Afraid the next Google update will kill your site's traffic? Already been hammered by one and trying to recover? Google unleashed a lot of updates this fall, and a lot of sites were negatively affected, especially those in the e-commerce and affiliate space. This talk will help you understand better how Google's machine-learning algorithms work. When Google rewards sites and when they "punish" sites by taking away their traffic. We will also look at how AI content might affect you going forward.
Understanding Google's machine learning algorithms will help you protect your site from the wrath of a Google update going forward as well as help you learn how to better grow your existing site traffic and revenue.
#ASW24
2. @schachin
Kristine Schachinger
Kristine Schachinger
• Started at a front-end dev & designer
Claim to Fame – Designed Reba McEntire’s site
• Started in SEO 2005
• Consultant 2009 – Present
• Some sites I have worked with:
GoodRx, Vice Media, Zappos, Instacart, Healthline, Jack in the Box, Discover,
USA.gov, Salon.com, Paychex,com, AndroidHeadlines.com, Patch Media etc
• Judge: US Search Awards, UK Search Awards, EU Search Awards
and since I said yes to all the Search Awards during the pandemic, there might be more.
• Specialties: Site Auditing, Site Recoveries, Technical SEO, and all the rest.
• Articles in: WIX SEO, Search Engine Journal, Marketing Land, Search Engine Land,
and Search Engine Watch -- among others.
• Speaker: BrightonSEO San Diego, iGaming, Affiliate Summit West, BarbadosSEO,
UngaggedUK/US, State of SearchLeeds, Pubcon, SMX, RIMC, SXSWi -- and others.
10. @schachin
Kristine Schachinger
Instead of strictly
creating algorithms to filter out SPAM & Usability Issues
now Google creates algorithms to
filter out low quality sites…
These are Quality Updates.
TIP! Quality Updates ALWAYS INCLUDE Technical SEO.
Update Summer.
Site Quality Algorithms.
13. @schachin
Kristine Schachinger
Medic is a misnomer.
Sites most hit fall under what is called “Your Money or Your Life” sites.
https://searchengineland.com/seos-show-mixed-results-following-google-march-2019-core-update-314600
15. @schachin
Kristine Schachinger
Takeaways.
But Google says it does not know what YMYL site
is in search, so how does it target them?
YMYL Sites
https://searchengineland.com/library/google/google-algorithm-updates
20. @schachin
Kristine Schachinger
Core Updates = Core Ranking Factors
Google is tweaking the
CORE RANKING FACTORS
to better surface the type of sites
as described in the Quality Raters Guide.
It is a QUALITY UPDATE.
21. @schachin
Kristine Schachinger
• Accuracy of Content
• Authorship
• Credentials of Authors
• Bios (unless a news site)
• Linking Out
• Links to Certifications
• Known Brands
• Reviews
What Medic or the CORE UPDATES are NOT.
22. @schachin
Kristine Schachinger
• Swear words in your comments
• Tone of your content
• Schema (while good not related to CUs)
What Else?
And ALSO are NOT.
30. @schachin
Kristine Schachinger
Core Updates = Core Ranking Factors
• Meaning of your Query
• Relevance of Webpages
• Quality of Content
• Usability of Web Pages
https://www.google.com/search/howsearchworks/algorithms/
32. @schachin
Kristine Schachinger
USUALLY - THE ISSUES WILL BE GLARING!
The issues also might have been on the site for YEARS
With NO NEGATIVE EFFECT!
FIX a CORE UPDATE?
35. @schachin
Kristine Schachinger
How to Fix Core Updates.
First
Make sure to CHECK YOUR QUERIES FIRST
Core Updates happen at the QUERY LEVEL
see where you lost IMPRESSIONS & CLICKS in Google Search Console
Don’t Use Third Party Tools.
36. @schachin
Kristine Schachinger
How to Fix Core Updates.
Next run a site crawl and find your technical issues
Make sure to fix TECHNICAL ISSUES FIRST*
because if Google cannot crawl and index your pages properly,
the REST DOES NOT MATTER.
*This includes internal linking, site architecture, & page speed.
37. @schachin
Kristine Schachinger
How to Fix Core Updates.
Next run a site crawl and find your technical issues
Make sure to fix TECHNICAL ISSUES FIRST*
Most important.
• Site Architecture
• Internal Linking
• In the past page speed, but with CWV may be
removed now, still make sure you fix it for PE.
38. @schachin
Kristine Schachinger
Takeaways.
Then Check & Improve
Next How to Fix Core Updates.
• Query Relevance & Content Quality
• Check Google Search Console for Query Drops
• Many sites drop on just a few CORE terms
• Are you pages RELEVANT to the entity searched
• Is your content meeting USER INTENT?
39. @schachin
Kristine Schachinger
Takeaways.
Important Note on Time to Recover!
• Originally you could not recover at all until the next update.
• NOW according to John Mueller, if they tweak ranking factors in
between major updates you may see SOME improvement.
• Still would not expect to see a full recovery without a new rollout.
How to Fix Core Updates.
43. @schachin
Kristine Schachinger
Google Helpful Content Update
“Our classifier for this update runs continuously, allowing it to monitor newly-launched sites and
existing ones. As it determines that the unhelpful content has not returned in the long-term, the
classification will no longer apply.
This classifier process is entirely automated, using a machine-learning model.”
https://developers.google.com/search/blog/2022/08/helpful-content-update
45. @schachin
Kristine Schachinger
Google Helpful Content Update
Main Points
• Ranking signal NOT an update
• First known ranking signal that has machine learning
• Continually rolling but with delays, so can take 2-3
months to catch-up with your site
• Sitewide but severity based on the number of issued
pages
• Other factors can lessen the devaluation (like
content quality on other pages)
• Seems to target what Panda and Penguin did with an
additional focus on the quality of “usefulness” or
“helpfulness”
• Is your content differentiating itself?
DALL-E image for “Angry SEO”
46. @schachin
Kristine Schachinger
Helpful Content + Page Experience
“Helpful content generally offers a good page
experience. That's why today, we've added a
section on page experience to our guidance on
creating helpful content and revised our help
page about page experience. We think this all will
help site owners consider page experience more
holistically as part of the content creation
process…”
https://developers.google.com/search/blog/2023/04/page-experience-in-search
HCU + Page Experience.
49. @schachin
Kristine Schachinger
Google
Myth: can’t detect AI content.
AI systems can predict that content is likely
created by AI.
How?
AI cannot create anything. It is only able to
use what is knows to detect patterns and then
in the case of content, use those patterns to
“write content”
So, AI can recognize patterns of how AI would
“write” and determine a likelihood that this
item is written by AI.
It is not 100%, but it can be done.
Google has an algorithm that detects AI
repurposed scraped content.
https://ai.google/static/documents/exploring-6-myths.pdf
50. @schachin
Kristine Schachinger
Google
Myth: can’t detect AI content.
AI systems can predict that content is likely
created by AI.
How?
AI cannot create anything. It is only able to use
what is knows to detect patterns and then in the
case of content, use those patterns to “write
content”
So, AI can recognize patterns of how AI would
“write” and determine a likelihood that this item
is written by AI.
It is not 100%, but it can be done.
Google has an algorithm that
detects AI repurposed scraped
content.
https://www.seroundtable.com/google-ai-plagiarized-content-34495.html
51. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
AI Content, Google, and the HCU.
Google says AI Content is okay IF it provides value and it not “spammy”,
But since it is writing what it trained on how does provide value?
52. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
How does Google define “Spammy” content?
AI Content, Google, and the HCU.
53. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
https://developers.google.com/search/blog/2022/08/helpful-content-update
Google and
the Helpful Content Update.
AI Content, Google, and the HCU.
54. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
https://developers.google.com/search/blog/2022/08/helpful-content-update
Don’t Believe Me – Believe Google!
55. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
https://developers.google.com/search/blog/2022/08/helpful-content-update
Don’t Believe Me – Believe Google!
56. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
https://developers.google.com/search/blog/2022/08/helpful-content-update
Don’t Believe Me – Believe Google!
57. @schachin
Kristine Schachinger
Some of the general parameters that are used to train language models include:
https://developers.google.com/search/blog/2022/08/helpful-content-update
Google and
AI.
Don’t Believe Me – Believe Google!
John Mueller On AI Images.
62. @schachin
Kristine Schachinger
Google
Myth: AI is approaching
human intelligence
“While AI systems are
nearing or outperforming
human beings at
increasingly complex tasks
like generating musical
melodies or playing the
game of Go, they remain
narrow and brittle, and lack
true agency or creativity.”
https://ai.google/static/documents/exploring-6-myths.pdf
63. @schachin
Kristine Schachinger
Google
THERE ARE THREE PLACES GOOGLE APPLIES MACHINE LEARNING
IN THE ORGANIC SEARCH ENGINE.
+ PRE-SCORING
LANGUAGE MODELS
+ AD HOC POST-SCORING
RANK BRAIN
NEURAL MATCHING
+ LIVE RANKING FACTORS
HELPFUL CONTENT UPDATE
THE BIG DADDIES! SGE and MUM ARE IN A CLASS BY ITSELF.
67. @schachin
Kristine Schachinger
Sesame Street and Search
What is BERT?
Natural Language Processing pre-training called Bidirectional
Encoder Representations from Transformers, or BERT.
Moving from NLU into early NLP
68. @schachin
Kristine Schachinger
Google
https://searchengineland.com/how-google-uses-artificial-intelligence-in-google-search-379746
BERT. ”BERT, Bidirectional Encoder Representations from Transformers, came in 2019, it is a neural
network-based technique for natural language processing pre-training. looking at the sequence of words
on a page, so even seemingly unimportant words in your queries are counted for in the result.”
• Year Launched: 2019
• Used For Ranking: No
• Looks at the query and content language
• All languages
• Language Training Model: Prescoring
• Very commonly used for many queries
• Can you optimize for it? No
70. @schachin
Kristine Schachinger
https://bensen.ai/elmo-meet-bert-recent-advances-in-natural-language-embeddings/
BERT, or Bidirectional Encoder Representations from Transformers, improves upon
standard Transformers by removing the unidirectionality constraint by using a masked language
model (MLM) pre-training objective. The masked language model randomly masks some of the tokens
from the input, and the objective is to predict the original vocabulary id of the masked word based only
on its context. Unlike left-to-right language model pre-training, the MLM objective enables the
representation to fuse the left and the right context, which allows us to pre-train a deep bidirectional
Transformer. In addition to the masked language model, BERT uses a next sentence prediction task
that jointly pre-trains text-pair representations.
There are two steps in BERT: pre-training and fine-tuning. During pre-training, the model is trained on
unlabeled data over different pre-training tasks. For fine-tuning, the BERT model is first initialized with
the pre-trained parameters, and all of the parameters are fine-tuned using labeled data from the
downstream tasks. Each downstream task has separate fine-tuned models, even though they are
initialized with the same pre-trained parameters.
Sesame Street and Search: BERT Definition
71. @schachin
Kristine Schachinger
LLMs can go forward and backwards
to predict an unknown (masked) term and/or sentence.
Also uses root words, so play for player/playing/played are the same
This allows them to derive context for what is being written.
Previous models were based on word vectors (entities and knowledge graphs)
LLM Transformers are Bidirectional
https://blog.google/products/search/search-language-understanding-bert/
72. @schachin
Kristine Schachinger
Sesame Street and Search: Why is BERT Special?
BERT can disambiguate words from the sentence and apply meaning forward and backward to those
words in order to predict a masked word using those applied contexts. This is SUPER EFFICIENT!
73. @schachin
Kristine Schachinger
Because BERT can go forward and backwards
to predict an unknown (masked) term and/or sentence.
Also uses root words, so play for player/playing/played are the same
Sesame Street and Search: Why is BERT Special?
https://blog.google/products/search/search-language-understanding-bert/
74. @schachin
Kristine Schachinger
Why are LLMs So Special?
Large Language modeling can determine the meaning of words in context
so it can better predict the next word in the sentence.
These sentences mean two different things forward and backward.
75. @schachin
Kristine Schachinger
How does this work? Transformers
What are transformers?
A transformer in language processing is a type of computer program
that is designed to understand and generate text.
It does this by using a special type of algorithm called self-attention.
Self-attention allows the program to look at all the words in a
sentence or a piece of text at once, and understand how they relate
to each other, rather than just one word at a time like traditional
methods. This way it can better understand the meaning of the text,
and can generate text that is more similar to how a human would
write.
77. @schachin
Kristine Schachinger
Simply put BERT or language modeling is
“Language modeling – although it sounds formidable –
is essentially just predicting words in a blank.”
78. @schachin
Kristine Schachinger
Why does it matter to us as SEOs?
It mostly doesn’t.
It was a breakthrough in Language Model
Processing, because it is …
+ VERY Fast
+ Uses fewer resources
+ Provides better understanding of content
83. @schachin
Kristine Schachinger
Rank Brain.
Rank Brain & Neural Matching & the
Document Relevancy Model (DRAM)
“Document relevance ranking, also known as adhoc retrieval
is the task of ranking documents from a large collection using
the query and the text of each document only.”
Rank Brain.
84. @schachin
Kristine Schachinger
Rank Brain vs Neural Matching.
Both are used to re-ordered the results post retrieval
according to “ad hoc retrieval” methods and ”dynamic relevancy”
Ranking with ONLY the document text
• https://www.searchenginejournal.com/google-neural-matching/271125/
• http://www2.aueb.gr/users/ion/docs/emnlp2018.pdf
100. @schachin
Kristine Schachinger
• When do you see it?
• Relationships between entities & search intent are weak or unknown
• -- enter Rank Brain.
• Behind the scenes, data is continually fed into the machine
learning process, to make results more relevant the next time.
• Can be combined with other algorithms such as neural matching
• No way to optimize for it
• BUT you can help prevent your page from getting one of these
results check your results for your queries.
Make sure Google is NOT CONFUSED.
Rank Brain.
103. @schachin
Kristine Schachinger
Google
https://searchengineland.com/how-google-uses-artificial-intelligence-in-google-search-379746
Neural matching. Neural matching was released in 2018 - expanded to the local search results in 2019.
Neural matching does specifically help Google rank search results and is part of the POST ad-hoc
ranking algorithms.
Links CANNOT affect this ranking sort.
• Year Launched: 2018
• Used For Ranking: Yes (but post scoring)
• Looks at the query and content language
• Works for all languages
• Very commonly used for many queries
• Applied post scoring ad hoc
• Can you optimize for it? Yes and No
110. @schachin
Kristine Schachinger
Rank Brain vs Neural Matching.
RankBrain helps Google better relate pages to concepts.
Neural Matching helps Google better relate words to searches.
• Rank Brain = page concepts
• Neural Matching = linking words to the page concepts
“…neural matching, – AI method to better connect words to concepts.” - Google
https://www.seroundtable.com/google-explains-neural-matching-vs-rankbrain-27300.html
113. @schachin
Kristine Schachinger
AI is ever-changing and unfixed.
Don’t waste the time and resources on gaming it.
But you can make it easier for the machine
learning to get it right.
Do you optimize for Machine Learning?
119. @schachin
Kristine Schachinger
Simple answer to a very complex issue?
Do your normal query research,
check the SERPs for Rank Brain issues
and then just write naturally.
Using specificity (topical hubs) PLUS
depth & breadth to create holistic content.
120. @schachin
Kristine Schachinger
Write holistic content? Does your content have depth, breadth, & semantic relationships?
Use terms that are semantically related. Image search is great for showing related terms.
124. @schachin
Kristine Schachinger
What is Structured Data?
Structured data for SEO purposes is on-page markup that
enables search engines to better understand the information
currently on your site’s web pages, and then use this information
to improve search results listing by better matching user intent.
130. @schachin
Kristine Schachinger
We can help give Google a clearer understanding.
That helps us help Google better answer
the questions users ask
and to better surface our content for those users
We give our data meaning
Google Understands
134. @schachin
Kristine Schachinger
Well Formed Text & Parsey McParseFace.
http://www.kurzweilai.net/google-open-sources-natural-language-understanding-tools
Ray Kurzweil on Google NLU
135. @schachin
Kristine Schachinger
Questions = Well Formed Text
https://ai.google/research/pubs/pub47323
“Understanding natural language queries is fundamental to many practical NLP
systems. Often, such systems comprise of a brittle processing pipeline, that is not
robust to "word salad" text ubiquitously issued by users. However, if a query
resembles a grammatical and well-formed question, such a pipeline is able to
perform more accurate interpretation, thus reducing downstream compounding
errors.”
139. @schachin
Kristine Schachinger
Takeaways.
• Think Search Queries NOT Simple Keywords
• Write in natural language
• Write using holistic content
• Focus on depth and breadth with related terms
• Add Structured Data
• Use well formed text (i.e. questions) when you can.
Takeaways.
142. @schachin
Kristine Schachinger
E-A-T.
EEAT IS Conceptual?
E-E-A-T has no
• Definition given by Google
• No ranking factors directly tied to it
• So, you are at best guessing what it means
• Why Guessing?
• Because it comes from the Quality Raters Guide which was
NEVER meant to be an SEO GUIDE.
143. @schachin
Kristine Schachinger
E-A-T Exception.
EEAT IS Conceptual with ONE Exception?
E-E-A-T = The E of Experience
Google does NOT care who your expert is, BUT if you have reviews, they
want reviews to have a personal component, so…
• Use ”I” first person terms
• Use video and images your company recorded and shot
• No stock!
• Have the person write about their personal experience with the
product