Upgrade to Pro — share decks privately, control downloads, hide ads and more …

LLM & SEO Automation in 2026

LLM & SEO Automation in 2026

Avatar for Marco Giordano

Marco Giordano

September 19, 2026

More Decks by Marco Giordano

Other Decks in Marketing & SEO

Transcript

  1. What We’ll Cover 1.The stack to do automate SEO tasks

    with LLMs 2.Content & Ecommerce use cases with AI (incl. Enterprise) 3.Common sense 😅
  2. AI Automation starts with a data foundation Before automating decisions

    or workflows, make sure the underlying data is structured, consistent, and trusted. Raw data Data model AI Automation
  3. Bring the data into one place Google Analytics 4 Google

    Search Console Google Ads Other business data BigQuery Dataform: clean, join, and model Trusted tables for AI and automation
  4. Ugh... no custom attribution and only 14 months in Explorations!?

    Sampling, thresholding... dude you must be joking! Only 16 months and hidden data? We can help!
  5. What is n8n? A workflow automation platform for connecting apps,

    APIs and data. Connects apps, APIs, and data sources Builds workflows from triggers, rules, and actions From a trigger to a completed workflow, with human approval where it matters. Automates repeatable processes with little or no code
  6. What is Hermes Agent? An open-source AI agent for getting

    work done across tools, tasks, and workflows. Plans and executes multi-step tasks Uses tools and connected services Self-learning agent that acts as your personal assistant. Remembers context and learns reusable skills
  7. Hermes Agent: the wrapper around tools Model decides Tools execute

    Hermes Agent orchestrates Call an API n8n Run a workflow Read or write data Hermes Agent Send a message or create a task Hermes Agent turns model reasoning into useful actions by calling the right tools at the right time.
  8. Claude (Code) is a nice intro for coding and basic

    automation: 👉 Creating apps 👉 Automate basic data extraction 👉 Be called by Hermes 👉 Help you with the setups
  9. Before creating content, 3 questions matter 1 Which new articles

    or pages are missing? 2 Do we already have a page for this topic? Without these answers, content teams risk missing opportunities, duplicating existing pages, or overlooking what competitors already cover. 3 What are our competitors doing?
  10. The problem content automation solves Research is scattered across sources

    Mockup visuals Research + synthesis Many sources Collating evidence takes time Ideas stay abstract without a visual mockup Research faster Goal: being faster with content production AND management. Bring different sources together Turn the brief into mockup visuals
  11. Running my automation Marco wants a content analysis of these

    terms. Got it, running the analysis. I send you the output!
  12. But the SERP data??? How many pages I need for

    a set of keywords SERP data SERP Clustering
  13. Where is the Google data? I know what to improve

    or where we stand. Let me check st what’s in our 1 party sources. I store all the data.
  14. My numbers after 1 month 10+ 1000+ 5+ Infographics generated

    files used as sources. long-form B2B articles published. Haven’t published articles since March 26 and new infographics since... a year or more!
  15. → → → AI costs (week) $2.5 + DataForSEO (week)

    $1 + VPS (week) $2.5 = around $24-25/month (even exaggerating here).
  16. The Content Problem for this B2B Client 10+ 100+ Languages

    Subdomains All of this should be considered into the final output to avoid duplications. ♾️ topics and/or keywords
  17. Every Google SERP has a language and a geo A

    Google Search or Shopping result isn’t universal: it depends on the language and the geographic market being queried. Different language Language Different result set Same keyword Different country Geo / country Different country To measure visibility accurately, we need to know where and in which language the search is happening.
  18. The websites aren’t all the same Across our 100+ websites,

    some are global, and some are local. Global Local No country-specific geo Mapped to a specific market Global websites are not mapped to any specific country or market. Local websites are mapped to a specific country or market. They still need to be included when scanning any country or geo. gbr = United Kingdom Global websites can rank instead of local ones and this must be recorded!
  19. The automation builds the right visibility set 1 2 3

    4 5 Country + geo input Language and SERP context Scan the 100+ websites Detect global + local domains present Visibility map Global domains Websites Local domains (for this market) Instead of checking every website manually, the automation identifies who is present in each market's results.
  20. How the research workflow runs DataForSEO F MS Forms People

    submit the brief Search data Power Automate Picks up and routes the form data n8n Prepared research or content draft Orchestrate the workflow DeepSeek LLM synthesis A simple form submission becomes structured research and an actionable draft. Combined and ready to use
  21. What the automation produces A file with 2 tabs, as

    simple as possible. NEW PAGES TAB EXISTING PAGES TAB Recommendations and competitor references based on the audience. GSC/GA4 data plus advice on how to improve the page.
  22. CONTENT WRITING STAYS HUMAN The automation does: research and summarization

    for you prepares drafts But would never be suitable for writing.
  23. Case 1: Same research engine, focused on ecommerce We still

    check the SERPs, but filter the results around the categories and products that matter. Keyword set Relevant SERP checks Which products are visible? Filter by product and category Which categories are covered? From broad content research to focused ecommerce decisions. Content and visibility opportunities Where are the gaps?
  24. Case 2: Who is more visible in Google Shopping? Use

    DataForSEO Merchant API data to improve product titles and benchmark competitor domains. Merchant API data Title optimisation See which domains are featured most often, understand why, and improve the titles that help products get found. Competitor domain analytics
  25. Reading the output via Telegram The automation sends a Markdown

    (.md) file to Telegram. Automated output The workflow produces the content and title options, packages them as a .md file, and sends the file to Telegram for review. Title suggestions Multiple title options are generated from the research. Human review The results are sent to Telegram for review and next steps. 9:41 AM
  26. ✅ LLM Automation goes beyond Claude Code and ChatGPT ✅

    You still need to do data engineering
  27. ✅ LLM Automation goes beyond Claude Code and ChatGPT ✅

    You still need to do data engineering ✅ SEO context + providing right sources is your job
  28. ✅ LLM Automation goes beyond Claude Code and ChatGPT ✅

    You still need to do data engineering ✅ SEO context + providing right sources is your job ✅ Human content writing is NOT to be replaced
  29. Resources & Links Everything I showed is yours -> Want

    this tailored for your business? Book a consultation