Google Updates and SERP Changes - August 2018

[yollo03]

System: The following 2 messages were cut out of thread at: https://www.webmasterworld.com/google/4909313.htm [webmasterworld.com] by robert_charlton - 3:47 am on Aug 1, 2018 (PDT -8)

-----

HERE WE GO! Get your sit belts on:

Google Search Algorithm August 1st Update Rolling Out Now; Might Be A Big One
Aug 1, 2018
https://www.seroundtable.com/google-search-algorithm-update-26141.html [seroundtable.com]




[edited by: Robert_Charlton at 11:57 am (utc) on Aug 1, 2018]

[aristotle]

I noticed that some people are saying that this Aug 1 update had a really big overall impact on the web. I'm not so sure.

Yes, it apparently did affect a lot of sites in a few sectors such as healthcare and finance. But most likely that's only a small percentage of all the sites on the web. For example, there was no noticeable effect on any of my six sites, which are in other niches. I just spent some time looking at search results in those niches, and still see the same sites as before in the top positions.

I've mentioned several times that google has been working on its algorithm for 20 years and should have reached a point where no more big overall changes are needed. Of course the web is evolving all the time and google will have to adjust its algorithm to keep up. The healthcare and finance sectors may have been flooded with a lot of new sites over the past year or two, many of them of mediocre quality and/or unproven trustworthiness. So google took action to try to weed out some of them.

This may have been a bigger update than others we've seen over the past few years, but people shouldn't be comparing it to Panda.

[EditorialGuy]

I've mentioned several times that google has been working on its algorithm for 20 years and should have reached a point where no more big overall changes are needed.

On the other hand, the transition to greater use of artificial intelligence may have an impact on what constitutes the "core algorithm."

Also (and this has more to do with small changes than big ones), machine learning is probably making things more fluid than they used to be. The monthly "Google Dance" was a batch process, but with machine learning, testing and tweaking can be occurring (with minimal human intervention) all the time.

[HammerDown]

In general, you shouldn't put nofollow tags on any internal links within your site.

Back in the days of PageRank sculpting I was obsessed with trying to divert "flow" by use of internal "nofollow."

Those days are long over. I did quite a bit of research and there is not one single reason you should use internal nofollow. If you see it in your source, get rid of it today. Even if you have a high number of "noindex" pages on your site, links to those pages should still be "follow." Google has stated this unequivocally.

[HammerDown]

I'm seeing a 20%+ increase in traffic since Friday. Health related (mostly) site.

Very interesting!

For anyone who has a site that was impacted for better or worse by the update, Len at telapost.com is asking to have read access to Analytics as he's looking at the data and promises a comprehensive writeup in about a week. In exchange he offers to give you a quick site audit.

[telapost.com...]

[yollo03]

Im in the finance section which was the most volatile. Dropping from position 30 to position 135. A huge knock, organic traffic is virtually gone. The odd part is that only me and another website got hit out of all the competitors ive been monitoring. Its going to take more time for me to study this.

Early analysis shows google targeted specific keywords. I am seeing some recovery but if one page was 130 yesterday and not its 105 today I cant call that recovery. I dont think there will be any recovery from this unless google make another update.

[HammerDown]

Not surprisingly, the data shows that Barry Schwartz was correct in dubbing this the Medic Update as health was the most significantly impacted niche.

"In this instance, Google has made drastic changes to the health industry's rankings. So much so, that on average sites moved four positions during the update period, which constituted a 72% increase when compared to the baseline data. Also similar to the position data showed above, the travel niche still, on average, only saw sites moving under three positions during the update (for a 37% increase in the average position metric when compared to the baseline data)." - [rankranger.com...]

[HammerDown]

I dont think there will be any recovery from this unless google make another update.

@yollo03 It appears Marie Haynes was correct when early on (I think the 4th) she said it "may be difficult to recover" a YMYL site that's been impacted. The only hope is to re-establish trust by building E-A-T.

[yollo03]

That or launch a new website, this is beyond repair in my opinion. I am going to make some drastic changes in my website. I will report if I see any significant improvements. I am aware of YMYL and EAT but those that rank above me dont. I find it hard to believe this is the only reason. Some of them are just pure trash.

Just a small question to everyone, I was mainly hit in desktops. Mobile also took a hit but not as big. Is it the same with your website?

[NickMNS]

@yollo03 no mobile and desktop appeared to be impacted equally in my case.

[HammerDown]

I am aware of YMYL and EAT but those that rank above me dont. I find it hard to believe this is the only reason. Some of them are just pure trash.

@yollo03 Google's E-A-T algorithm is apparently pretty complex. Have you looked at their backlink profile? Maybe it's strong?

I'm in a real holding pattern. I have tons of work I could do on content, but what's the sense as long as Google has decided my site is not E-A-T? I'm hoping to have something more concrete soon from the insiders I trust. Til then I'm tuning out all the hyperbole being posted in this thread such as "all the experts are wrong. This update is 100% about..." or "Google is lying and trying to screw the world."

[HammerDown]

I keep seeing tons of bulls#!t eBay links in health-related SERPs and see they're up over 20% since the core update which completely flies in the face of YMYL-related E-A-T. Makes no sense!

[Halaspike]

My organic traffic is up by 30%. Some days ago i deleted over a thousand very old posts that weren't getting any traffic & used a 410 gone error code so Google will remove the pages ASAP.

I don't know if it has anything to do with the increase in my organic traffic or it's just a coincidence.

Has anyone ever experienced growth by pruning thousands of old posts?

[jmorgan]

@Halaspike Yes, low quality pages decrease the overall quality rating of a domain.

[lostshootingstar]

"Overall quality" or "Site wide quality" is a bit of an enigma for me; I don't fully understand why it would have much weight. Ive always felt that each page should be evaluated in more of a vacuum.

Say a website has 100 pages about wigets. 98 of the pages are "thin", or otherwise low quality. Not terribly useful to anyone.

But two of the pages are objectively the highest quality, and most relevant information about wigets that exists on the internet.

Why, in theory, would Google demote the best results for a query because other pages on the site people don't care about are not as useful? It seems... spiteful? Cutting off their nose to spite their face.

For example, who cares if a website has 4000 thin user profile pages without much on them? If the pages aren't useful, then don't rank them. It seems quite simple.

What am I missing? Has it ever been confirmed by Google that a site wide "quality score" exists?

[MayankParmar]

"Yes, low quality pages decrease the overall quality rating of a domain".

The sub tag and category pages are supposed to do that.

[mosxu]

Desktop traffic has technically disappeared after this update, the mobile traffic is about local shopping and not mcommerce [apnews.com...]

Location is important on mobile nothing to do with quality content

[Sandy01]

Hello,

During this month i see what, my websites are not crawling even there Sitemap.xml file exist.
At the same time tried to submit fresh Sitemap.xml file but not reflecting. Please suggest any alternative.

Thanks.

[jmorgan]

@lostshootingstar I think it was to prevent authority sites like ebay, etc. dominating the SERPs with thin-content pages. You know, the stuff us lowly webmasters would complain about. :)

I know where you're coming from. But in a way, it does force us to think about what we will, and won't, allow Google to index.

[Martin Ice Web]

Last friday some new quality update started in our niche. Lets say all is upside down. Serps are a total mess.
winners are:
-brands with only little content
-brand pages that have only the keyword on the page and that do not cover the topic
-sites with not E.A.T.
-sites with little html code and very little content

Conversions from google are down by 90%.
Amazon sales are up by 70%. ( <-Thanx google)

To the last updates i posted that it couldn´t be much more worse but this one hit the ground.

"Yes, low quality pages decrease the overall quality rating of a domain".

The sub tag and category pages are supposed to do that.

I think i should depend on the vertical the site is in.

if you take a serach engine, then the serps pages are like kategory pages but they are most important for the SE site, even if they are full off links, scraped content and ads.

[yollo03]

Google is ranking now all pages as one. If you have thin content pages they will affect the overall ranking of the entire website. Anyone has the guts to de-index thin pages and see if the ranking improves? I dont.

[MrBlack]

UK widget price comparison site got hit big in the August 1st update. It appears they tweaked something today to finally finish it off. It is now dead. And of course it is a very good, completely accurate and well researched website that adds a lot of value. It's previous positions have been replaced with a load of old rubbish. I guess they just want everyone to use their inferior product price comparison service.

[JesterMagic]

@HammerDown

I did quite a bit of research and there is not one single reason you should use internal nofollow.

I can think of several. To prevent a search engine from visiting form pages like for new comments or forum posts. To prevent a search engine from visiting a link that performs a certain action like pressing a like or Thank you button.

[JesterMagic]

BTW any one notice "Google Fixes Video Carousels To Show Them Less Often" [seroundtable.com...]

I haven't really seen a difference in our niche all the ones that appeared in the March update are unfortunately still there.

[HammerDown]

Anyone has the guts to de-index thin pages and see if the ranking improves? I dont.

@yollo03 It's admittedly very difficult. The site I bought recently was in just that situation. The "iceberg" if you will. Glen Gabe (I think) first described it as an iceberg where the 10% above water that you see is what Google values and sends traffic to. You have an area near the surface that with work can be brought to surface and then you have almost 90% below the surface that's a total liability.

It took someone to convince me to do it but it was a great idea and I finally got the nerve and looked at Analytics and ended either taking the content from those pages and moving to other pages, then using a 301 to point to the new page and finally removing the old URL in GSC, or deleting the page altogether or simply moving its status back to Draft to work on at a later time.

I can't recommend it highly enough. Just don't forget to either 301 the URL to a new page or at the very least the homepage. Doing this is more critical than new content. Once completed you've pruned a bunch of dead limbs and now Google sees a much more tight, compact footprint and your PageRank is nice and contained.

[HammerDown]

I can think of several. To prevent a search engine from visiting form pages like for new comments or forum posts. To prevent a search engine from visiting a link that performs a certain action like pressing a like or Thank you button.

@JesterMagic The meta "nofollow" tag would accomplish none of those things. If you're using internal "nofollow" for that purpose you're hurting yourself. You need to do some basic research on what the tag is for in 2018. If you don't want to do the research, trust me when I tell you do NOT use internal "nofollow."

[HammerDown]

Just now looking at my GSC data for landing pages on August 12. Did any of you in YMYL who got significantly impacted on the core update see a very nice uptick in overall position on the 12th? I'm talking 25%.

[EditorialGuy]

@HammerDown: Why not just use "noindex" on those pages?

[HammerDown]

@HammerDown: Why not just use "noindex" on those pages?

I thought about that but for a few different reasons I chose not to just noindex them. First, I had internal links pointing to them which made no sense to send juice to them. Also, many had good content but the articles were sometimes 300-400 words and I had very similar articles on the topic elsewhere that were ranking so I merged the two and 301'ed the old URL to the new while removing the old in GSC.

However, I was only dealing with less than 100 articles. If I had 1,000 or even several hundred, right now I'd "noindex" them all, get googlebot crawling the site and then deal with them on a one-on-one basis.

One thing is for sure, I wouldn't spend one minute on new content until the rest of the "iceberg" was dealt with. Once you have a nice, tight, compact PR container, then work on the new or refreshing the existing.

There's a real great article on this somewhere I read.

[NickMNS]

Do you know why the tip of an iceberg sticks out of the water? It is due to the difference in density between the water and ice.
What's the point?
If you trim off the 90% of an iceberg and place the top 10% back in the water it will sink to the point where only 10% (of the 10%) will remain afloat.
And the point?...
IMO trimming content is risky. It should be obvious that not all content is "tip of the iceberg" quality, but not all "less than perfect" content needs or should be de-indexed. Granted, some very low quality content, if it exists can and probable should be removed. (One must then ask the question, why was it there in the first place? But that is another topic of discussion.)

What I'm saying is the content quality is not discrete, good, okay, bad. It varies from great to horrible and every website has content that is spread along that spectrum (except mine of course, my content is all great, LOL).The fundamental problem is that the determination of what constitutes great vs horrible is subjective and what you find great may not meet that same standards as Google. Just look at the SERPs, would you expect to see some of the stuff one see ranking above your site. So where does one draw the line?

Being too critical will cause good content to be removed from the index. The value and ranking signals provided will be lost. Think back to the iceberg analogy, the particular page may have been below water but it provided the material, the density to keep the tip afloat. At the opposite end, leaving too much poor quality content in the index could send signals that reflect poorly on the whole site (unlike the iceberg this will cause less of the tip to float).

The one safe solution, the one recommended by Google is instead of deleting or no-indexing content, improve the content. Re-write the content fix the issues. Obviously on a big site this is a lot of work. By fixing the existing content one maintains the same volume of but reduces its density thus allowing more of the tip to rise above the water.

As a side note, my first reaction with respect to the discussion about "How to Geek" that occurred earlier in this thread was that they had mentioned that previous to all their problems they had trimmed a lot of content and continued to see a gradual decline in rankings until this latest update were the gradual drop become sudden. My thought is that they probably trimmed a bunch of content that was likely not great but still provided value.

And remember, just because I sneezed this morning at the moment that it started to rain does not me that my sneeze caused the rain. (or maybe it did hmm.... I always thought I had special powers)

[HammerDown]

I wasn't talking about a radical pruning based on arbitrary opinion.

What I did was repurpose content on pages that had rec'd 0-5 organic visits over the last 120 days. Absolutely zero doubt that Google placed little or no value on the content and one look at most pages left little doubt why. If/when there was good, unique content I moved it to similar pages or put the status of the page back to Draft so I could potentially use it later.

My point is, if there's a site with thousands of pages of low-quality content getting little or no organic traffic, there's no harm in doing something about it. Status quo is pulling the entire iceberg down.

[HammerDown]

Without getting too off-topic here, I found the original article about pruning content here: [oncrawl.com...]

[NickMNS]

What I did was repurpose content on pages that had rec'd 0-5 organic visits over the last 120 days. Absolutely zero doubt that Google placed little or no value on the content and one look at most pages left little doubt why.

Respectfully your are making a big assumption. You are equating the content's ability to drive traffic to your site as the same as the content's ability to send ranking signals. No traffic simply means that there has been little search interest for the topic covered by the content. If tomorrow some news story breaks about that topic and there is renewed interest you may begin to see traffic. But traffic or no traffic, if the content is good and it is indexed it is most likely providing valuable ranking signals.

My site relies almost exclusively on long tail searches (those are search that occur infrequently, not search terms that a comprised of many words). 99% of my site sees no traffic for month if not years at a time. But I am unable to predict which page will receive traffic on any given day. If I followed your logic, I would delete 99% of my site and only keep the most frequently visited pages. This would be great, 100% of my site would be indexed, I could have a smaller server and on and on... But I would eliminate 90% of my landing pages and in the process decimate my traffic. My situation is not common but the underlying logic applies even to a typical blog like pattern. Traffic and ranking signals are two different things.

[NickMNS]

@HammerDown
The author of the article you link to above clearly has no understanding about the physics of icebergs:

Removing the bottom half of this iceberg will cause the rest to “rise up,” making more of it visible above the surface.

I would be cautious when taking advice from someone that is not astute enough to realize that their own analogy does not hold.

[koan]

So if you have a site with articles and a forum section (with varying degree of post quality, which is normal), the forum section will penalize the article section? It should be no-indexed or blocked with robots.txt?

I encountered similar problems back in the Panda days, but after no-indexing all the user profile pages (usually just a bunch of links to their posts), that seemed to do the trick with the next updates. However, there's no way for me to be 100% sure that's what solved it.

[JesterMagic]

@HammerDown nofollow works as I want it. I have the noindex tag on my forum editor page for new posts. The problem was Google was still visiting the page for every single forum topic because the topic id gets appended to the url. Sure Google did not indexing the forum editor page but they still visit the page for every new topic. Adding nofollow for the new posts url solved this problem. This way Google doesn't waste my resources for serving the content and I don't waste theirs having them visit a page that doesn't belong in the search results.

[HammerDown]

The author of the article you link to above clearly has no understanding about the physics of icebergs:

@NickMNS I don't care.

[HammerDown]

@HammerDown nofollow works as I want it. I have the noindex tag on my forum editor page for new posts. The problem was Google was still visiting the page for every single forum topic because the topic id gets appended to the url. Sure Google did not indexing the forum editor page but they still visit the page for every new topic. Adding nofollow for the new posts url solved this problem. This way Google doesn't waste my resources for serving the content and I don't waste theirs having them visit a page that doesn't belong in the search results.

The "nofollow" meta tag has nothing to do with Google "visiting" your page.

[NickMNS]

@HammerDown

@NickMNS I don't care.

That is the problem, you should care...

[Dimitri]

I hope I am not going to embarrass NickMNS, but I am agree with him.

[HammerDown]

Note to self: before accepting SEO advice, determine author's understanding of iceberg physics. Some guy on the internet called NickMNS says I should care. Another guy on the internet called Dimitri agrees with him.

[Dimitri]

This is not it. If you base a demonstration, on an argument which is wrong, then you can't assume that the demonstration is right. What's the point of taking the example of an iceberg, and make it sounds "oh look how smart example this is" , if at the end the iceberg "physics" is totally wrong ? It's juts making the demonstration stupid and to make the buzz. This is like saying, a plane if flying , so it means the metal is less heavy than the air...

[Halaspike]

I also agree with the iceberg theory, pruning thing.

[NickMNS]

The issue is a fundamental issue of quality. Given the ongoing discussion about the quality of posts. I feel this is very relevant, as relevant a signal as spelling week as weak when one means to use the word that describes a span of time and not strength. (week vs weak is a reference to a post in another thread [webmasterworld.com...]

Removing ice from the underside of an iceberg will make the tip of the iceberg smaller (volume of ice above water). There are only two ways to make the tip stick out more, one is to reduce the density of the ice that makes up the iceberg and two is to add more material to the underside of the iceberg. The second method is exactly the opposite of what the author of the post is suggesting.

An analogy is supposed to be used to frame a complex issue in simple and understandable terms, if the logic of the analogy is flawed how can you trust that the author fully grasps the more complex underlying concept that is being described.

A flawed analogy is a sign of poor quality, the analogy used in the article was flawed. Feel free to follow the author's recommendations if you feel it is the right thing to do.

[yollo03]

A page that has 1.6k words got hit after the update, was well positioned in google before. When I say hit I mean besides the ranking the number of keywords google actually allows for that page is less than 40, was over 1k before the update. I see no reason why not to cut content from it, its pretty useless now.

[bwnbwn]

yollo03 not to be a smart ass but an article with 1.6k words give me an internet break, and now since this was hit cut it. IMO should have been cut a looooong time ago. Problem with the new and coming group stuff it with all BS keywords picts, ads and what ever and expect not to get demoted, please give me a break. Your all looking for a quick buck the internet, algo and REAL PEOPLE feel differently.

Your money train just crashed.

[yollo03]

Not really. If I was relying on ads then yeah but I dont have any ads. This particular page was up for 4 years so this is not the type of pages you are talking about, also invalidates your claim that I was after a quick buck.

[jmorgan]

Can I ask what niche you are in @NickMNS?

[jmorgan]

I see no reason why not to cut content from it, its pretty useless now.

Are you basically admitting you have/had bad content?

Not to sound like a broken record, but I think you're spending too much time trying to figure out how to game Google, rather than helping the users who land on that page.

[BushyTop]

Anyone else seeing any movement again this morning. Not good for us again :(

[yollo03]

Its not bad content. Like I said, it served me well for 4 years. But google decided other sites were under rewarded so I got pushed down the chain. Cutting content wont help from that angle but if not getting keywords anymore then whats the point of keeping it until the next algorithm update?