The BEST SEO Tutorial for Businesses 2026 (Full Guide)

Author: Caleb Ulku | Published: 2025-12-09 | Source: transcript (42:04, 80.5K views at ingest)


Summary

The definitive, end-to-end statement of Caleb’s local-SEO system — the source most other SEO pages in this wiki aggregate from. He frames Local Ranking Factors (Relevance, Proximity, Prominence/Authority) as proximity, relevance, authority (a “three glasses” analogy: each business’s glass must overflow; proximity is uncontrollable, so you win on the two you can control, relevance and authority). Local SEO is a “different game” — you rank the Google Business Profile, not website URLs; the map pack takes 60–70% of clicks.

Diagnosis first: run a Local Rank Map & Top 3% Metric (GBP position at 169 points across the city; green = top-3, yellow = 4–10, red = untrusted). The key metric is the Top 3% — the green share of the map — which drives the entire content decision (topical vs. geographic).

Foundation: fix the three GBP mistakes — too few categories (use up to 10; 2–4+ min, chosen with GMB Everywhere), too few services (20+, categorized under categories), and empty fields (750-char description, 20+ photos, attributes, posts). Then align the GBP landing page via 7+1 consistency signals (GBP–Website Alignment (Consistency Signals)): title tag and H1 = primary category + city, Google Maps embed, secondary categories in subheads, review widget, address and phone matching character-for-character, plus local-business Schema Markup (Structured Data). Automate upkeep with LeadSnap (52 scheduled posts, geotagged photos, AI review responses, verified citations).

Build the Core 30: ~30 pages mirroring GBP structure (homepage → category pages → service pages), connected by editorial (in-content) links, which pass authority where nav/footer links don’t. Target keywords are category+city and service+city.

Authority via links: after 2023’s AI-content flood, Google leans harder on links as a quality filter (he traces this to Brin & Page’s 1999 PageRank insight). Two link types: “not AI slop links” (1 medium-quality external link per page, ~$35 via his icecreamtruck.shop) and local authority links (chambers of commerce at $200–300/yr — join up to 10 — plus youth-sports/charity sponsorships).

The decision loop: 30 days post-launch, re-run rank maps. Compare your Top 3% to the top competitors’ (your goal); target 25–50% of their level as your topical-relevance threshold. Below it → build topical supporting content (FAQs from People Also Ask — reworded, not copied — and Reddit). At/above it → build geographic content (landmark pages for positions 4–6, using only places Google Maps recognizes). Repeat monthly forever. Claude is his preferred writing tool.


Key Claims

  • Local ranking triad: proximity, relevance, authority; proximity is uncontrollable, so win on relevance + authority (the “three glasses”).
  • You rank the GBP, not URLs; every website page exists to help the GBP rank; map pack = 60–70% of clicks.
  • Local rank map + Top 3% metric is the diagnostic that decides topical vs. geographic content; getting this decision wrong wastes months.
  • GBP: up to 10 categories (2–4+ min), 20+ services categorized under categories, fill every field (750-char description, 20+ photos, attributes) — completion is itself a ranking signal; ~30-min audit beats 90% of competitors.
  • 8 landing-page consistency signals including character-for-character NAP match and local-business schema.
  • Google invented the transformer in 2017 (“the T in ChatGPT”); both Google and ChatGPT reward consistency — being on both = customer “sold before they call.”
  • Editorial (in-content) links pass authority; navigation/footer links barely do.
  • Links as post-2023 quality filter: “not AI slop links” (1/page, ~$35) + local authority links (chambers $200–300/yr, join up to 10; sponsorships).
  • Topical-relevance threshold = 25–50% of the top competitors’ Top 3%; below → topical content, at/above → geographic content.
  • Rework People Also Ask questions (don’t copy verbatim — Google has caught on); Reddit for real local questions.
  • Results: Plano plumber 17→2.35 in 2 weeks / $50K next month; Chicago client green across ~5M people.

Notable quotes

“You are not trying to rank blog posts. Any SEO agency pitching you on weekly blog posts, run away from them.”

Why citable: His signature contrarian position stated at maximum strength — the sharpest expression of the “structure over blogging” thesis that recurs across the corpus (and the one most in tension with mainstream content-marketing advice).

“If we’re not writing blog posts, what do we charge our clients for?” [agencies in his comments] … “The priority for these SEO agencies isn’t getting you results. It’s finding something to charge for.”

Why citable: Names the incentive misalignment in the agency model — a structural critique, useful as client-facing/content material (with the caveat that he’s selling the alternative).


Connections

Entities mentioned: Caleb Ulku, Google, Google Business Profile, LeadSnap, GMB Everywhere, Claude, Reddit · Larry Page, Sergey Brin, Apple Maps, Bing, Yelp (minor) Concepts referenced: Local SEO, Local Ranking Factors (Relevance, Proximity, Prominence/Authority), Local Rank Map & Top 3% Metric, GBP–Website Alignment (Consistency Signals), Core 30, Local Link Building & Authority, Schema Markup (Structured Data), Entity-Based SEO


Contradictions / Tensions


Notes

Treat this as the canonical reference for the SEO pillar; when other transcripts conflict on detail, prefer this one for the fully-worked system.