clients?


    2. Bounce rates & time on page


    Try these engagement signal issues:

    Is a high bounce rate a good or bad engagement signal?

    Is a low time on page a good or bad engagement signal?

    The only correct answer is it depends...

    If you search for a phone number, address, business hours, correct business name, product availability or price, etc. a high bounce rate and low time on page should indicate a useful page that answered the search query in one click.

    A few years ago G. implemented an extensive PR campaign to alert us to "micro moments" searches. If memory serves, 25% of mobile searches were said to be people wanting short, simple quick answers. Low bounce rates and high time on page would suggest poor satisfaction with landing page quality for these.

    If you have access to the Analytics reports of a shopping or catalogue site you can see huge variations in the same site for the same engagement metrics.

    Many searchers will have progressed through their purchasing process to the point where they search for a product or service category. When they land on the page that lists the individual products they will likely click through to one or more products/services. These types of pages tend to have a low time on page but and a low bounce rate.

    If a searcher has progressed to where the desired product name is known and they search for that, this searcher is likely to land on the product page. There is likely less need to click through to other site pages so these can result in high bounce rate and low time on page.

    Astute readers may ask, why would a product page with detailed content have a low time on page?

    This goes to another "engagement" signal problem. If site visitors do not visit 2 pages from the same site, Google Analytics CANNOT REPORT ANY TIME ON PAGE. So, every single page visitor generates "0" time on page.

    These issues do not stop many SEO companies playing loose with the terms improving "engagement signals" as if they were ranking signals.


    3. Pogo-sticking & Dwell Time Engagement Signals

    A SE user survey was commissioned early this year. Amongst other factors it reported that only 8% of SE users returned from a search query click to the search results.

    That makes the metric potentially unrepresentative, unreliable, misleading and as useful as a glass door on a dunny.

    I find it interesting that the guy who commissioned the SE user survey was perhaps the main proponent of the "engagement" theory. It seems to me he's been very quiet on the theory ever since...

    "Dwell time" as a metric would be killed by Analytic's inability to report time-on-page AND the fact that almost no one "dwelt", so that G could use it as a representative signal.
    Last edited by JohnAimit; Nov 27th, 2017 at 02:48 AM.
  18. #10
  19. No Profile Picture
    Newbie
    SEO Chat Explorer (0 - 99 posts)

    Join Date
    Nov 2017
    Posts
    5
    Rep Power
    0
    A sigh of relief. Finally Google shared something that may be helpful for us. Good share!

Similar Threads

  1. How am I violating Google quality guidelines
    By nixies in forum Search Engine Optimization
    Replies: 2
    Last Post: Jul 24th, 2014, 04:35 AM
  2. What to do if my pages violate Google's quality guidelines
    By imlearner in forum Google Optimization
    Replies: 10
    Last Post: Feb 28th, 2012, 11:04 PM
  3. Google`s Quality Rater Guidelines Leaked
    By SEO Chat in forum SEO Chat Articles
    Replies: 0
    Last Post: Jan 13th, 2009, 09:00 AM
  4. Question about duplicate content and Google Quality Guidelines
    By mwilson75 in forum Google Optimization
    Replies: 0
    Last Post: Nov 21st, 2005, 03:21 PM

IMN logo majestic logo threadwatch logo seochat tools logo