Welcome to Crawl or No Crawl - this is Carolyn Holzman and today is June 24 - Day Number 659 of the indexation research project.

The last two Crawl or No Crawl reports are really important to understand moving forward -

Topical Judgement
[ Ссылка ]

[ Ссылка ]

When I started this research Google was dealing with some broken systems - javascript rendering was out of commission for at 3 months in 2021. Getting new content indexed was broken late 2021 into 2022 - and since last Dec 5th - things shifted again where desktop primary crawler sites had a different new content indexation than smartphone ones.

The reason the last 2 reports are critical to understanding moving forward what this research might be revealing.

In them I mentioned that I think there is a judgement between the index and serving components of the indexation system. And it looks like a topical judgment -

There is data from this research to suggest that this may be how the Helpful Content System works.

If you’ve been hit by the HC system - it is likely that you’ve unknowingly introducing content that is off topic as far as Google’s algorithm is concerned.

It’s not likely that it’s because you used an AI to generate your content - all you’ve done is use a predictive language by determining which word should be next. And it’s possible if you’re writing about things that a lot of sites are writing and you’re not getting indexed - it’s likely you’ve hit on a syndication filter more than a penalty for using AI.

So what if topical precision is the measurable aspect to Helpful Content Update? On topic content is rewarded by indexation and serving/ranking - off topic content may get indexed but will not be served or ranked.

In this testing, once The topic is settled by a series of content within the topic - indexation and serving by its keywords is returned.

We’ve been saying for a long time that google ranks pages for quite some time - and on one level that remains true - but I’m seeing evidence that they may not serve content that is off topic to the content as GOOGLE understand it to be -

For example - business attorney and business litigation attorney are NOT the same topic - think about it - our brains make that connection - not the math.

So how does this work? I went back to the developer documents and found some interesting distinction.
[ Ссылка ]

The system generates a site-wide signal that we consider among many other signals for use in Google Search

This classifier process is entirely automated, using a machine-learning model. It works globally across all languages.

Periodically, we refine how the classifier detects unhelpful content. When we do this in a notable way, we share this as a "helpful content update" on our Google Search ranking updates page.

So a reasonable person can interpret that as the classifier is tuned for topical aka helpful content that clearly runs all the time - and when they tighten or loosen those tolerances that’s an update.

Between this and the introduction of the InspectionTool Crawlers - the cost savings - the limitations within those tools - they can gate the introduction of new content - and rescoring - I believe this is why we’re seeing so much volatility.

I’ve adjusted my statements on non indexed content - new data is providing evidence that barring any technical issues at Google, it is likely related to your content - but not in how people describe that content.

Not “quality” content as in HOW you wrote it - but if its off topic in relation to your other content.

I use AI all the time to generate test content and it is getting indexed as easily and quickly as human written content.

Likely this is going to impact everyone who is not what we might call a high-level SEO -

Not everyone working in SEO is that. This has huge impact for agencies that are working on SOPS that don’t take this into account because they have do not have someone in that little basket at the very top of the ship who they significantly more for than an seo implementer - they may be making the matter worse with all that new content package they sold their clients. Sorry.

In situations like there they will be at a loss to understand why pushing out blog content for their clients and not getting any of it indexed or they start to see content drop out that was ranking well previously - if that content contained enough signals for more than one topic - or worse getting links from off topic pages - it will feel like a house of cards.

The weather is going to get worse until we look at the serps and start listening to the people who spend hours a day looking at what’s going on.

Indexation just got more interesting folks.
[ Ссылка ]

Related Topics:
Helpful Content System

свернуть