Google September 2019 Core Update

[MayankParmar]

---

System: The following 22 messages were cut out of thread at: https://www.webmasterworld.com/google/4962430.htm [webmasterworld.com] by brett_tabke - 9:39 am on Sep 24, 2019 (cst -5)

---

Holy #*$!, official core update will be rolling out later today!

[twitter.com...]

[mosxu]

@glakes

You are correct if you do go 100 miles away and your phone is not tracked you see different results but only if your site has not been visited before from the device you searched.

So technically any new buyer only if has not visited your site before will eventually find you on the second page.

AI is picking winners here but also if you are on second page for fresh buyer searches what is the traffic you are getting and plenty of?

[glakes]

AI is picking winners here but also if you are on second page for fresh buyer searches what is the traffic you are getting and plenty of?

In my previous post I did not dive into the signals that Google publicly uses for smart bidding/targeting or the creepy signals they likely use and don't disclose (data obtained from Chrome browser history, Gmail, etc.) or how retargeting may play out in organics as well. That's alone is worthy of its own thread. My guess is the same technology/features Google uses for paid ads is also to some degree applied to organics as well. This may be why conversions from paid ads drop substantially during major updates when filtering rules may be off which allows ads to be displayed on very low performing queries.

I rarely visit the second page of any search result, whether it be Bing (my preferred search engine) or Google. Mostly tire kickers/info seekers are visiting page two or bots. They don't buy anything, though being on the top of page two has its benefits with far less clutter.

We'll see how this plays out today. It's very quiet on Amazon, and few orders are coming in. Could it be Google finally squashed some of the massive Amazon crowded SERPS in my industry? Still too early to tell IMO.

[mosxu]

@glakes

And yet not all the websites are subject to revenue personalisation. Some players have steady results/ads everywhere you search from.

Maybe our social scores are not high enough.

[StupidIntelligent]

DailyMail not only got a boost but increased its traffic double digits after this update. The myth of user experience being a ranking factor should only be believed by authentic retards.

[robert76]

I suspect for anyone who is a Howard Stern listener that it is the cause for DailyMail traffic increase.

[ichthyous]

I was away and not paying attention to the vagaries of Google...I see a big traffic bump from Sept. 7th or so and it has remained higher since then. I suspect regular back to school traffic pattern...if there is an update underway now I am not see increased traffic but my visibility is increasing rapidly in SEMrush while other top competitors are either rising sharply or dropping sharply...like some kind of rotation. I did absolutely nothing...in fact was on vaca for three weeks and didn't even look at Google, SEMrush or my website.

[Selen]

Those who have been negatively impacted by the March / June updates and experience recovery on this update - did you change anything substantial?

[MayankParmar]

Recovered completely from June diaster.

In June, July and August - I worked on:

EAT improvements (rewrite author bios, about us page).
Nuked thin pages (not all, but some).
Removed above header ad unit on mobile.

I lost all hope in mid-August and gave up late last month.

[ichthyous]

I did absolutely nothing...didn't even look at my site all of September, but it's zooming up. Let's face reality, Goole has us all on a merry-go-round now. Some months we will go up for no reason and other months down for no reason.

[HereWeGo123]

Beginning the evening of 24, we saw an uptick in many KWs across the board, those positions continued to sustain themselves, and now are too. From a SERP perspective, we see a significant improvement, the complete opposite of what happened on the morning of March 13, as I woke up to the news of a broad core update. We also see an uptick in SEMRush search visibility. However, that has not translated into significant traffic gains. Maybe it will later, maybe it won't. But traffic is subtly on the uptick, but nothing that would resemble a HUGE win from that perspective.

[edited by: HereWeGo123 at 3:41 pm (utc) on Sep 26, 2019]

[vphoner]

I am seeing a big drop in the last two days.....20-30%, and I already was hit by March and June updates. It just never ends. And I have put up more content inbetween. Very discouraging, as I never seem to make progress, only sliding down into the abyss. I guess quality content does not matter anymore. Is this algorithm still jumping around? Any chance of recovery in the next couple of days?

[skaterpunk]

Google has us all on a merry-go-round now. Some months we will go up for no reason and other months down for no reason.

For the people sharing positive gains or recovery…means others have lost. And, as said, the cycle continues without knowing when, where, and how.

EAT improvements (rewrite author bios, about us page).
Nuked thin pages (not all, but some).
Removed above header ad unit on mobile.

I wouldn't put any stock in these helping with the recent recovery. Pure coincidence.

[Selen]

For bets, I'd put money on removed ads (= decreased bounced rate).

About us / bios can be easily gamed or "rewritten"; they are probably ignored by Google now.

[edited by: Selen at 3:50 pm (utc) on Sep 26, 2019]

[cabsco]

@Selen Yep.

1. Nodex/Nofollow all category and search pages.

2. Completely remove AdSense and Analytics on the site.

3. Re-edited/updated deep rusted old articles with typo and bad formatting on the site.

4. Remove over 1000 old articles about discontinued products (specs/price sort of pages) that are no longer use or available in the market.

5. Add "about author" bio snippet at the bottom of all articles with a link to our "about us" page included.

6. Removed/updated broken internal and outgoing links on the site.

7. Used " Google URL Parameters" to remove indexed url parameters (over 900 navigation/search parameters) that I cannot "noindex" from search.

Its been work since my site was penalized in June. The recovery is slow and is no where near my pre-june traffic volume, but I'm glad am seeing some improvement.

[edited by: cabsco at 3:54 pm (utc) on Sep 26, 2019]

[Selen]

Yep, #2 seems strong to me (especially removing above the fold ads which decrease bouncing rate).

#1 - now Google says long-term noindex meta becomes like a 404 page, so it may be risky.

[edited by: Selen at 3:59 pm (utc) on Sep 26, 2019]

[vphoner]

Hi Cabsco, do you feel that noindex the category and search pages helps a lot?

Why remove analytics from site?

When you remove old articles, they may have valuable incoming links to them, which is why I am reluctant to remove them. How do you address that?

Does bio snippet at bottom of page help a lot?

[seomotionz]

@MayankParmar Congrats man! My rankings are also recovering but at a very slow rate..

[cabsco]

@vphoner A lot of these pages were indexed in search (search/category) and these pages contains little to no information (just product images with 50 character snippets). Google may regard them as thin pages that's why I noindex them.

I remove analytics and adsense to increase site speed. Moreover, adsense earned me nothing since June so why keep it if by removing it I can increase my users engagement and lessen pageload.

The old articles I removed (mainly specs/price pages) do not receive traffic anymore. They were stagnant. Most of them are also thin pages. I cross-checked to make sure that I do not remove the ones with tangible backlinks.

Adding bio/author snippet may help in the overall user experience depending on the site/niche, as Google briefly mentioned in August. But I also think a strong proving "about us" and "privacy policy" pages are also now very important since EAT was introduced.

[edited by: cabsco at 4:25 pm (utc) on Sep 26, 2019]

[whoa182]

@MayankParmar

Nice to hear! :)

I pretty much abandoned my site after June update (almost) because I had made a lot of improvements after getting hit the first time in August 18. I did also make some changes to bio, more contact information, and improved a few pages, but nothing major.

While I've not seen a complete recovery like you, many of the pages are that got thrown to page 10+ are now 2nd page.

I guess this is a good sign, even if it's not translating into a huge increase in traffic just yet. It makes me want to work on my site again... Or maybe I should have continued despite what happened. :/

[seomotionz]

@whoa182 same here! I hope when all the rolling is done, it is going to be better.

[HereWeGo123]

@vphoner - Part of our process of when we remove old, dead, thin articles, is we check if they got any backlinks using a service like ahrefs.com. If they do have links, then we would simply still unpublish that URL, but 301 it to the homepage OR to another relevant high-quality article. That way you can still salvage some ink equity.

If done correctly, and strategically, while weighing out many other aspects of your site, in theory, pruning can work miracles for your site. Too much of old, dead, thin content can weigh the rest of your site down to the point that even the good pages won't perform well in search at its max potential.

[StupidIntelligent]

This is a Google Test.

For a few highly commercial queries where CPC is very high $5 plus per click; in certain regions I'm getting only one ad on the top and 3 ads at the bottom.

Looks like Google has listened or like Bill Lambert said, could be because all the filters that reduce traffic would be down for the next few days.

[glakes]

Maybe our social scores are not high enough.

That's Google - keeping us guessing while they laugh to the bank. Can't be social scores though as my company has a very strong social presence as compared to our competitors. We also sell directly to Government agencies inside the USA and out, so good .gov backlinks and buying traffic from .gov IP addresses should provide some solid quality signals. But to Google it is not about quality as much as it is money. To a thirsty person walking in the desert, tap water satisfies thirst even though natural spring water would taste much better. Unfortunately Google's users know no difference unless they step out of the box Google has them in.

To some degree I think Google stress tests commercial websites. We have EAT then Google proceeds to STARVE us of converting traffic. The thing is we won't go away. Thankfully we still get orders by phone, fax and email which to a large degree are outside of Google's prying eyes.

[browndog]

I seem to have been hit after doing well from previous updates.

[HereWeGo123]

@ Mindbodygreen – What niche are you in?

[Travis304]

@HereWeGo123 - I can speak to pruning bad content quite well: I did the same, exact thing to one of my eComm sites. Lots of thin blog posts and poor performers. Back in June/July, I pruned *a lot* of content. It was a little scary hitting the "delete" button on about 35% of the site's content. But, I double-checked each page/piece of content had no backlinks, little to zero traffic in a 6-month+ time period, and no rankings (or bad rankings, e.g. not on page 1 but on pages 2, 3+) for its target keyphrase(s).

Within 60 days traffic has almost (or by now, perhaps more than) doubled. Posts that I felt should have performed better, have started to. Some new content instantly ranks near spot #1 now. Rankings for target keyphrases climbed up toward spots 5 and under. Of course, Google's volatility since July has also contributed to fluctuations. And this latest update is another story.

[Milchan]

I am eating my Bill Lambert words as we speak

I might have ask you to share some of those with me as it looks like I wont be able to afford to eat any more after this update! All sales have stopped dead. Just hoping it settles back in a day or two and I get some sales at least.

[Selen]

Just because there are no referrals to a page from Google doesn't mean there was no human traffic to that page.

[StupidIntelligent]

Bill Lambert has been outed. He seems to be just a Google contractor, who sits in the meetings done by the search team. More here [seroundtable.com...]

[StupidIntelligent]

@Milchan - Surely with hundreds of thousands of visits per day; you're sure to get at least a few sales. Are you saying a site that gets close to 5 million visits per day has its revenue stop dead. That's improbably but not impossible. Google be damned.

[samwest]

The only recovery I am getting is by pumping up my social media posts to site pages and doing some visitor retargeting.
Luckily that produced a handful of sales which will provide bread and water for another week.

[hodl2themoon]

The filters seem to be in place now, from 50k organic users per day to 100... nice one Googlito

[HereWeGo123]

What are all those “filters” everyone keeps talking about in terms of organic search ? I’m still not getting it. Thanks!

[hodl2themoon]

See, I was hit from the June update, 95% of traffic vanished, suddenly I get all my traffic one day before the September algo update began, today the traffic is gone again. Bill Lambert guy is onto something here...

[StupidIntelligent]

That goddamn Bill was right once again. He is the real deal, but he keeps pushing total noindex for Google.

[glakes]

What are all those “filters” everyone keeps talking about in terms of organic search ?

I think the discussion revolves around what Bill Lambert said on the SE Roundtable blog:

I explained, prior to an update you will see traffic & metrics like you are used to. This is because while the core algo is updated the various "filters" designed to take your traffic away are not live. While the update rolls you will see flux. Post update you will see traffic levels around where they were prior to the update BUT these will slowly drop away. We are still in filter drop mode. Geo targeting will be out for the next few days too (as it was last week with the test

Filters, by their nature, are designed to further refine the search results. The problem with what Bill Lambert said is as one page loses another one tends to gain. A geographic filter might be the exception. Personally I would not mind if Google filtered out low quality products, Amazon (40% of sellers are from China peddling low quality products) and low quality blogs that are designed for ad or Amazon commission revenue. I don't see filters as a bad thing as they can target and demote specific pages that meet a certain criteria more efficiently then if the process were done via algorithm. The problem is many webmasters feel their sites are being filtered out of the SERPS to favor big brands. In ecommerce, the biggest brand is Amazon yet they produce few items and are essentially an extremely large affiliate earning commissions from each sale. Regardless, according to Bill Lambert filters will be dropping the 26th or 27th or possibly the coming week. In other words, what he is saying is that the traffic people are seeing now is the peak and will be scaled back once the filters are put back in place. Whether or not Bill Lambert's comments are real is open to debate. There was a real Bill Lambert that was accurate in the past, though anyone can impersonate him on the SE Roundtable blog since he posts as an unregistered guest.

like I said, wait for the 26th/27th. In fact going by meetings today this will still be up in the air until next week.

[Milchan]

There was a real Bill Lambert that was accurate in the past, though anyone can impersonate him on the SE Roundtable blog since he posts as an unregistered guest.

that said though , he did predict this update and also there was a post from Barry from SERT that confirmed that the same Bill Lambert posted from the same IP etc when people where questioning if there was an impostor - not saying it never happens but the genuine one does seems to post often enough.
I know Barry watches these threads also , so maybe he can comment on if it seems to be the same user everytime or if anyone ever tries to impersonate - I would think that maybe if someone was doing that , at least quite a lot, then they would get called out by the real Bill Lambert anyway.

[HereWeGo123]

@glakes - thanks! My vote of course. Do you think (or anybody here) that this logic can be attributed to the fact that on certain days, whether it’s this update, or other updates., confirmed or otherwise, or on just regular days, when organic rankings appear stable or even better than usual, the traffic is still lower? Just curious.

Not to say that personalization doesn’t exist, But whatever happen to the mindset of just that sometimes certain keywords at certain times are simply not searched by users? Or some days you just don’t rank as highly as other days? Or some days when google is releasing something, there’s a lot of flux in many of your keywords, which results in less people naturally landing on your page?

Just very interesting. I get it when it comes to e-commerce using AdWords, or if you’re targeting subject matters related to particular locations.

[BoredMeteor]

I can always tell when they're doing an update by looking at my stats. Things will jump up for a day or so, then my daily stats look like a step ladder going back down, often lower than usual (those are the updates I remember most, anyway, you know how it goes). Never looks very organic. That's why I checked here today and ah, well, there it is.

I don't know who Bill Lambert is (I don't follow this stuff too often), but what he says about traffic returning to "normal" and then falling back down rings true to me. Not saying it is, though.

Worst thing is that rankings seemingly haven't changed at all, but everyone's already mentioned personalization and all that, so it's useless to bring up, I guess.

As others have mentioned, I can 100% say that all those featured snippets and "Also Asked" and other nonsense have had a VERY negative impact on my site ever since they were introduced, and that kind of organic rank suppression has become the new status quo. I've also seen plenty of scraped/rewritten content in featured snippets. Can't stand them.

[StupidIntelligent]

@boredmeteor - Has the traffic always been going down for you, or do you also see ladder up after updates?

[samwest]

Filter effect was clearly seen here too. After 4 days of zero, had four sales yesterday and clearly robust human traffic. Its right back to $#!t drip traffic and non cenverting again. This clear manipulation is sickening.

[BoredMeteor]

@StupidIntelligent - I've seen ladders up, yes. I had that happen last year. I don't always pay attention to my stats, though. If I see things go back up in a few days, I'll post an update.

It's always an ebb and flow over here. Sometimes it's a seasonal thing, but not always. Sometimes I'll have that "step ladder" down with an update and it'll stay there, and that's when I get frustrated. Like there's this perfect proportion of traffic that gets chipped away every day, until it reaches a new baseline. Hoping that's not the case this time.

It's also possible that these ebbs and flows aren't 100% related to Google updates. I'm open to that possibility, I guess. Search Console shows me (right now) that average rankings are exactly the same, but impressions have dropped steeply over the last day or so. Could just be coincidence, maybe people aren't searching much for my keywords this week. I really don't know.

But when there's nothing else to account for it (seasonal, viral posts, etc), I just assume it's Google shenanigans.

[browndog]

Since I started my site in 2002, I have never seen numbers as low as today. I have over 1,000 articles, and right now I am sitting on 5 visitors. When I search, it's brand, brand, brand. Mostly insurance companies who are ranking.

I don't know if I can do this, but I'd really appreciate what I post not turn up on SEO round table. That is a friendly request.

[darthman]

Sharing some of our results:

- We’ve a seen a ~40% drop in organic traffic across our city pages and content section. Despite the fact that our guides and articles are well-researched and optimised.
- Some of the service pages that dropped were replaced by other websites' content guides.
- Some of our high-ranking guides were replaced by forum discussions.

Can anyone targeting multiple locations share their experience?

[StupidIntelligent]

@browndog - What were your highest visitors per day since 2002, when Google was still in love with you.

[StupidIntelligent]

@darthman - Search twitter for John Mu, where he specifically mentions Google hating on multiple pages created for several cities.

That being said, it doesn't have much issues with sites like Groupon or Yelp.

[TeresaD]

Over on twitter SEO Roundtable there are now 2 Bill Lamberts arguing with each other!

Edited to add it is discus not twitter

[seroundtable.com ]

[darthman]

@StupidIntelligent - Good point - We've made each location page as unique as possible (e.g. original city data, local reviews, advice etc). And some high ranking locations (with a lot of authoritative links) have dropped too. Competitors who follow similar strategy haven't been affected as much as us.

[StupidIntelligent]

Apparently original Bill Lambert posted this today:
----------
Today many will see poor metrics, high bounce rates & low session data.

The filters are being reapplied.

Next week could be big, the data will be analyzed and if they think they can push it harder they will.

---------------

Michael Martinez had this to say about him:

-------------
..Maybe he works in a Google kitchen, or the mail room. Not everyone who works at Google works on the algorithm teams. He could be a Webmaster support specialist, an advertising rep, a trainer, a conference logistics tech, one of the IT guys, or anyone of hundreds or thousands of jobs that don't directly contribute to the crawling, indexing, and ranking systems.
-------------

[glakes]

he did predict this update and also there was a post from Barry from SERT that confirmed that the same Bill Lambert posted from the same IP etc when people where questioning if there was an impostor - not saying it never happens but the genuine one does seems to post often enough.

I did not see the post where Barry confirmed Bill was the same poster as before, but it's good Barry took the time to respond. It does look like some of his posts were deleted from the one post about the update rolling out. And it looks like there is a lot of Bill bashing going on in another. Which is why coming here to learn of the latest news is a far more pleasant experience.

@HereWeGo123

Do you think (or anybody here) that this logic can be attributed to the fact that on certain days, whether it’s this update, or other updates., confirmed or otherwise, or on just regular days, when organic rankings appear stable or even better than usual, the traffic is still lower?

Not to say that personalization doesn’t exist, But whatever happen to the mindset of just that sometimes certain keywords at certain times are simply not searched by users? Or some days you just don’t rank as highly as other days? Or some days when google is releasing something, there’s a lot of flux in many of your keywords, which results in less people naturally landing on your page?

Many of us operate in industries where search volume and conversions will vary by day. For example, the products my company manufactures and sells to consumers are often researched during the week and purchased on Saturday and Sunday. The reason for this is the average consumer needs to do some investigating to determine which of our products will work for them, and if they fail to do this then what they order will not work. On weekends B2B searches and conversions mostly come to a stop since most businesses are closed on weekends. Likewise, a website about movies will likely see increases in traffic on the weekends while ski resorts will likely see traffic increases after a snow event. These are all naturally occurring and expected events. When it comes to personalization, that's a different story since ranking can change by the minute.

Since we know little about personalization as it pertains to organic search results, we can look to Adwords as an example of the signals they use for paid ads. Those "contextual" signals are publicly posted and are:

Device
Physical location
Location intent
Weekday & time of day
Remarketing list
Ad characteristics
Interface language
Browser
Operating system
Demographics (Search and Display)
Actual search query (Search and Shopping)
Search Network partner (Search only)
Web placement (Display only)
Site behavior (Display only)
Product attributes (Shopping only)
Mobile app ratings (coming soon)
Price competitiveness (coming soon for Shopping)
Seasonality (coming soon for Shopping)

See: [support.google.com...] under automated bidding signals

Expand the signal categories on the page above, and Google will provide examples. Under the demographics signal, Google provides the following example:

For a toy retailer, bids may be adjusted if someone has been identified as likely being a parent and is more likely to convert on an ad promoting a new line of educational toys.

This is what Google posts, but there are many other signals that the Chrome browser and other data collection points can provide Google. For example, if a person visits Amazon then goes to Google to search for a product, should Google display Amazon pages in the SERPS at all or demote them? A user that came from Amazon to Google performing a product search likely has a high intent of buying and whichever ads and organic results are displayed are all competing for this buyer. If a different user came to Google from a webpage that describes how to build a picnic table, that user is less likely to buy a pre-assembled picnic table.

The machine learning Google uses for smart bidding in paid ads gives us some insight into how they may handle personalization for organics. Though the output may be different, Google has to profile every visitor in order to serve ads and that info can easily be carried over into what they display in organics. We can check our rank using a different IP address/browser in ten different locations across the USA, at the same time, and at least in half the queries our keyword will rank in different locations.