Google Updates and SERP Changes - April 2021

[goodroi]

---

System: The following message was cut out of thread at: https://www.webmasterworld.com/google/5028704.htm [webmasterworld.com] by goodroi - 5:38 am on Apr 1, 2021 (utc -5)

---

With Google continuing to evolve its ranking factors and growing the influence of AI influence on the SERPs, the updates seem to be more frequent (due to so many moving parts) but less drastic (due to most parts only being a small cog in the Google machine).

[NickMNS]

One needs to take these metric with a grain of salt. Google may be making this a ranking factor but worth how much? If the https ranking factor or the mobile ranking factor are any measure, this is unlikely to make any noticeable difference.

My main website is some what special in that I have many pages (in the millions) and technically each page is the same as any other page. The data is populated by single data source and whether the page displays data for entity "A" or entity "Z" it really doesn't change anything in terms of delivery or speed. The other thing that makes my website special is that I don't have a single landing page or small set of landing pages. Instead traffic trickles in to many pages. As a result I get nice wide sample of pages. Google's "core web vitals report" doesn't rank all my pages the same as one would expect. Instead I get a roughly even distribution of "poor", "needs improvement" and "good" pages. What this tells me is that the metrics are not just a function of the webpage but also the network, and depending where, when and who is accessing your pages.

I must assume that the metrics are based on the typical few pages with many page views model. In which case the many page views would average out to a pretty accurate estimate. But in my case, the many pages with few pages I get to see first hand how the metric falls apart.

We will see how this plays out, by I'm expecting more of the same, a lot of hype with nothing to show.

[topaz]

@universenet that has been my experience on two sites. the first is responsive HTML with separate amphtml pages. the second is native amphtml. both run adsense and have 93% and 94% "good pages" in GSC.

[universenet]

@topaz
And do you use http/2 ?
Auto ads or manual?
How many ads per page?

[saladtosser]

I moved several sites to a new server that supports HTTP/2 and a while back and see no change in rankings for these sites, positive anyway, some were hit hard by DEC core update.

My sites are 100% AMP, tested removing AMP on one and had no change either way.

The technical changes I've made to the sites are in full.

1) Dedicated Server (super fast and expensive)
2) Dedicated IP address for my sites.
3) Mobile responsive
4) Minimised inline CSS/HTML
4) img srset, 20 image versions per image. (WEBP with JPEG fallback so 40 images per image in total)
5) JPEG/WEBP compression
6) Cloudflare Web Accelerate
7) All metrics between 96-100% on google chrome audit and pass CWV in green
8) SSL
9) HTTP/2
10) Correct schema
11) Every page on the site is grammatically perfect.
12) All pages are run through Grammarly plagiarism checker & Copyscape to ensure unique content.
13) social following (Facebook followers 28k and 200+ youtube) and 1000's of Facebook page likes (doesn't do anything for organics)
14) 301 to SSL, 301 www to non www
15) Broken link checked, no broken links anywhere.
16) Good internal linking structure in content and menus.
17) HTML5
18) Lighthouse Accessibility/Best Practices/SEO scores = 100, performance 96 (except on one AdSense sites because Adsense it crap, even AMP AdSense)

I'm a hand coder, so I've pushed technical SEO to its limits. HOWEVER, MY CWV doesn't pass on one of these sites but only because of AdSense.

From what I can tell, anything technical for SEO, grammar correct unique content, UX, speed and social signals are such lightweight signals your better off focusing your time on organic link acquisition once you have good content obviously, then if you have a lot of money and time to spend on the site improve these other things which = good UX, not from a ranking point of view (as I said before you will jump through hoops for practically nothing, but to improve UX to retain users on your sites longer).

Links still seem to trump everything technical/social I'm doing above combined x 3 IMO. Sites that are above me not implementing half or a quarter of the things have next to no social signals, a couple of them arent even mobile-friendly. But they have been around 20+ years so have acquired backlinks back in the day (before smartphones) they wouldn't have today with the current state of their sites.

This is just all my opinion and experience, maybe other people have had great results from doing these things!

[edited by: saladtosser at 9:08 am (utc) on Apr 27, 2021]

[christianz]

I'm a hand coder, so I've pushed technical SEO to its limits. HOWEVER, MY CWV doesn't pass on one of these sites but only because of AdSense.

One of my sites is static with no javascript at all and and everything inlined (inlined stylesheet) - basically AMP level of instant-ness. And, of course, it fails CWV because it is AdSense site. This has been the case ever since they introduced pagespeed insights and there is nothing we can do about it. Other ad networks are not much/any better in terms of damage done to pagespeed/CWV.

What I can't understand is why are these ad auctions/tracking/delivery happening on main thread? Most of my ads are rendered in such a way that they don't affect the layout of other DOM elements (interstitial, popup, sticky sidebar etc). There is zero reason why any code loading them should run on main thread and block interactivity. Yet that's how AdSense and other networks are built and there is nothing I can do.

But they have been around 20+ years so have acquired backlinks back in the day (before smart phones) they wouldn't have today with the current state of their sites.

I would value 20+ years of history and backlinks far far far higher than any technical SEO and mobile friendliness. Any good web developer can create unlimited amounts of perfectly optimized but low quality, low worth, low value-added sites. 20 year old sites that are still live and still supported by OG enthusiasts and experts - those are gems that should be prioritized. Not corporate media sites attacking every keyword with SEO-tailored garbage, not auto generated, spun, fake spam sites. And definitely not social media trash.

[saladtosser]

>>>One of my sites is static with no javascript at all and and everything inlined (inlined stylesheet) - basically AMP level of instant-ness. And, of course, it fails CWV because it is AdSense site. This has been the case ever since they introduced pagespeed insights and there is nothing we can do about it. Other ad networks are not much/any better in terms of damage done to pagespeed/CWV.<<<

You know the crazy things, lighthouse and page speed tell you to remove unused AMP scripts (/v0.js(cdn.ampproject.org) as they are slowing down the page and loading junk you don't need, but if you remove that JS you cant have AMP.

How anyone was ever going to convince clients to uptake AMP when googles own test tool flags it is a mystery, AMP was doomed from the start with googles own test flagging it as a problem and they could have easily resolved it with a line of code in the lighthouse, but n..... no, let's cripple our own initiative from the get-go.

>>>I would value 20+ years of history and backlinks far far far higher than any technical SEO and mobile friendliness. Any good web developer can create unlimited amounts of perfectly optimized but low quality, low worth, low value-added sites. 20 year old sites that are still live and still supported by OG enthusiasts and experts - those are gems that should be prioritized. Not corporate media sites attacking every keyword with SEO-tailored garbage, not auto generated, spun, fake spam sites.<<<

Agree in terms of as a base you have to have good content first and foremost, then work on the other stuff... even though Google doesn't understand it themselves apparently, I was just pointing out not to get hung up on the tech stuff expecting much of a difference, even if you check every box! Work on your content and make that perfect first, then basic UX to retain users if/once you get them, then to rank on SE's organic links, links, link, then come back to everything else techy if you feel like it, but from SEO point of view you don't need to go crazy and distract yourself from content creation and organic link building if SEO is your core goal. (of course, you should work on it for your users regardless of SEO, but don't expect much google recognition if that's your main motivation)

Still some sites above me aren't held up on good tech or even content, spelling mistakes and very sparse textual information in fonts visually impaired people cant read.

[edited by: saladtosser at 9:36 am (utc) on Apr 27, 2021]

[mzb44]

I'd still like to stress the fact that one should not overly obsess about page speed or the CWV update. I sincerely doubt you will see any changes at all once the update hits, unless your site is genuinely terrible from a technical standpoint.

[TalkativeEditorial]

Google Analytics having a little freakout for anybody else?

[renatovieira]

Yesterday and today high traffic. it seems that this latest update has finally benefited me. It is too early to celebrate. Adsense with high conversion, in other words, quality traffic.

Weather forecast, Global

[ichthyous]

@renatovieira It seems that your site and mine always rise and fall around the same time. I had a big surge yesterday and today is also up, but perhaps not as much. Yesterday a lot of the traffic increase was classified as "direct". USA still lagging behind UK, AU, and EU traffic, but a definite increase. My ranking also increased several days in a row, instead of the perpetual up/down. The problem is, I have still lost most of the high volume terms that I had in January, so recovering a few low volume terms doesn't make that much of a difference. Also, we'll see if it lasts.

[edited by: ichthyous at 3:37 pm (utc) on Apr 27, 2021]

[yollo03]

My impressions are up and so is my ranking, pretty much like those that said they are recovering. I am on the same boat. I am still nowhere to close to pre-December core update's mayhem but it is on the right path, hope this continues.

Regarding https insufficient coverage, I fixed it, it is not a bug. I have it in green now. Don't expect google to fix it but who knows.

I am not using adsense but those that do, if the issue with core web vitals is the js script, did you explore serverless scripting or cloudflare workers (it's the same, different terms by different CDN brands)? It's basically js scripts that load at the edge of your CDN. I have no idea if it will fix it, anyone tried it?

[RedBar]

Everyone has problem with mobile pages

No they do not!

Why are you having problems? ALL of us have had years to adapt / adopt / use this.

I was developing and testing html5 all through 2012-14 before converting all sites, I started 9 years ago.

Granted my biggest sites only run into a few thousand pages yet I did do it by myself.

If ones's million+ pages are that valuable to the world and you have teams of developers making loads of money, then get them converted, honestly, I'm fed-up with this whinging when you have KNOWN for years the realworld scenario.

[yollo03]

I score on average 94% for all mobile pages, on desktop it's 99%.

[renatovieira]

@ichthyous - You're right. Our graphs are very similar, and it is strange to be in totally different niches. Anyway, we can only wait :-|

[mzb44]

Indexing issues again.

Barry Schwartz pointed out that Garry Illyes commented recently that a page needs to pass some quality checks before being indexed and if you aren't being indexed you need to improve quality.

So far on the site in question everything was always indexed within minutes and the most recent pages we posted the last week-ish are in line with what has been posted before. All real real expert stuff, research by specialists etc.

So perhaps this latest "indexing quality check" stuff they did (if Illyes is referring to some recent change) is to absolutely no ones' surprise broken (like core updates, google news, discover) and is now not indexing good content either in some cases.

[universenet]

No they do not!

Why are you having problems? ALL of us have had years to adapt / adopt / use this.

RedBar
I did not see that I have problem with any pages
I said that many have problems with mobile pages
Did you see score of some top websites in the world?
Example BBC news?
ok, some people like reddit.com so you can see that too
Terrible score, terrible
Is easy check who has problem with mobile
and many has problem with desktop too
many top websites in the world

[lammert]

Having a low score in Google's test tools is not the same as having problems with a site. Reddit and BBC do perfectly well regarding traffic, SERPs and user experience. Don't fall for Google's tricks and chase their latest gadgets. Do what Reddit and BBC do: build sites for humans, not for bots.

The only thing Google said is that soon they will use core web vitals in their ranking. But they didn't say how, and didn't say how much. I am sure it won't be their primary ranking signal. And it could even be used only as a secondary signal when other signals already caused doubt on the quality of the pages.

[universenet]

lammert
what you said looking intereting and VERY logical
But why google want spend so many resources on Core Web vitals if that is not important?
We need be clear that reddit has much traffic because google algorythm give this
and in last time, what is very googd to know google has problem with news websites too (and not only google)
but all that websites depends very much on traffic from google
If internet will be sad place without reddit? I think no
We need be clear next too that google never losing
Google always has much possibility for place ads on any website , change traffic over nite,
and change internet
All other depends of google, is not many websites who has strong direkt traffic, example 80% or more
So, google can play with every website how want
So, why google should not decide that everyone should to be perfect in everything?
I think that ranking even of top websites will be changed soon ...
Why?
becuase do not depends of them, google depends only about advertisers and they paying to google more and more

[lammert]

Google has spent a lot of resources on failing projects in the past. Writing some scripts to measure site performance and usability is peanuts compared to what they invested in Google+ or Project Loon for example.

They are aware that a lot of sites have a bad user experience (not in the last place through their own asynchronous AdSense ads that cause the layout to hop around :)) and try to condense good and bad user experience in a few easy numbers which webmaster can use to make changes to their site. By dangling a carrot in front of them, they hope webmasters will improve their sites. But I don't see the SERPs totally disrupted by the core web vitals update.

The big brands will probably not be affected by the core web vitals at all. They have too much other strong signals to support their SERP positions.

[NickMNS]

@Lammert

asynchronous AdSense ads that cause the layout to hop around :))

Let's be fair, the hopping around is not really Adsense's fault. AdSense is not responsible for designing the page layout, that is webmaster's responsibility. Are you suggesting we revert back to synchronous javascript calls for the ads? I guess that will reduce the jumping around, by replacing it with a mostly blank screen.

But then again, with AdSense aggressively marketing flawed product's like "Auto Ads" that are supposed to use "AI" to optimize the ad placements within the page layout it is no wonder that the webmaster community feels this way. "AI", you gotta love it, like the time it placed an ad in the Google Search box. It certainly was an optimal location, I'm sure that ad got a lot of clicks.

Otherwise I agree, as I said in my previous post the inclusion of this metric wont be noticed in the rankings. They have likely already included the metrics into the ranking algorithm a while ago and fully tested it, all that is left now for Google to do is to sell the kool-aid.

[TalkativeEditorial]

Saw a slight Discover uptick yesterday - importantly, for the first time since whatever happened in early March, relevant (new) content surfaced. Will keep expectations in check, however, as this might be fleeting.

Has anyone else who has been nuked on the feed seen similar over the last 24 hours?

Edit: I hasten to add that my personal feed is still wildly unreliable and was showing as empty on a few occasions so far today.

[MayankParmar]

My crawl rate suddenly dropped from 20K a day to 1700 in the last two days. Is this normal? Discovery is fine, Refresh is dramatically down.

[edited by: MayankParmar at 9:27 am (utc) on Apr 28, 2021]

[mzb44]

My crawl rate suddenly dropped from 20K a day to 1700 in the last two days. Is this normal?

Same here. Might just be something general and not related to your site in particular.

[yollo03]

My crawl rate has been consistently low since December core update. It is slightly higher now but still low. Maybe they tightened the crawl budget.

[mzb44]

My crawl rate has been consistently low since December core update. It is slightly higher now but still low. Maybe they tightened the crawl budget.

Same here but then it started to pick up again around end of Feb and early March. - Is this what you see as well? Just wondering if this is just something general or if Google decided to love me again.

[yollo03]

The spikes I've seen were probably due to algorithm updates but overall it is still low. It was at the end of March and mid-April.

[TalkativeEditorial]

Had a dramatic drop on 21 April and 16 March - but seemingly 'normal' now.
The Discovery/refresh ratio has always been utterly lopsided so I don't ever pay attention to that

[gatormark]

@christianz

I would value 20+ years of history and backlinks far far far higher than any technical SEO and mobile friendliness. Any good web developer can create unlimited amounts of perfectly optimized but low quality, low worth, low value-added sites. 20 year old sites that are still live and still supported by OG enthusiasts and experts - those are gems that should be prioritized.

Preach!

[mzb44]

If you were wondering why Google is not in a hurry to fix all these issues. Why we're an extremely small minority who believe that G search is broken and why no one cares that mainstream sites are favoured unfairly.

"Google overall saw a 34% earnings increase year-over-year. Google was up 32% for its search advertising revenue category, year-over-year." - [seroundtable.com...]

All the things G did last year, core updates, mainstream site boosts, non-mainstream site demotions. etc. seem to have been a massive success. G is more successful than ever.

This should be a strong wakeup call for anyone who still hopes that the 2020 updates and changes will be reverted because they were "broken". Also to those who still believe Google is killing the golden goose by being hostile to non-priority non-mainstream publishers.

Expect them to triple down HARD on all of the above in 2021, as it seems what they did last year was amazingly successful. The next core update will probably be massive and many people on this - and other SEO forums - will be extremely disappointed when it hits.

If they want to hit the same numbers and revenue increase percentages in 2021, they absolutely must crack down again hard with some more massive updates. We're probably still riding the effects of the December update. But soon I believe they will have to release another one to boost H2 2021 numbers as well. Probably around June, so they can show growth early H2 2021. Expect it to hit even harder than the December one.

[ichthyous]

@mzb44 That is correct...as long as revenue increases for Google after these changes this is going to get worse. There is no limit to the content Google can shuffle, if most smaller sites vanish or their business goes under it doesn't matter at all. Unfortunately, paying Google to recover lost organic traffic through running ads does not work. The traffic is poor quality no matter how much you target it, and the cost is far too high to pay for a wide range of lost terms. What that means is that many businesses that flourished online for years have an expiration date stamped on them now.