Everything Google Lied to Us About

Transcript

1. 1 1
2. 2 2 Download this deck: https://speakerdeck.com/ipullrank
3. 3 Salutations! I’m Mike King (@iPullRank)
5. 5 Things Google Has Lied to Us About
6. 6 6
7. 7 7 Google Has Owned Everything It Would Need to Become SkyNet
8. 8 Just Create Great Content, We’ll Figure it Out He said it so many times, I’m not going to cite a source.
9. 9 9 We Don’t Understand Content, We Fake it
10. 10 10 We Don’t Use Clicks in Rankings
11. 11 11
12. 12 12 We Use Clicks in Navboost aka GLUE -Pandu Nayak
13. 13 13 Here’s Google Telling On Itself
14. 14 14
15. 15 15 Y’all owe @randfish an apology. Go ahead and send it right now.
16. 16 16 Don’t Make Machine-Generated Content For Search
17. 17 17 Hey News People, Use this Generative AI Tool For Your Articles
18. 18 18 Ads and Organic Search are Separated like Church and State
19. 19 19 Oh, Here’s a Thread Where the Ads Team is Asking the Search Team to Juice Ads
20. 20 20 Using Expired Domains Don’t Give You an Advantage …you can get that domain into Google; you just won’t get credit for any pre-existing links. -Matt Cutts So if the content was gone for a couple of years, probably we need to figure out what this site is, kind of essentially starting over fresh. So from that point of view I wouldn’t expect much in terms of kind of bonus because you had content there in the past. I would really assume you’re going to have to build that up again like any other site. -John Mueller
21. 21 21 Then Why are Just NOW Making Expired Domains Spam?
22. 22 22 Google is not here for us; we are here for Google.
23. 23 23 Google’s Search guidelines are aspirational and we help them the enforce them.
24. 24 24
25. 25 25 SEOs are Google’s unpaid workforce.
26. 26 Things SEO Lies to Itself About
27. 27 27 There’s A Bigger Generational Shift Happening Too..
28. 28 28
29. 29 29
30. 30 30 40% of People Leaving ChatGPT Go to Google
31. 31 31 Google Still Dwarfs Everything, but More People are Using More Channels
32. 32 32 It means there is fragmentation in how information needs are being met.
33. 33 33 Our understanding of how Google works is out of date…
34. 34 34 In SEO we still think Google is Here
35. 35 35
36. 36 36 Search Engines Work based on the Vector Space Model Documents and queries are plotted in multidimensional vector space. The closer a document vector is to a query vector, the more relevant it is.
37. 37 37 TF-IDF Vectors The vectors in the vector space model were built from TF-IDF. These were simplistic based on the Bag-of-Words model and they did not do much to encapsulate meaning.
38. 38 38 Relevance is a Function of Cosine Similarity When we talk about relevance, it’s the question of similar is determined by how similar the vectors are between documents and queries. This is a quantitative measure, not the qualitative idea of how we typically think of relevance.
39. 39 39 Google Shifted from Lexical to Semantic a Decade Ago
40. 40 40 Very very few SEO tools are offering analysis that aligns with how Google works today.
41. 41 41 Google Has Been More Like This Since Hummingbird
42. 42 Semantic Search is Fueled by High Density Embeddings …just like large language models. A lot of what Google has always been trying to do is more real.
43. 43 43 This Allows for Mathematical Operations Comparisons of content and keywords become linear algebraic operations.
44. 44 44 Words are Converted to Multi-dimensional Coordinates in Vector Space
45. 45 45 We Went from Sparse Embeddings to Dense Embeddings
46. 46 46 Word2Vec Gave Us Hummingbird
47. 47 47 I Wrote About How You Can Get These With Screaming Frog and OpenAI Vectorize your content as you crawl. https://ipullrank.com/vector-embeddings-is-all-you-need
48. 48 48 I Talked About How You Use Google Sheets to Do the Analysis Cosine Similarity is the measure of relevance: https://ipullrank.com/cosine-similarity-knn-in-google-sheets
49. 49 49 Since the Introduction of BERT, Google Has Looked More Like This
50. 50 50 Google Has Been Using Public About Models Since 2020 This is why some of the search results feel so weird. A re-ranking of documents with a mix of lexical and semantic. https://arxiv.org/pdf/2010.01195.pdf
51. 51 51 The way SEO build links doesn’t work as well we think it does anymore.
52. 52 52 Google Has Always Aspired to Leveraging Link Relevance in Meaningful Ways
53. 53 Dense Retrieval You remember “passage ranking?” This is built on the concept of dense retrieval wherein there are more embeddings representing more of the query and the document to uncover deeper meaning.
54. 54 54 Dense Retrieval is Scoring down to the Sentence Level
55. 55 55 You need to focus on building more relevant links than higher volumes of links.
56. 56 56 Under the SGE Model, Google is Structured Liked This
57. 57 57 All of this is a huge problem because SEO software still operates on the lexical model.
58. 58 58 No one is actually improving content when they do the skyscraper technique
59. 59 Indexing is Also Harder It’s not being talked about as much, but indexing has gotten a lot harder since the Helpful Content update. You’ll see a lot more pages in the “Discovered - currently not indexed” and “Crawled - currently not indexed” than you did previously because the bar is higher for what Google deems worth capturing from the web.
60. 60 60 I Believe This is a Function of Information Gain Conceptually, as it relates to search engines, Information Gain is the measure of how much unique information a given document adds to the ranking set of documents. In other words, what are you talking about that your competitors are not?
61. 61 61 In conclusion: “More content” is no longer inherently the most effective approach because there’s no guarantee of traffic from Google.
62. 62 62 The only content you should be making
63. 63 63 I’m Leaving Y’all with Four Actions Today 1. How to Prune Your Content 2. How to Use LLMs 3. How to Appear in LLM based Search Engines 4. How to Think About Relevance
64. 64 The Content Pruning Process
65. 65 65 Pruning and Optimization Work Quite Well Together
66. 66 Aleyda Has a Process Aleyda’s workflow is a great place to work through whether your content should be pruned or not. https://www.aleydasolis.com/en/crawli ng-mondays/how-to-prune-your-website- content-in-an-seo-process-crawlingmon days-16th-episode/
67. 67 67 We like automate to get to a Keep. Revise. Kill. (Review.)
68. 68 68 Content Decay The web is a rapidly changing organism. Google always wants the most relevant content, with the best user experience, and most authority. Unless you stay on top of these measures, you will see traffic fall off over time. Measuring this content decay is as simple comparing page performance period over period in analytics or GSC. Just knowing content has decayed is not enough to be strategic.
69. 69 69 It’s not enough to know that the page has lost traffic.
70. 70 70
71. 71 71 The Content Potential Rating (CPR).
72. 72 72 Content Potential Score
73. 73 73 Interpreting the Content Potential Rating 80 - 100: High Priority for Optimization 60 - 79: Moderate Priority for Optimization 40 - 59: Selective Optimization 20 - 39: Low Priority for Optimization 0 - 19: Minimal Benefit from Optimization If you want quick and dirty, you can prune everything below a 40 that is not driving significant traffic.
74. 74 74 Combining CPR with pages that lost traffic helps you understand if it’s worth it to optimize.
75. 75 75 Step 1. Pull the Rankings Data from Semrush Organic Research > Positions > Export
76. 76 76 Step 2: Pull the Decaying Content from GSC Google Search Console is a great source to spot Content Decay by comparing the last three months year over year. Filter for those pages where the Click Difference is negative (smaller than 0) then export.
77. 77 77 Step 3: Drop them in the Spreadsheet and Press the Magic Button
78. 78 The Output is List of URLs Prioritized by Action Each URL is marked as Keep, Revise, Kill or Review based on the keyword opportunities available and the effort required to capitalize on them. Sorting the URLs marked as “Revise” by Aggregated SV and CPR will give you the best opportunities first.
79. 79 79 Get your copy of the Content Pruning Workbook : https://ipullrank.com/cpr-sheet
80. 80 How to Kill Content Content may be valuable for channels outside of Organic Search. So, killing it is about changing Google’s experience of your website to improve its relevance and reinforce its topical clusters. The best approach is to noindex the pages themselves, nofollow the links pointing to them, and submit an XML sitemap of all the pages that have changed. This will yield the quickest recrawling and reconsideration of the content.
81. 81 81 How to Revise Content Review content across the topic cluster Use co-occurring keywords and entities in your content Add unique perspectives that can’t be found on other ranking pages Answer common questions Answer the People Also Ask Questions Restructure your content using headings relevant to the above Add relevant Structured markup Expand on previous explanations Add authorship Update the dates Make sure the needs of your audiences are accounted for Add to an XML sitemap of only updated pages
82. 82 How to Review Content The sheet marks content that has a low content potential rating and a minimum of 500 in monthly search volume as “Review” because they may be long tail opportunities that are valuable to the business. You should take a look at the content you have for that landing page and determine if you think the effort is worthwhile.
83. 83 How to Use LLMs for Content
84. 84 84 What is Retrieval Augmented Generation?
85. 85 85 It’s Not Difficult to Build with Llama Index sitemap_url = "[SITEMAP URL]" sitemap = adv.sitemap_to_df(sitemap_url) urls_to_crawl = sitemap['loc'].tolist() ... # Make an index from your documents index = VectorStoreIndex.from_documents(documents) # Setup your index for citations query_engine = CitationQueryEngine.from_args( index, # indicate how many document chunks it should return similarity_top_k=5, # here we can control how granular citation sources are, the default is 512 citation_chunk_size=155, ) response = query_engine.query("YOUR PROMPT HERE")
86. 86 86
87. 87 87
88. 88 88
89. 89 89
90. 90 90
91. 91 91
92. 92 92
93. 93 Integrate Promptitude with Zapier or Make
94. 94 94 If you’re building your own custom stuff, use Gemma. Vertex is so expensive.
95. 95 The threat of Google’s Search Generative Experience (SGE) aka AI Overviews
96. 96 96 Queries are Longer and the Featured Snippet is Bigger 1. The query is more natural language and no longer Orwellian Newspeak. It can be much longer than the 32 words that is has been historically in order 2. The Featured Snippet has become the “AI snapshot” which takes 3 results and builds a summary. 3. Users can also ask follow up questions in conversational mode.
97. 97 97 AI Overviews are a Function of Retrieval Augmented Generation (RAG)
98. 98 98 The Search Demand Curve will Shift With the change in the level of natural language query that Google can support, we’re going to see a lot less head terms and a lot more long tail term.
99. 99 99 The CTR Model Will Change With the search results being pushed down by the AI snapshot experience, what is considered #1 will change. We should also expect that any organic result will be clicked less and the standard organic will drop dramatically. However, this will likely yield query displacement.
100. 10 0 Rank Tracking Will Be More Complex As an industry, we’ll need to decide what is considered the #1 result. Based on this screenshot positions 1-3 are now the citations for the AI snapshot and #4 is below it. However, the AI snapshot loads on the client side, so rank tracking tools will need to change their approach.
101. 10 1 10 1 Context Windows Will Yield More Personalized Results SGE maintains the context window of the previous search in the journey as the user goes through predefined follow questions. This will need to drive the composition of pages to ensure they remain in the consideration set for subsequent results.
102. 10 2 SGE is Susceptible to Spam and Lower Quality Sites
103. 10 3 10 3 Luckily Users Love it So Much
104. 10 4 10 4 Ranking in Search Generative Experience is more about relevance than the other signals.
105. 10 5 How to Appear in LLMs
106. 10 6 10 6 Blocking LLMs is a Mistake. Appearing in these places will be recognized as brand awareness opportunities very soon.
107. 10 7 10 7 What is Mitigation for AI Overviews? 1. Manage expectations on the impact 2. Understand the keywords under threat 3. Re-prioritize your focus to keywords that are not under threat 4. Optimize the passages for the keywords you want to save
108. 10 8 10 8
109. 10 9 10 9 We Can Also Show You Per Keyword How You Show Up
110. 11 0 11 0 It’s all about the Fraggles.
111. 11 1 11 1
112. 11 2 11 2 The Fraggles Show What SGE Used for the AI Snapshot
113. 11 3 11 3 Scroll to Text You can capture the copy used to inform the AI snapshots by scraping the Scroll to Text copy from the page.
114. 11 4 11 4 There’s a Nearly Linear Relationship Between Fraggle Relevance and AI Snapshot Appearance Relevance against the chunks to keyword: Relevance against AI Snapshot:
115. 115 11 5 Check out MarketBrew’s Free Tool to Help
116. Embrace Structured Data There are three models gaining popularity: 1. KG-enhanced LLMs - Language Model uses KG during pre-training and inference 2. LLM-augmented KGs - LLMs do reasoning and completion on KG data 3. Synergized LLMs + KGs - Multilayer system using both at the same time https://arxiv.org/pdf/2306.08302.pdf Source: Unifying Large Language Models and Knowledge Graphs: A Roadmap
117. 117 11 7
118. 118 11 8
119. 119 11 9
120. 120 12 0 They Share Their Prompts in Their Code The GEO team also shared the ChatGPT prompts that help them improve their visibility. You can augment them and put them to work right away. https://github.com/GEO-optim/ GEO/blob/main/src/geo_functi ons.py
121. 121 12 1
122. 12 2 Roll the Credits
123. 12 3 12 3 Get your threat report: https://ipullrank.com/sge-report
124. Thank You | Q&A mike@ipullrank.com Award Winning, #GirlDad Featured by Get Your SGE Threat Report: https://ipullrank.com/sge-report Play with Raggle: https://www.raggle.net Download the Slides: https://speakerdeck.com/ipullrank Mike King Chief Executive Officer @iPullRank