Google Updates and SERP Changes - September 2017

[martinibuster]

---

System: The following 4 messages were cut out of thread at: https://www.webmasterworld.com/google/4860963.htm [webmasterworld.com] by robert_charlton - 3:25 am on Sep 2, 2017 (PDT -8)

---

How do you compare your pages with those in the top 10?

User intent.

[RedBar]

I'm more than 50%, closer to 75%!

[Lake]

Looking at Moz's top-view metrics, Google's overall domain diversity is the lowest it has been this whole year as of yesterday.

[Shaddows]

We had a horrific morning at 50% down, but it's back to normal for the last hour

[Cralamarre]

No change on my site with traffic. This morning has been pretty much the same as yesterday, which was pretty much the same as last week and the week before. I was hoping for a bump in traffic yesterday now that the school year has officially started, but it didn't happen. Maybe today will be better. Surprisingly, my AdSense revenue has improved quite a bit, beginning last Wednesday or Thursday, even though my traffic has not changed.

[samwest]

Sunday & Monday were the best converting days in many months. Absolutely no change in traffic volume, just "buyer quality". Tuesday and it has reverted back to non buyers. In two days the usual weekly quota was exceeded, so the bet is that it will be "off mode" for the rest of the week to balance out. If so, there's more empirical evidence to back the TT theory.

[martinibuster]

I wrote well researched and content rich. I did not say SEO optimized.

My response still applies.

Google's algorithm does not care about how well-researched or how rich in content the a web page is. That is not a part of the algorithm. I am trying to help you understand the algorithm better so that there will be more clarity as to why some sites rank well and why some sites fail to rank.

I will restate what I wrote: A major focus of the algorithm is satisfying the intent(s) of a search query.

Good luck,
;)

Roger

[Cralamarre]

@martinibuster, if I'm understanding you correctly, you're saying that if someone types in "How to do this?", all Google cares about is finding an answer that says "Here's how to do that". Is that right? Everything else in the article may be of interest to the reader but it's of no interest to Google's algorithm.

[Shaddows]

if I'm understanding you correctly, you're saying that if someone types in "How to do this?", all Google cares about is finding an answer that says "Here's how to do that". Is that right? Everything else in the article may be of interest to the reader but it's of no interest to Google's algorithm.

Not quite.

Everything else in the article may be interesting but may not be of interest to that specific person at that specific time.

If Google has the aggregated data that says that most searchers are satisfied by the short, succinct answer on one site, they will definitely not return a long-form page on another site. It may be full of interesting facts, but requires more attention than Google thinks the how-to user has to spare.

[Shaddows]

Also of note is that Google will hedge their bets by returning various types of response in one SERP, to cover alternative possible intents. These will still be personalised, so my set of possible intents will be different to your set of possible intents, so we will have a different set of results.

Generally, Google does not exercise editorial judgement- it doesn't scale. The model you, they model the intent of people who make that search, and they use whatever technique to calculate what sites satisfy people like you searching for things like this.

[Cralamarre]

@Shaddows, interesting. That would certainly explain why I see so many short, 1 or 2 paragraph articles, and even articles listing everything in point form within <li> tags, doing so well in the SERPs. For me personally, I like to know as much as I can about a subject. But I know from my experience with others that I'm in a small minority. Most people just want the answer.

[RedBar]

Most people just want the answer.

Tis the Internet for dummies!

[Awarn]

Adwords do not lock you in those top positions. Adwords is a waste of money anymore. Those positions are locked to make make the results look somewhat normal. I know the one pays nothing in adwords but they don't move off the first page on the top keyword. Now I can compete very well on secondary words. How is structured data working for others in eccom? I know mine is right and Google only picks up a fraction (200 of 11,000). This is just flat out manipulation. Another strange thing on those first page locked sites is speed is terrible.

[Shaddows]

articles listing everything in point form

Many new media sites and a big proportion of click-bait advertising specialise in "listicles".

The fact is that many, many people like these, so no wonder Google likes them too. Check this out:
8 shocking reasons listicles are Gold Dust [webmasterworld.com]

Totally intentional- illustration only

[martinibuster]

you're saying that if someone types in "How to do this?", all Google cares about is finding an answer that says "Here's how to do that". Is that right?

Yes, because the user intent is clear.

However, this is all dependent on how much data there is to sift through. For example, I can search for how do I cook with X kind of cooking tool. If there aren't many people searching for that phrase, Google can still know that this is an informational query and return some forum posts, blog posts, etc. But at the bottom half of the SERPs you'll see commercial results. That's because there is less degree of certainty, the cause of which is not a whole lot of people making those queries.

Also take note that ranking factors can play little to no role in some of those queries. There are algorithms outside of the Core Algorithm that can decide to promote other pages to the top, regardless of their poor ranking factors score, based on their ability to best satisfy a query. These are the Modification Engine algorithms. I wrote a blog post about that a couple months ago. I call them quirky serps. SEOs constantly get hung up trying to diagnose competitors and often come to the wrong conclusion because they're still stuck in the 200+ ranking factors paradigm.

Ever see a web page ranking that has few links or maybe poor quality links ranking? Ever see a web page ranking that has very little information and even a lot of images? That's the modification engine working to promote a site that has poor ranking credentials.

EXAMPLE:
I saw one site earlier this year start to rank because it was a California based site. A little digging around and I discovered that most people making this particular query lived in guess where? If you guessed California you're right. My observation was that Google promoted this clean and non-spammy site that had poor ranking credentials because most people typing that commercial query were from California, making this site more relevant.

That site had little content to explain its high rank. It had very poor inbound links. You will never diagnose why this site ranked unless you look at it as I did, through the lens of user intent.

That's why I say that the role of the 200+ ranking factors in powering the ranking ability of a site are diminishing. There are things going on outside of the core algorithm related to user intent that is overruling links, heading tags, etcetera.

[martinibuster]

Google will hedge their bets by returning various types of response in one SERP, to cover alternative possible intents.

That's correct. But it's more than just hedging. Google knows the exact percentages or ratio of users who mean this, that and the other when they type a query. It knows the different intents for certain queries*. What you see in the SERPs is the exact order by user intent by searcher popularity.

So if you do a search for Jaguar, the top result where I'm at is Jaguar the automobile. That's because MOST people mean the auto when searching for jaguar. Then an informational result about the animal and so on.

I downloaded a research paper that mentioned Jaguar specifically and it listed the percentages of people who mean auto, animal and so on. When I compared those percentages to Google's SERP for that query it was an exact match.

*Google can know the intent of approximately 85% of search queries, as statements out of Google have implied that 15% of queries are unique and never seen before.

[NYCTech]

Does anyone know whether pagerank would continue to pass through a 301 redirect if the "Remove URLs" tool is subsequently used to remove the initial page? We have a massive duplicate content issue that is not resolving quickly and want to remove an entire directory, but we want the quality pages to continue ranking, and they're in that directory.

The plan is to publish a new subdomain with the same title as the directory. Pages in the directory that we think are important will be redirected to the subdomain (ie website/directory/Page123 would become directory.website/Page123). Once that's been processed, we want to use the Remove URLs tool to remove that whole directory, removing a lot of duplicate/thin/low quality content from the index (it's all noindexed now, but there's a ton of it, so it's taking forever).

Does this seem like it would work?

[seoskunk]

@martinibuster excellent stuff - keep going please - you make perfect sense.

Does this seem like it would work?

No sorry I think you're replacing one problem with another, the subdirectory will take an age to rank and have to attract links almost as a separate entity, It would help if you could explain what is causing duplicate content and how the following is not an option;
1. Using canonical tag
2. 404,410 the pages

[flatfile]

@NYCTech have you checked the "Misusing the Remove URLs tool" section of this page [support.google.com ]?. That might answer your question.

[peaceofmind]

I've seen a decrease of online visibility and organic visits on 1st Sept and and a jump back on 3rd Sept. Some strange things happening but don't have enough tools to see exactly what pages are affected.

[RedBar]

My decline in traffic continues into today and it is all the lack of USA visitors with yesterday being 56% of US average when Wednesday is usually one of my busiest days.

Now what are you doing Google, penalising non-US owned/hosted sites in the .com SERPs?

I am sick to death of their perpetual manipulation of results in favour of US sites/companies/their friends.

[Cralamarre]

This morning, I received an email from the Google Search Console Team telling me that I can now use their new "Index Status" report to easily see which pages on my site are indexed by Google. I clicked the big blue link that says "See which pages are on Google", and.... I get a page with a 403 error message telling me I don't have access to the page. Good job.

[NickMNS]

As of right now I am seeing the new version of search console!

[Peter_S]

I get a page with a 403 error message telling me I don't have access to the page.

May be you need to sign in first.

[Cralamarre]

I am signed in. I get the same 403 error when I try clicking the same link in my actual Search Console. Everything else works fine.

[NickMNS]

@Peter I was getting the same error. There is a link in the top left sidebar "Try the new Search Console", click on that (if it appears). If not you may need to wait a few minutes until your account updates.

[Cralamarre]

Thanks NickMNS. I'm not seeing the link for the new version yet, but I'll give it some time. On the plus side, my Google traffic seems to be up a little bit today.

[NickMNS]

I noticed that the link only appears in one of my accounts. So it seems that the role out is limited.

[Cralamarre]

There we go, I'm viewing the new Search Console now and everything is working.

[NickMNS]

I started a new thread to discuss the new GSC [webmasterworld.com...]

[Peter_S]

The vacations period seems to be finally finished. I was expecting to see some significant increase of traffic after Labor Day, but it remained quite. But today, my "human" traffic is up by a solid 30% from North America and Europe.