Knowledge base v2.13.0 · Community edition · data through 2 October 2026
Topic · AI and Search
Day 1's first lightning session showed how practitioners use AI in SEO work. An agency SEO director uses AI only where it enriches work the team has already defined: people set the strategy buckets, scoring and rules, and AI pulls in the data and applies them, which cut a competitor keyword analysis from three days to a few hours; his lesson was that AI should not decide what good looks like. A second speaker showed how LLMs get search data wrong, for example calling a move in average position from 8 to 9 an improvement, and argued for receipts that trace every figure to its data request, for never letting a model check its own output and for giving it fixed metric definitions. A community speaker applied Pascal's line about false windows built for symmetry to LLMs, which sometimes are not trying to produce the most accurate figures or interpretations of them. Other talks showed an LLM loop that rewrites structured data until Google's structured-data test passes, a small model (Claude Haiku) choosing keywords under fixed rules (no brand terms, no navigation terms, at most two keywords per page), and monthly Search Console reports automated step by step with AI as a coding partner. Google's hosts closed by calling AI a tool that can be used well or badly and advised always checking what AI tools do. On Day 2 a community speaker described 'prompt journalism': a prompt-assisted workflow that cleans, maps and restructures recorded interviews into an article and social posts, to make more of material that already exists rather than to create more content. Author’s view: a fix loop that stops when Google's test passes proves only that the markup is valid, not that its values are true, and the Search Console API itself is free, so the cost check before automating is about usage limits and the paid parts of the pipeline.
Things in this topic 7
Counts are claims that name the thing. All things
What to do
A community speaker's approach to AI in agency SEO work is to use it where it meaningfully elevates and enriches existing work rather than replacing it, keeping full strategic oversight and control of the output.
“Human judgment and strategy is the default, and AI simply pushes it at scale.”
Wording checked against the slide or recording
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Asking AI to produce whole SEO deliverables from scratch, such as creating content or auditing a website, is where people get stuck; break a task into its elements and decide where AI can and should play a part, a community speaker advised.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker designs AI-assisted deliverables like a flowchart, keeping guardrails in place and keeping the prompts succinct and measured, so the team controls the process end to end.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
In an agency's competitor keyword analysis, a deliverable defined before AI, the team set the strategy buckets and the keyword scoring metrics itself and used AI only to pull in the data, never to make a judgment.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
With AI pulling in the data, a competitor keyword analysis covering 26 competitors and 79 data points per keyword went from three days of work to a few hours, a community speaker reported for his agency.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
For a content audit of a client blog with over 3,000 posts, an agency defined the parameters, scoring and weightings manually and had AI apply them, producing posts grouped by category, possible cannibalisation flags, a priority list and a list of posts to check by hand.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
The key lesson of a community talk on AI automation: AI should not decide what good looks like; people set the rules for what is good and AI applies them.
“it didn't decide what good looked like. We did that.”
Wording checked against the slide or recording
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
In a post-migration analysis of over 200,000 lines of search traffic data and more than 45,000 pages, where many URLs had been merged into one, an agency set the rules for what counted as the same page and used AI to match old and new addresses and roll the traffic up per page.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker builds SEO recommendations without AI and then uses AI to turn them into mock-ups and visuals, which help win buy-in from clients and internal teams.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
An agency built its own AI visibility tracking tool so it could control the methodology: the team wrote pages of flowchart-style prompts (if this happens, do that) and tested manually, and AI only generated the code.
Speaker James PowleyIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker's example of an LLM error on Search Console data: a chatbot called an average position that went from 8 to 9 an improvement, although position 1 is the top, so the move from 8 to 9 is a decline.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
How often an LLM gives a wrong answer about data depends on the model, but how often the mistake is noticed depends on the person using it, a community speaker said.
“how often can you notice the mistake? And that depends on you.”
Wording checked against the slide or recording
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
In agency marketing-data pipelines, as in financial forecasting, one misplaced figure travels downstream and compounds into later analysis and decisions, which is why LLM errors on search data must be caught early, a community speaker warned.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker reported that a code refactor of his company's AI analysis product left data fetching correct but made the analysis shallow while the answers still read well, with the same model, so polished output is no proof of correct analysis.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
The first rule for using LLMs on search data, according to a community speaker, is never to let the model check its own output.
“not let the model grade its own homework”
Wording checked against the slide or recording
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker's first guardrail for LLM answers about data is a receipt: every figure the model gives must be traceable to the data request behind it, so the answer can be verified.
“If there is no receipt, there is no reimbursement.”
Wording checked against the slide or recording
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
In a community speaker's verification system, each claim in an LLM's data answer is marked verified, flagged (a hallucination), or skipped or not found.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Claims that cannot be verified go to a judge, models from other vendors that score them; a failing claim is struck through with the reason shown, and the system annotates and proposes corrections rather than rewriting the answer, a community speaker described.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Give an LLM your metric definitions as fixed, protected context, for example that position 1 is the top, and leave nothing for the model to guess, a community speaker advised.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Before using AI-generated numbers, for example in a client meeting, always ask the AI for the receipts and the reason behind every number, because AI hallucinates; LLMs present their reasoning as a forward chain, so the user has to check it backwards, a community speaker said.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community demo showed a script that opens Chrome with a saved session, pastes a URL into Google's Rich Results Test, waits about 15 seconds, then screenshots and saves the result and retries on failure (the demo's recording is largely unintelligible; the tool name is a best reading).
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
In a community demo, a fix loop gave a large LLM (Claude Opus) the current markup, the errors Google's test reported and Google's documentation, had it write new JSON-LD and re-ran the test, with at most three attempts; the demo's page passed on the second.
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community demo matched model size to the task: a small, cheap model (Claude Haiku) chose keywords from a page's URL, title, H1, description and schema types, while a large, expensive model (Claude Opus) fixed code and wrote the summary.
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
At a community speaker's agency, monthly client reports took each SEO manager one full workday every month, contained technical errors and were not read by anyone, which led the team to automate them.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
To automate SEO reporting with AI-assisted coding, a community speaker advised against aiming for an app or a perfect dashboard: pick one repetitive task, define the smallest useful output, ask the AI what is possible, then build and debug one step at a time.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker who is not a developer automated monthly SEO reports with Python in Visual Studio Code, using Claude and ChatGPT as coding partners and Google Cloud for access to the Search Console API.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
An agency's automated monthly SEO report follows four simple steps: the data comes in, a script processes it, AI summarises it and the result goes onto a dashboard.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Before writing code to pull Search Console data for one property, check which data you actually need, what the API can provide and what the API costs, a community speaker advised.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker gave the AI a detailed description of the tech stack and requirements and started tiny, with one script, one client and one output, before having the AI write the code.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
When debugging AI-written code with AI, always read the code and understand what went wrong, so you can fix it yourself next time, a community speaker advised.
“Don't trust it blindly, and try to understand what is happening in your code”
Wording checked against the slide or recording
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A good AI summary in an automated SEO report needs cleaned, good-quality data, a clear definition of what is wanted and a detailed, explicit prompt, a community speaker said.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
The resulting monthly report showed Search Console data on one side and an AI-written summary of what happened that month on the other; an in-house AI specialist built the dashboard from the SEO manager's code.
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
To start automating SEO work with AI, a community speaker said you need no expertise, only one painful task, a clear outcome and the patience to iterate.
“you don't need to be an expert; you just need to try.”
Wording checked against the slide or recording
Speaker Kira BreuerIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Closing the lightning talks on automation and AI, one of the session's Google hosts called AI a tool that can be used well or badly: it can even be used for content generation, which none of the talks featured, and it is fantastic for coding and analysis but less so for other tasks.
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
One of the Google hosts of the AI lightning talks advised always checking what AI tools do and focusing on what the business and its users need.
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community speaker applied Pascal's line about makers of false windows built for symmetry, whose rule is to make pleasing figures rather than to speak accurately, to today's LLMs: sometimes they are not trying to produce the most accurate figures or the most accurate interpretations of them.
Speaker Rafael KovashikawaIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
A community demo's script used both input modes of Google's Rich Results Test: the URL mode for public pages, which Google fetches itself, and the code mode, into which the script pasted the page's HTML, for private pages and pages Google cannot fetch (best reading of a largely unintelligible recording; the second kind of page was heard as 'dead', possibly 'dev').
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Used byrequirement DEV-SDA-11
In a community demo, the small model choosing keywords (Claude Haiku) followed fixed rules, no brand terms, no navigation terms and at most two keywords per page; the Spanish example was 'zapatos de mujer' and 'comprar zapatos de mujer' (women's shoes, buy women's shoes).
Speaker not identifiedIn Day 1, 13:10 · Lightning session A: Automation and AIEvidence transcript
Author's view: the Search Console API itself costs nothing, so the cost check advised on stage before automating reports is about the API's usage limits and the paid parts of the pipeline, such as storage and AI model calls.
Author Ibrahim AnjroAnnotates Day 1, 13:10 · Lightning session A: Automation and AI
Author's view: an LLM fix loop that stops when Google's test passes proves only that the JSON-LD is valid, not that its values are true; Google's generative AI content guidance asks for AI-generated structured data to be fact-checked as well as validated.
Author Ibrahim AnjroAnnotates Day 1, 13:10 · Lightning session A: Automation and AI
A community talk described 'prompt journalism' as journalistic and content work, such as interviews and podcasts, done with a prompt-assisted workflow whose aim is to optimize material already recorded, which holds expertise, quotes and context, rather than to create more content.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
Used byglossary term Prompt journalism
The first step of a community speaker's prompt-assisted workflow is cleaning the interview transcript of fillers, a step that needs editorial and SEO judgment because a published page is read by many systems whose way of consuming it is unknown.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
A community speaker cited W3C guidance that, when cleaning a transcript, it is usually not appropriate to correct the speaker's grammar or other mistakes.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
In the mapping step of a prompt-assisted workflow, an AI system splits an interview transcript into chapters with timestamps, headings and entities; in a community speaker's example it correctly extracted the subsea cables 2Africa and Equiano from a passage on Internet infrastructure in Africa.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
A community speaker said the chapters and entities extracted from an interview transcript can then be used to generate metadata and structured data for the published content.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
The second step of the prompt-assisted workflow applies the newsroom's inverted pyramid (the five Ws and H first, then attribution and quotes, context and secondary text); because it is a fixed structure, it can be written as a prompt that generates the article.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
From one unified interview transcript, a community speaker's workflow produces a full article for the website and shorter angles for LinkedIn, Substack, X and YouTube.
Speaker Patrick DomanicoIn Day 2, 14:00 · Lightning session F: MediaEvidence transcript
A community speaker applied Pascal's line about makers of false windows built for symmetry, whose rule is to make pleasing figures rather than to speak accurately, to today's LLMs: sometimes they are not trying to produce the most accurate figures or the most accurate interpretations of them.
A community speaker reported that a code refactor of his company's AI analysis product left data fetching correct but made the analysis shallow while the answers still read well, with the same model, so polished output is no proof of correct analysis.
A community demo's script used both input modes of Google's Rich Results Test: the URL mode for public pages, which Google fetches itself, and the code mode, into which the script pasted the page's HTML, for private pages and pages Google cannot fetch (best reading of a largely unintelligible recording; the second kind of page was heard as 'dead', possibly 'dev').
A community demo showed a script that opens Chrome with a saved session, pastes a URL into Google's Rich Results Test, waits about 15 seconds, then screenshots and saves the result and retries on failure (the demo's recording is largely unintelligible; the tool name is a best reading).
In a community demo, the small model choosing keywords (Claude Haiku) followed fixed rules, no brand terms, no navigation terms and at most two keywords per page; the Spanish example was 'zapatos de mujer' and 'comprar zapatos de mujer' (women's shoes, buy women's shoes).
A community demo matched model size to the task: a small, cheap model (Claude Haiku) chose keywords from a page's URL, title, H1, description and schema types, while a large, expensive model (Claude Opus) fixed code and wrote the summary.
Google's Rich Results Test help says the tool tests either a page's full URL or a pasted code snippet (Code instead of URL); all page resources must be reachable by an anonymous user on the internet, so resources behind a firewall or a password are not available to the test unless exposed, for example through a tunnel.
A community demo showed a script that opens Chrome with a saved session, pastes a URL into Google's Rich Results Test, waits about 15 seconds, then screenshots and saves the result and retries on failure (the demo's recording is largely unintelligible; the tool name is a best reading).
Google's guidance on generative AI content warns that AI output can contain hallucinations and says AI-generated metadata, including structured data and image alt text, should be fact-checked before publishing and the markup validated.
In a community demo, a fix loop gave a large LLM (Claude Opus) the current markup, the errors Google's test reported and Google's documentation, had it write new JSON-LD and re-ran the test, with at most three attempts; the demo's page passed on the second.
In a community lightning talk, agency founder Nik Vujic presented how his agency feeds Search Console and GA4 data to an LLM agent to support decisions for its clients.
A community speaker who is not a developer automated monthly SEO reports with Python in Visual Studio Code, using Claude and ChatGPT as coding partners and Google Cloud for access to the Search Console API.
The community workflow's third step feeds the exported Search Console dataset into the agency's LLM agent, which queries the dataset and returns the evidence rather than a sample.
An agency's automated monthly SEO report follows four simple steps: the data comes in, a script processes it, AI summarises it and the result goes onto a dashboard.
Nik Vujic's slide said no agent is needed to start: pull the Search Console data, import it into any LLM and query it in conversation; his agency built its agent to have everything in one place.
An agency's automated monthly SEO report follows four simple steps: the data comes in, a script processes it, AI summarises it and the result goes onto a dashboard.
Review the filters and regexes that Search Console's AI-powered configuration proposes before trusting the numbers: Google's launch post calls the feature experimental, warns that AI can misinterpret requests, and limits it to the Search results Performance report (not Discover or News) and to configuration, not sorting or exporting.
The first rule for using LLMs on search data, according to a community speaker, is never to let the model check its own output.
Validate markup in the Rich Results Test before it reaches templates, and monitor it after release
Rests on 8 claims, 1 of them in this topic