Google Algorithm Updates - 2019 Ongoing Discussion

Joined
Jul 5, 2019
Messages
62
Likes
65
Degree
0
Being in one of the niches that @Ryuzaki mentioned, it looks as if there was something on 08 November. Difficult to analyse because of the impact of seasonal changes, but, certainly as far as my sites were concerned, the overall site trends reverted to their usual seasonal movement the next day.

I'll try and have a more in-depth poke around more long-tail kws and rankings for other sites when I have a moment...
 
Joined
Nov 18, 2014
Messages
193
Likes
141
Degree
1
I run around in a lot of the YMYL niches, and we're seeing lots of ups and downs, for large KWs, at that. It's a widespread update, for sure. I'm sure longtail is a part of it, but things that have ranked well in the top 3 for years are no longer there, and there's actually quite a bit of spammy looking results in several instances of medium volume words. An interesting update.
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
I'm down 30%+ in organic traffic, so this is a pretty big nasty update.

Still only one site affected.

I am not too worried as I don't think the SERPs are better now, so probably will get back to speed eventually.

Running PPC ads in the meantime to try to recoup some of the earnings. Maybe that was the whole idea with this update here close to Black Friday. Some guy calling himself "Bill Lambert" claimed that, don't know if he was the real Bill Lambert or a poser.

I'm going to go ahead and do a full content upgrade again, hopefully that helps. I'm also going to begin adding more news-like content with the hopes of building a larger social media following and a list.

I've already increased my site speed, but I'll get some specialist to look at it as well.

I'm going to pay for some ecommerce certifications and stuff to increase authority.

Besides that, I don't see what I could be doing differently.
 
Joined
Jul 5, 2019
Messages
62
Likes
65
Degree
0
Not really any noticeable difference in traffic or income so far. If I was going to point to anything to take a look (from what I can see looking at some long-tail) at then it might be traffic from countries which do not have your site language as their main one (such as, for example, English-language sites with lots of Scandinavian visitors) and possibly ranking transactional kw phrases on otherwise informational sites. (Think 'Cheap red widgets' on Wikipedia.)
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
Not really any noticeable difference in traffic or income so far. If I was going to point to anything to take a look (from what I can see looking at some long-tail) at then it might be traffic from countries which do not have your site language as their main one (such as, for example, English-language sites with lots of Scandinavian visitors) and possibly ranking transactional kw phrases on otherwise informational sites. (Think 'Cheap red widgets' on Wikipedia.)
Well now, that got me thinking (lightbulb moment).

I have a generic domain extension, which can be used in a few countries, but isn't a country specific domain extension.

Only the affected site doesn't have the specific country exenstions.

I hope that's not the case though, because I don't see how I can target it. Hreflang is only for alternate sites right?
 

andreint

BuSo Pro
Joined
Oct 20, 2014
Messages
152
Likes
189
Degree
1
Well something to take note of, Chrome and Chromium will soon have "Speed Badges".


Sauce: Moving towards a faster web

I have a feeling this update may have something to do with speed now...
I just checked, and a couple of sites that I own (that dropped 40% during this update) have 90+ scores on PSI and more than awesome TTFB on WPT, so I'm not sure. Confusing as fck
 

CCarter

If they cease to believe in u, do u even exist?
Staff member
BuSo Pro
Boot Camp
Digital Strategist
Joined
Sep 15, 2014
Messages
2,446
Likes
5,717
Degree
6
I just checked, and a couple of sites that I own (that dropped 40% during this update) have 90+ scores on PSI and more than awesome TTFB on WPT, so I'm not sure. Confusing as fck
I am starting to think it is what @ToffeeLa is talking about. Are you able to do a country comparison from before and after?
 
Joined
Oct 23, 2018
Messages
60
Likes
162
Degree
0
Had recouped from the -20% September update, and the Friday update reversed the gains. Now back at -20%, maybe even slightly more.

This update seems to have impacted A TON of medium-size bloggers. The Mediavine group on facebook has a post with 200+ comments on people who have been hit with 20-50% drops, mostly in the recipe and travel niches. Most of them seem to be in the 50k-500k PV per month range. A lot of them say that they've been replaced with junk results. That's subjective though, everyone thinks that their site is the best one.

I'm not sure how to interpret the results. I guess I'll just keep building links, adding content, updating old posts and hope that this turns around. Hopefully, after a few weeks, we'll see some plausible theories on what exactly this update did.

I noticed that in the search console's new tab "speed (experimental)" all of my pages are slow on mobile and moderate on desktop. On Google's page speed test I get 60-75 on mobile and 90-96 on desktop for my article pages, which I think is pretty good. Maybe I should spend some time improving the speed even more. I'm sure when the Chrome update finally rolls out, having "this page usually loads slowly" will be bad for UX.

Oh and I looked at location data, and nothing seems to be changed. Still get around 55-60% from the US, 3-7% from Canada, UK, and Australia, and the rest is peanuts.

One thing that I did do two weeks ago, was increased ads, and added autoplay slideshow videos with ads. I've been tracking the user signals, and nothing seems to have changed from this. I get even more pages/user now, and bounce rate/time on page hasn't changed. My RPM's jumped through the roof from this. If I was getting 15-20$ before, now I'm getting 32-37$ RPM's. Maybe this update has something to do with ad density? Not sure, too early to tell.
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
As far as I can tell one of my main competitors, who I track, also dropped.

This is an absolutely brutal update btw, definitely seems sitewide.

One page just yesterday finally got to 2nd position for a broad keyword, then today plummeted back to 8th, so what does that mean? The bot got around to indexing it and then decided, from new ON PAGE criteria, that it didn't like it?
 

Cash Builder

BuSo Pro
Joined
Jan 14, 2017
Messages
337
Likes
367
Degree
2
I’ll chime in with what I’ve noticed from this update. I’m one of the lucky ones, organic traffic to my site has increased by >45% after this update.

The increases started on the 8th November, and it is sitewide. Almost every page got an increase in ranking, some of them huge. There’s a page that I couldn’t budge past the 8th page since I started the site that has jumped up to 20th for its main keyword.

Longtail keywords and the main shorttails have both improved.

With regards to site speed, my site is pretty slow. I did some work to speed it up last month, but with Ezoic ads on the site it’s damn near impossible to get it any faster. Just checked my Search Console speed tab and 100% of my URLs are marked as slow on mobile and desktop. Most of my pages have an average FCP of 3.5 seconds, which is just above the recommended 3 seconds.

As for the country, my site is a .co.uk targeting the UK SERPs, as are my two main competitors. None of us has experienced any drops with this update. I checked my analytics and my traffic from other countries has remained the same. There are still lots of US sites doing well in the UK SERPs. Some of them even have prices with dollar signs in the titles and they don’t seem to have been affected. Both SERPs are English language so that could explain why there are no drops.

Not really sure yet what this update was about, I was thinking page speed but going by the comments in this thread I’m not so sure.
 

CCarter

If they cease to believe in u, do u even exist?
Staff member
BuSo Pro
Boot Camp
Digital Strategist
Joined
Sep 15, 2014
Messages
2,446
Likes
5,717
Degree
6
On Google's page speed test I get 60-75 on mobile and 90-96 on desktop for my article pages, which I think is pretty good.
No. You need mobile to be at 90 to even ranking in some very competitive niches. I’ve recently worked with a very large brand and they wanted to hit 90 on mobile which I thought was curious- then they showed this, lets just say, technology I’ve never seen in my life - and the sites in the top 3 all had 90+ for mobile pagespeed score.

Then recently I completed about 42 consulting packages and threw in pagespeed for comparison sake - any serious keyword being ranked for - the websites that were in the #1-3 position were coming in at 90+ for mobile, if not 90+ they had some crazy metrics for authority and sick backlinks they could get away with an 80+ score.

So as the easiest common denominator - you can control your mobile pagespeed and get it to 90+, that way at least you’ll have an advantage if Google determines everything else equal.

People in the past have criticized us for “obsessing over pagespeed”, but then I turn around and see all these sites in low digits for mobile and it like - what are you expecting??

Everyone is on mobile. I am literally writing this on my mobile. And there is a 90% chance you are reading it on mobile.
 

mikey3times

BuSo Pro
Joined
Aug 25, 2018
Messages
103
Likes
81
Degree
0
No. You need mobile to be at 90 to even ranking in some very competitive niches. I’ve recently worked with a very large brand and they wanted to hit 90 on mobile which I thought was curious- then they showed this, lets just say, technology I’ve never seen in my life - and the sites in the top 3 all had 90+ for mobile pagespeed score.
Is this even possible if you are running display ads? The networks kill us with requests and payload size.
 

Ryuzaki

女性以上のお金
Staff member
BuSo Pro
Digital Strategist
Joined
Sep 3, 2014
Messages
3,767
Likes
7,156
Degree
8
Is this even possible if you are running display ads? The networks kill us with requests and payload size.
Yes. You just need asynchronous ads, lazy loading images / iframes / anything you can, etc. All you have to do is prioritize the loading above the fold. That could mean inlining some CSS in the <head>, having multiple CSS files with prioritized loading orders, dealing with FOUT (flash of unstyled text), server pushing CSS and JS files, etc. It's all about getting the above the fold loaded. The rest doesn't matter with these scores.
 
Joined
Sep 3, 2015
Messages
376
Likes
173
Degree
1
I got a small bump from this update but my reported pagespeed is and always has been atrocious. 99% due to the ad units I run I think.

All my pages have this issue in GSC FID issue: longer than 300 ms (mobile)

I am at the limit of my technical expertise with regard to speeding up the site - I believe it is as fast as possible on a server level. Grunty af dedicated instance, Wordops, nginx, caches, etc. I think I would need someone to help with the tasks @Ryuzaki mentions above. Anyone have someone they can recommend?

Also, are the speedsters here using Cloudflare? Does it help? I have no use it for it other than potential speed improvements.
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
Anyone have someone they can recommend?
I have someone helping me now for $130 / hour, so I hope they know what they're doing. I know this guy has been talking speed optimization for close to 10 year so I think so. I'm looking forward to seeing what the effect will be of his changes.
 

andreint

BuSo Pro
Joined
Oct 20, 2014
Messages
152
Likes
189
Degree
1
I have someone helping me now for $130 / hour, so I hope they know what they're doing. I know this guy has been talking speed optimization for close to 10 year so I think so. I'm looking forward to seeing what the effect will be of his changes.
If you end up liking the results, please share the contact info via PM
 
Joined
Nov 18, 2016
Messages
69
Likes
60
Degree
0
I have someone helping me now for $130 / hour, so I hope they know what they're doing. I know this guy has been talking speed optimization for close to 10 year so I think so. I'm looking forward to seeing what the effect will be of his changes.
Hey bernard , please can you PM me his contact? Thanks
 
Joined
Oct 23, 2018
Messages
60
Likes
162
Degree
0
John Mueller gave us some insight into the 8. November update in the latest webmaster hangouts.

Link: https://www.searchenginejournal.com/john-mueller-november-2019-update/

To sum up: It wasn't BERT, it wasn't about user experience, and it wasn't about speed. Instead, he says that it's related to overall relevance. They've updated the SERPS to show better quality, more relevant, and more authoritative sites.

Not sure how to interpret this though. It sounded like it's about EAT, links, authority, and maybe content pruning. What are your thoughts?
 

CCarter

If they cease to believe in u, do u even exist?
Staff member
BuSo Pro
Boot Camp
Digital Strategist
Joined
Sep 15, 2014
Messages
2,446
Likes
5,717
Degree
6
Not sure how to interpret this though.
John's explanation is literally the definition of vague. I suspect the reason no rank tracker was able to detect it is because it has something to do with country specific origins.

Example, if a user is tracking "black BMW 5 series" in the USA (Google.com), and the top 10 SERPs change for that keyword the rank trackers all have their "weather" reports. The thing is since no tracker can monitor everything Google, it will only show weather reports for keywords in their system. However if a Canadian user was coming in to a website in the Top 10 of the USA websites by using "black BMW 5 series" - unless the tracker is tracking the Google.ca version, the tracker would not know.

So following this logic, if Google simply stopped sending Canadian traffic to USA website for the "black BMW 5 series" query, any only sends that traffic to Canadian websites - what you would see is "no changes" in the weather reports.

And I suspect that a lot of the queries are long-tail and therefore BERT related - meaning trackers wouldn't even be tracking those terms as well. Cause who in the right might is going to track "black BMW 5 series with M5 package"? No one (well there are the outliers).

So when users say "long-tails are not impacted" - How do you know? Last I checked there is absolutely no way Google tells you all the keywords you come up for within Analytics nor GSC (in GSC you are limited to the first 1000 queries).

So my running theory: Google targeted BERT related terms, long-tail, in this update that were being sent to different regions versus the region of the user that's querying.
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
So my running theory: Google targeted BERT related terms, long-tail, in this update that were being sent to different regions versus the region of the user that's querying.
That would fit me, but I think speed could still be a part of it. Except I don't target non-native regions, I just have a non-native generic tld.

What can you even do to tell Google what region a page belongs to anymore? The href lang tag is meant for translated pages right? They removed the country targeting from GSC.
 

CCarter

If they cease to believe in u, do u even exist?
Staff member
BuSo Pro
Boot Camp
Digital Strategist
Joined
Sep 15, 2014
Messages
2,446
Likes
5,717
Degree
6
Except I don't target non-native regions, I just have a non-native generic tld.
I never use any of that SEO language/regional targeting stuff. I don't think it's about "targeting" non-native regional traffic, it's more that you were getting them since Google didn't have anything better to give in the user's native search results. But again this is still a running theory. I haven't done the deeper queries yet.
 
Joined
Dec 31, 2016
Messages
624
Likes
452
Degree
2
I never use any of that SEO language/regional targeting stuff. I don't think it's about "targeting" non-native regional traffic, it's more that you were getting them since Google didn't have anything better to give in the user's native search results. But again this is still a running theory. I haven't done the deeper queries yet.
I don't see how this should affect me though. I only have native language on my site, so it's only native. The only difference is the TLD is not native.