Google Updates and SERP Changes - December 2020

[jacobjack]

Though there is not change in SERPs, traffic is down due to holiday season. Looks like in December, it will go down more.

---

Previous Monthly Thread [webmasterworld.com]

[edited by: goodroi at 10:55 am (utc) on Dec 1, 2020]
[edit reason] New month, new thread [/edit]

[mzb44]

If this 'update' sticks it can't be long before the mainstream media cottons on to the fact that searching Google is a waste of time since all that will be produced is the same small number of sites

The mainstream media will absolutely love this because it's their sites that will be ranking for everything.

[MrPink]

I haven't seen any changes so far. I would be interested to see if anything changes this weekend. Some of the success I've seen has been due to the fact of work that was done before the algorithm update happened about two weeks earlier.

[westcoast]

Large entertainment/education site, 25 years old.

We got slammed hard (-25%) in the May update. Google for some reason has, year by year, for over 10 years now punished us even though our content and quality has only improved. I've always chalked it up to some bug that snowballs over time and has our site caught up in some "old domain" data hairball.

New update.... Early signs look decent right now... traffic is up around 15% so far this morning. Not nearly enough to make up for May, but it's partway there. Most of that gain appears to be coming in from desktop, rather than mobile. Also, UK traffic gain is outperforming US traffic gain. And, much of it appears to be long-tail... so pages that google had pretty much dropped from the index in May appear to be bringing in a bit of traffic again.

If this sticks... and that is a big, BIG if, then this would be only the second core update that has improved our traffic in the last 10 years.

Fingers crossed it can keep up on this trajectory... Part of me is bracing for the bugged filters to be applied....

[loliover]

Seems to me like Google is going all in with Machine Learning since May.

The problem about that is that while it may work really well when it works, when it screws up and for example does not understand user intent, it can lead to some really atrocious results.

So I guess all we can do is hope that Google will figure things out a few updates from now... so in about 4-8 years. Until then some SERPS will simply be much worse than before machine learning, but I guess since there is no competition Google can do that.

[mzb44]

I am seeing so far an around -20% decrease, which generally means it will settle around -30% or -40% over the course of the next few weeks.

Now, just a little bit of background:

- September 2019 core update resulted in +200% increase
- January 2020 core update was +100% increase
- May 2020 core update was -60% decrease.

Before the May update the site had around 150 referring domains and around 200 pages.

After the negative May update we seriously overhauled everything, which meant:

- substantial design and UX changes
- content restructuring, keeping only what's actually truly relevant and useful for visitors
- rewriting everything with experts in the niche and not freelancers
- hiring an editor to further proofread and improve readability and flow
- published several pieces of real genuine original research

The result:

- Now after 7 months the site has around 1,000 referring domains linking to it. That's like almost 900% increase.
- The best part: ALL of these are 100% real and organic. We did not build them. We did not even email people or contact them on Twitter or something. ALL these links just came naturally to the original research pieces we published.
- Linking sites include such as BusinessInsider, Venturebeat, Tomsguide, Wired, major tech industry blogs, etc.

Effects of the update:

- Potentially up to -40% decrease.

Wow. Just wow.

So, now, after all these obviously substantial changes (otherwise why would 900 new - and major! - sites voluntarily link to it) in comparison to how the site was before May (when it had a meagre 150 links + freelance writer content), according to Google the site is now actually much worse and deserves much less traffic than even after the May update.

Sure, it did not get a major boost. But the site now is supposed to be even worse than before? Why? How?

Just. Wow.

You guys have no idea how much work went into this.

I am unsure what else to say or think about this.

[MayankParmar]

I get backlinks regularly for the 'original' work I do, but no improvements at all. I wasn't hit by May and January update, but I was still expecting improvements (inclusion in Top Stories). Over the past 12 months, I have focused more on original reporting, which almost doubled the natural links I have. I have never done any link building manually (as far as I can recall).

I have links from 9K referring domains and my DA is 70 on Ahrefs. I have links from 80-98DA sites. And the site will turn five years old on December 5.

It's disappointing that Google is not taking any steps to reward original news coverage. Scammers buy expired domain and scrap content, and they appear in Top Stories. I have more links and better/original than scrappers, but I'm excluded. Big brain algorithm? :)

[christianz]

@mzb44 Do you only have articles or you also have lots of thin content URLs or user generated thin content? My only explanation is that Google is trying to (in addition to just being greedy and boosting their mainstream media friends) remove spam which is a real problem in age of AI generated content, and they regard anything that is not long winded word soup as spam. Database driven information sites probably have suffered a lot and keyword stuffed, bloated (or sometimes even spun) text pages gained, because, hey, they are quality and good to E.A.T...

I should probably add "2021" to every heading and nVidia AI generated fake headshot with fake LinkedIn profile as author below all of my pages. Bet you would double my traffic.

The worse these SERPs get the more webmasters will be forced into blackhat techniques.

[Chickensalami]

And that’s what’s scary. Google continues to say “ just focus on creating good content “ and when we do, we still get hit. When improving pages speeds and web vitals, hit. Improving EAT, hit. I feel we’re just getting the run around at this point. It’s extremely unsettling knowing that the business you built by the required standards doesn’t guarantee security.

[JorgeV]

Hello,

900% increase.

It's more like 566% increase, but it's not the point, ... I always believed that, too many changes, in a too short time, sends a negative signal, because it doesn't follow a natural path.

Something simple to look, how much people are coming from backlinks? If links are not driving traffic (people clicking on them), then it means these links are of no value or interest for humans. So if they are of no value for humans, then I suspect Google is not (no longer) valuating them either.

[mzb44]

@mzb44 Do you only have articles or you also have lots of thin content URLs or user generated thin content?

Only articles. And only articles that have a clear scope and are written with real visitors in mind and try to solve visitors' problems / answer their questions right away without fluff and without any "made for SEO X word count target" bs. Generally, the articles are between 800-4,000 words (length is determined by what is actually warranted to fully and concisely explain the topic at hand but at the same time not going on on tangents).

There are user comments, yes, but everything is very tightly moderated and only relevant and actually useful comments are accepted that genuinely add something to the discussion.

The fact that it got such a significant number of real and genuine links without building them (not even link outreach!) should speak for the quality of these recent improvements. Honestly I was genuinely surprised to see around +100 new referring domains every month, all truly natural and from legitimate and important sites.

But I suppose Google disagrees and in their opinion the site is now actually even lower quality than before.

[mzb44]

I always believed that, too many changes, in a too short time, sends a negative signal, because it doesn't follow a natural path.

Yeah, but except that we just made the site better in accordance with Google's recommendations.

All the links and stuff we had no control over. Those sites themselves decided that our site deserves to be covered and linked to due to the content it provides. It literally follows a natural path. We had no manual intervention in relation to the links.

Something simple to look, how much people are coming from backlinks?

I'd say a significant number.

When it was featured on sites like BusinessInsider, Venturebeat it got thousands of visitors from those stories and links. Even more than search traffic at that time.

It regularly gets now 15% of the traffic from those backlinks from those linking websites. Every day. I don't know if others would consider that to be a lot or not. I think it's definitely significant. We are talking about hundreds of link referral visits per day.

[ichthyous]

Well whatever this is seems to have bumped up traffic somewhat, 18% for USA and 24% for UK. All of my main landing pages that had been languishing for weeks prior to this are all much higher today. Could possibly be that links are being weighed again more heavily. Let's see if it lasts.

[ichthyous]

Is anyone else having huge spikes of traffic from China hitting their site? The visits are coming from IPs all over China and hitting different pages on my site all at the same time, not one from one IP address or location. The only way you can really tell that it's a scraping attempt or some other not legitimate traffic is that the visits all last a few seconds and are all Google Chrome 78.0 browser / Windows OS / 1920x1080 screen res. I have been seeing this over and over for maybe five to six weeks now. Today I blocked all visits from China to stop the surge. Anyone else seeing this activity?

[edited by: ichthyous at 6:47 pm (utc) on Dec 4, 2020]

[westcoast]

Fascinating watching the algorithm work through specific URLs...

One of our main keywords that had held the #2 to #4 rank for 15 years was thrown way back onto page 7 in the May update (Had to have been a bug -- nothing else makes any sense. Nothing had changed on this end to warrant a 70 position drop). Last night it was on page 6... this morning page 5.... a couple of hours it moved up to page 4. Now it's just gotten onto page 3...

I still maintain that there was a serious bug in the May core update which caused all sorts of irrational and unintended moves. It appears that perhaps that may have been resolved and it is in the process of repairing itself right now.

[ichthyous]

I would refrain from making definitive conclusions about this update yet...it could very well still be in progress, and it is rare that Google ever fully rolls back anything.

[MayankParmar]

Has anyone noticed a drop in crawl rate on December 2?

[RedBar]

USA traffic is within my normal range at the moment however my widget SERPs are looking more than bonkers ... for one important widget search I see 4, yes FOUR, Pinterest listings at 2, 4, 5 and 6 ... that is NUTs Google.

'Tis early days yet, fingers crossed!

[worker]

I'm running tests using search phrases that I use every month or so. I'm finding fake search results on the FIRST page of google search results where the links and the description look valid, but when clicked, they redirect to non-related sites (i.e. different URLs, different text, different sites, with NONE of the text in the search result appearing anywhere in the site code).

Basically, Google search results are being 'gamed' in a significant way. To manage to get fake search results on the first page of search results is unbelievable. Apparently, Google is unable to crawl through the link to the ultimate destination, or the scammers identify when it is Google and they change to trick it. I'm not sure how this is happening, but Google search results are clearly manipulated.

In the example I'm referencing above, I had 10 organic search results and 2 were fake. There was also the Pinterest issue on the same page, but I'm much more concerned at Google having 20% fake search results in this test than the other lower quality links on the same page.

[aristotle]

I'm finding fake search results on the FIRST page of google search results where the links and the description look valid, but when clicked, they redirect to non-related sites (i.e. different URLs, different text, different sites, with NONE of the text in the search result appearing anywhere in the site code).

Could be a hacked site.

[EditorialGuy]

Nothing obvious here in terms of update impact.

Our Google traffic was up 21 percent today compared to a week ago, and our overall traffic (in a year-over-year comparison) is better than it's been in months. Still, that trend started days before the update rolled out and may simply reflect an increase in search demand for our topics. We have a travel-planning site, and--as I've mentioned in earlier posts--news about COVID-19 vaccines may be igniting dreams about 2021 travel.

[TalkativeEditorial]

Wary of being too optimistic, but early signs are interesting. Things can be quite volatile for some of our tracked KWs, and a few tanked on Thursday, but they are now slowly stabilising.

Far more intriguing is that traffic is stable and consistent (and in some comparisons to the same period over the last few months) dramatically improved.

That said, we did switch to full AMP earlier in the week, so too early to draw any definite conclusions from either.

[mzb44]

A popular tech publisher says their traffic dropped dramatically after recent updates and December update is not helping:

Tech seems to have been butchered during this update.

I am tracking several tech niche sites that for the past 5-10 years were dominating but now all are bleeding and are taken over by the Techradars, PCMags, Comparitechs and all the other big name mainstream publishers. While the site in your example does have great metrics, it's not a "big-name" publisher like techradar and the others.

I have long suspected that Google is gradually pushing the YMYL specific parts of their algo/AI onto all general niches and searches too. They started with medical, finance etc. and now core update after core update they are gradually rolling it out globally on all types of searches.

Seems like it was tech's turn this time around and now you're mostly only seeing the aforementioned sites for most tech related searches - sites that in Google's view are probably the equivalent of gov and university sites in the medical field.

[edited by: mzb44 at 9:25 am (utc) on Dec 5, 2020]

[MayankParmar]

The top publishers that you mentioned are part of the same company. One company owns all the popular tech news sites and they all interlink. They also explicitly state "as our sister site reported" in their articles and link back to each other. The entire tech news industry is dominated by one or two big companies.

The winners are mostly top publishers and new (spam) site. Losers are independent old sites.

[sofie77]

Hello,

here we go again, 7 month later. This was my post from the 6th may 2020:

we lost in May 2018,
we lost in August 2018 (Med),
we recoverd half in March 2019 (core update),
we lost the recovery from March in September 2019 (core update)
in January 2020 we recovered to the same positions as march 2019 (core update)

May 2020: we lost everything again. At this moment we got hit harder then 2018.

This is the badest und most painfull update for our sites I ever saw and this during corona where earnings are -60% already.

NOW:

Good news. We recoverd 60-70% from the May 2020 Core Update. I hope this sticks. Its not a full recovery but better then nothing. It gives hope.

Its a smart game. You get 80% taken, get 40% back, you have 40% effective loss but the customer is happy with it.

[mzb44]

Congrats, sofie77. That's great news!

What changes and updates (if any) did you do after the May update?

[sofie77]

Congrats, sofie77. That's great news!

What changes and updates (if any) did you do after the May update?

thank you but I think its to early to be a little bit happy. This rollercoaster since 2018 takes my trust away.
Well, we did alot between 2018 and may 2020 and of course 20 years before 7 days a week... but we stopped working on 6th may 2020. So in the last 7 month we did the daily work and made NO new projects or big improvemts. Our websites are great and the members and visitors love them. This has to be enough (for us).

We made changes for the loggedin users. We focused a bit on user experience but I am 100% sure, this has nothing to do with this core upate. The small recovery (related to 2018) has todo with the impact of good links and other factors. Unless there are no serious bugs or crawling problems on the site, there is nothing you can do. Says google itself related to core updates.

Since May 2020 our live has changed a bit. The very good years (before 2018) are gone and the trend is completly visible. We dont sit anymore 10 hours a day infront of the PC. We try to find our peace in nature, with kids and dog. Everything else is a plus...but I don't want to make my mood dependent on google anymore. Well, since I heard about the new core update I fallen back into old patterns and checked keywords. I haven't done this for 7 months (almost ; )

If the core updated sticks this way for us, we have only a grace period until the next update.

[MayankParmar]

News + affiliate or product reviews + affiliate is a common strategy and if that's the reason for the downfall then the competitors should also suffer.

[mzb44]

It definitely was / is a common strategy. But perhaps Google doesn't want this anymore?

Also, this seems to be a gradual process. A lot of affiliate / made-for-display-ads sites were hit in May. More now in December. There will probably be more come next core update.

This seems to be the trend. Just because some were not hit this time around doesn't mean they won't be next.

[JeepersCreepers1]

Decrease -40% still work on. Website with own technology, not based on WP.

[edited by: JeepersCreepers1 at 1:44 pm (utc) on Dec 5, 2020]

[christianz]

Google wants to rank real and legitimate entities and organizations and sites that have a general purpose beyond just "let's make pages to add affiliate links / earn display ads revenue".

It would be great if this was true, but its not. They want to rank specific sites belonging to handful of large media companies. Just look at what they are doing in YouTube, which used to be fair playing field but now is dual-class cast system (corporate content floats to top).

AI is not smart enough to differentiate "sites with purpose beyond ads" and sites just made for ads.

In best case scenario Google is not as evil as I fear and all they are doing is trying to filter out spam, but clearly they are throwing the baby out with the bath water, judging by how many well aged, independent, quality sites have been hit.