Google Rewrites Quality Guidelines

[netmeg]

I'm going to post this link here, because it's from a credible source (WebmasterWorld user jensense) and because it touches on some of the things we've been discussion lately - particularly the Knowledge Graph. Interesting, and worth a read.

[thesempost.com...]

Here's the analysis on "Supplementary Content"

[thesempost.com...]

[edited by: brotherhood_of_LAN at 2:35 pm (utc) on Jul 11, 2014]
[edit reason] Added extra link [/edit]

[n0tSEO]

That's right, Selen!

Also, many academic pages with amazingly helpful content are nothing but text with some basic HTML formatting. Yet, they provide great value!

I really can't see a lot of wisdom in these new guidelines for quality raters.

[brotherhood of LAN]

They are useful, and relative to all the others serving similar content they may be the best quality. IMO, it's more of an edge case though. Smart people read those kind of pages and Webmasterworld :o)

The edu walls of text do tend to be collated in very logical orders too, they're pretty much a syllabus (which has a facet of usefulness in itself and the navigational structure describes it).

[netmeg]

Raters have never had the ability to alter a specific sites ranking. That comes from Google and from former Raters that I know.

[Shepherd]

Are there any indications as to whether or not the raters have the power to alter the position of the reviewed site in the SERPs?

+1 Netmeg, raters have no direct control over the serps. I see google using the raters to evaluate the algo in this manner: algo came up with 10 search results. 10 raters looked at those 10 search result, each gave 9 of the sites listed a very high score and 1 of the sites a very low score. The algo gets a 9 out of 10 score for that set of search results. rinse and repeat.

This is why I say, IMO, this guideline and the raters are more about rating the algo then individual websites. google wants to get to the point where the algo puts up what a majority of people think are quality websites in the top ten results.

[Jenstar]

Quality raters do not impact the search results at all... And here is it straight from Google...

Raters’ judgments do not directly impact Google’s search result rankings. While a rater may give a particular URL a score, that score does not directly increase or decrease a given website’s ranking.

Instead these scores are used in aggregate to evaluate search quality and make decisions about changes.

From [static.googleusercontent.com...] (PDF, in preface)

[EditorialGuy]

Jenstar: Never let facts get in the way of a faith-based discussion. :-)

[superclown2]

Raters have never had the ability to alter a specific sites ranking. That comes from Google and from former Raters that I know.

Thanks Netmeg, that's useful to know. So presumably they report back to Google who then wait a while to collate the information they will get from this edition of the guidelines and then make relevant changes to the algo? This would suggest that although the quality pointers are not necessarily factored into the current algo, they may well be in a future one.

[netmeg]

Google pretty much is not interested in anything that doesn't scale, and by scale I mean in a ginormous way. One rater who got up on the wrong side of the bed that morning isn't going to affect your site in the slightest.

I expect Google is always collating information and crunching data to get a bigger picture than we will ever see. There's no waiting; it's gotta be a constantly moving target.

[JD_Toims]

One rater who got up on the wrong side of the bed that morning isn't going to affect your site in the slightest.

Except the site(s) will be included as part of the pattern Google tries to detect as "low quality" algorithmically.

Meaning: no, raters don't "directly impact rankings" as in, low rating == immediate drop in rankings [just like they say], but raters can [do] contribute to the patterns Google looks for algorithmically, which means even though raters may not have a direct/immediate effect on a given site's rankings, raters have a direct effect on the patterns Google tries to detect, so their ratings do impact rankings for sites.

A "low quality" rating for a site means the site "pattern" will very likely be part of the pattern Google tries to eliminate from the results algorithmically, so even though it's indirectly, raters and a rating for a site very likely *do* impact a site's rankings via the site's pattern(s) being included in an new or updated algo to detect a given level of "quality" algorithmically, whether the attempt to detect the "quality pattern" of a given site by a specific portion of the algo is low, medium, high, etc.

[EditorialGuy]

Except the site(s) will be included as part of the pattern Google tries to detect as "low quality" algorithmically.

One tiny, tiny drop in the bucket.

[JD_Toims]

And being "a drop in the bucket" is inconsequential to the point of not even being worth thinking about, or expressing empathy for since it *must* be the site owner who was wrong and had a bad business plan to lose Google traffic, because it's definitely *not* Google's raters or algorithm based on their raters that cause a drop in rankings, well, unless someday the drop mixed in with the rest of the bucket is ours, then it might be a bit different...

Saying raters have no effect on rankings is both true and false at the same time -- It's easy to "play it off as inconsequential/nothing" especially when the drops [site(s)] in the bucket aren't ours, because what Google states is technically true, even though a bit of thought and insight tell us their statements about the impact of raters are a bit misleading at best.

[edited by: JD_Toims at 3:02 am (utc) on Jul 13, 2014]

[aakk9999]

We do not know how it is done. It could be that a few raters are given the same page to analyse and that some sort of standard deviation is used to discard assessment that is at odds with other raters rating the same page.

One thing that I have noticed reading the latest guidelines is that there is much more emphasis on the website even though the page is evaluated.

[JD_Toims]

We do not know how it is done.

Except they all have the same guidelines, so a site "not conforming" to those could easily be rated the same way by many, even if the site has a great impact and following...

BTW: Where's the easy to find contact/email/address link here on WebmasterWorld that conforms with the guidelines? I see Home, Tools, Terms, Privacy, Report Problem, About, Library, Newsletter, etc. [they're all only in the footer] and there's not a "contact" link or a "customer service" link I can easily find -- Based on the guidelines we've been reading, how should I rate this site? Hopefully, thoughtful/insightful people will see my point in the preceding question.

[edited by: JD_Toims at 3:19 am (utc) on Jul 13, 2014]

[aakk9999]

I was commenting on one rater that got out of the wrong side of the bed influencing the results. I did have impression from one of Google's previous guidelines that there are more than one rater rating the same page.

One more observation from reading the guidelines: It says pages with no Main Content deserve the lowest rating. Many CMS systems produce "empty pages" when an URL with non-existing parameter value has been requested. Therefore a technical error may produce many "Lowest" pages on the domain, which all together may be weighing the site down.

[JD_Toims]

I understand, but I think saying "one rater doesn't/can't have an impact" is false -- Raters all have the same guidelines, and if one "gets up on the wrong side of the bed" and their rating is "low" or "low+" rather than the "more often medium" they aren't "way out of line" and their rating could definitely influence ratings to be lower than they would have been if "the norm" was "med" and they rated a site at "med+".

Generally [not you specifically aakk9999]: Think about averaging test scores or grading on a curve -- One score can definitely impact things when it's not "out of line" enough with the rest to be discarded, depending on the exact method used.

[lucy24]

[xkcd.com...]

Where's the easy to find contact/email/address link here

Does it have to be identified with the exact word "contact"? I mean "have to" by Google standards, which might be either stricter or looser than ordinary-common-sense standards. Similarly, does the "Privacy Policy" link have to use that exact term? Mine just says "Legal Stuff" :(

[micklearn]

Oh, I am sure there are some SEOs or friends/spouses of SEOs who are quality raters, but from what I understand, they try to avoid anyone with more than a passing knowledge of the industry.

Can you shed some more light on that (what you know)? I can't imagine that many jobless non-SEOs go out searching for a "quality rater job" on any given day.

<snip>

[edited by: goodroi at 1:03 pm (utc) on Jul 13, 2014]
[edit reason] Time for somebody to reread the forum rules [/edit]

[JD_Toims]

Does it have to be identified with the exact word "contact"?

We all have access to the guidelines, so you tell me -- Does this site here we're all frequenting have the "qualities" expected/dictated-to a Google-hired, non-seo, non-webmaster, "normal person" to be rated consistently as "highest quality", or not?

[edited by: JD_Toims at 5:05 am (utc) on Jul 13, 2014]

[Planet13]

@ lucy24:

Thanks for the laugh.

You (and the cartoon) are so right though.

My family goes to a particular house of worship on Sundays.

This house of worship also has a brunch out in the courtyard. People who want to support this house of worship bring their food and make it available for people to purchase. They then donate the money the make to support the house of worship.

So, there are like a gazillion reviews on yelp for the house of worship - specifically all for the food offered at the Sunday brunch.

No one who goes to the brunch actually goes and listens to the sermons. They couldn't tell you anything that goes on INSIDE. They just know about the food.

[lucy24]

Does this site here we're all frequenting have the "qualities" expected/dictated-to a Google-hired, non-seo, non-webmaster, "normal person" to be rated consistently as "highest quality", or not?

Well, it's a bit of a trick question innit, because normal people aren't expected to hang around here ... but google clearly loves the site. Which I guess comes around to Planet13's point about the house of worship.

:: wandering off to figure out what they consider a ymyl page, although I'm morally certain I haven't got any ;) ::

[lucy24]

:: returning after slogging through the first 40-or-so pages* ::

Sometimes it is impossible to figure out the purpose of the page despite your best effort to do so.

Well, there goes my whole site down the tubes.

I am more than a little puzzled by the way they keep citing a custom 404 page as an example of "highest quality". Haven't they been saying all these years that you shouldn't want your error documents to get indexed? Sure, it factors into overall site quality, but this is about pages.

That All About Explorers site really is funny, though :)


* When it says "Proprietary and Confidential" on every single page, it's kind of a hint that we aren't really supposed to be reading it. And I'm not entirely confident that an offline copy won't still find a way to disappear eventually.

[Martin Ice Web]

Do i get it right if i say, a quality checker has to review your site to make a "EAT" rating?

[lucy24]

I hope that isn't what it means, because my impression was more that they're spot-checking the algorithm. If human raters' results are measurably different from the computer-generated result, then they have to go mess with the algorithm again.

:: noting with great irritation that the Wayback Machine (recommended several times in the raters' guidelines) has taken to sending auto-referers, which is enough to get them auto-blocked from my site ::

[netmeg]

I am more than a little puzzled by the way they keep citing a custom 404 page as an example of "highest quality". Haven't they been saying all these years that you shouldn't want your error documents to get indexed? Sure, it factors into overall site quality, but this is about pages.

Because it's about user experience, not site quality.

The Google Raters can't possibly rate every site on the net. It *is* just a spot check to look for anomalies to improve the algorithm. At least in theory. All the other various factors that go into where your site shakes out are still there. And I'm sure some of our sites have never been rated and maybe never will.

Don't go overboard.

[EditorialGuy]

Let's not forget that we're talking about quality raters, not the Webspam team.

Let's remember, too, that the system we're discussing gives scores for high quality, not just for low quality. And it's a lot harder to find good pages that can be used as "seed pages" than it is to find crap in the cesspool.

If Google's use of human quality evaluators can help search engineers identify patterns that make it easier to rank good pages better than bad ones, that's a huge win in my book--not just for Google, but also for Google's users and for site owners who produce Shinola instead of #*$!.

[hasek747]

Quality raters do not impact the search results at all... And here is it straight from Google...

Jenstar: Never let facts get in the way of a faith-based discussion. :-)

"Do not directly impact" (google's words) is a far, far cry from "do not impact at all" (your words).

[Planet13]

"Do not directly impact" (google's words) is a far, far cry from "do not impact at all" (your words).

You are probably right.

- Algo team tells group of quality raters to look at a group of sites that rank well for a competitive phrase.

- Quality raters consistently find a couple of crappy pages in the group.

- Google algo team / spam team looks at those crappy pages to figure out why the algo is erroneously ranking them (and other low quality pages) highly.

- Google algo / spam team adjusts algos so those types of pages are no longer ranking as high anymore.

[EditorialGuy]

- Quality raters consistently find a couple of crappy pages in the group.

- Google algo team / spam team looks at those crappy pages to figure out why the algo is erroneously ranking them (and other low quality pages) highly.

It can work the other way, too: The algo team may well be looking for high-quality pages (which are much, much harder to find than crap pages are).

[Planet13]

"It can work the other way, too: The algo team may well be looking for high-quality pages (which are much, much harder to find than crap pages are)."

Yes. Good point.

[TapFam]

Thank you for posting, NetMeg!

I appreciate this article now, before I have too much time invested into my website as I've started over again. I'm glad I never got involved with some of the old techniques that won't work in the future, anyhow. What baffles me is, how many top websites have too many long titles and URLs, No descriptions, no H1 tags, and no ALT attributes yet they still rank high in the SERPs.

It will be interesting to me, to see if Google's Quality Raters start giving low scores to websites that don't have the proper tags. Many authority websites post that proper tags are needed yet they don't have them on their own websites and they still get traffic. Am I missing something? Even Google.com doesn't follow the recommended guidelines, their title is too short, has no heading tags, no ALT attribute and low content on its home page.

I guess when a website automatically get enough traffic, then the webmaster no longer has to worry about the details.