SEO Believer: Master Strategy Reference

Created: 2026-05-31

Purpose: Single reference for the positioning, geographic, competitive, and case-study strategy decisions made in research and planning. Pulls together the deeper documents: the competitor report, the Colorado scorecard, the 11-firm deep dives, the site build blueprint, and the AI-whitespace test.

Companion files in research/:

• competitor-research-report.md (best-in-class site gap analysis)
• colorado-competitor-scorecard.md (real Colorado rivals)
• colorado-seo-geo-deep-dives.md (11 Colorado firms torn down)
• site-build-blueprint.md (how to build a site that beats Volume Nine)
• ai-whitespace-test-boulder-denver.md (live AI-citation evidence, 3 verticals)

---

1. Positioning and differentiator

SEO Believer is "the SEO consultant in Colorado who runs your campaign like a lab." Evidence-based, scientific background, AI-assisted workflows that let one operator deliver agency-quality work.

• Proof: Bone Voyage from DR 0.8 to 62 on zero paid advertising, 4,000+ dogs placed on organic traffic. Frame the method, not the raw metric (a nonprofit earns links a service business never will).
• Differentiator: GEO is the credential layer, not the headline. Annette tracks AI citations weekly on her own properties and ranks her own properties first.
• Voice: Morgan Housel / James Clear. Contractions on. No em-dashes. No marketing register. Plain-spoken expertise over hype.
• ICP: Boulder/Denver service businesses, $150K to $5M revenue, owner-operated, reads but too busy to execute.

---

2. Competitive landscape (the real cohort)

Benchmark the firms a Front Range owner actually compares, not the tool giants (Ahrefs, Moz, Semrush are software companies, not consultancies, useful only as models for citation assets).

The two real threats:

• Volume Nine (Denver): the benchmark. 20 years, 25+ quantified case studies, and shipped GEO infrastructure (llms.txt, an /ai-instructions page, a free GEO Grader tool, AI-attributed-revenue case studies). Much of what we plan, already live. Weakness: Denver-only, no north Front Range, AI proof skews ecommerce/B2B not local service.
• Boulder SEO Marketing: the closest rival on our exact positioning. Multi-office, true Boulder footprint, founder-led by a visible AI/SEO personality (Chris Raulf), runs an "AI SEO & GEO Summit," real case studies. The firm most likely to come up when our prospect comparison-shops.

The field overall: crowded but shallow on GEO. Most firms have simply existed longer (domain age, reviews, indexed pages), not done better work. On AI search they split: a few ship it (Volume Nine, Boulder SEO Marketing, Firestarter, Taylor Scher for B2B SaaS), some only talk it (Webolutions, Tactical SEO, Subsilio), most ignore it (COSEOCO, NewStar, Blue Frog).

The open seat nobody holds: not one of them turns their own brand into a public, dated, falsifiable AI-citation case study, for local service businesses. That is the strongest available differentiator, and also the most copyable, so speed and consistency are the moat, not the idea.

---

3. Geographic strategy (the core decision)

The five small Front Range satellites are NOT big enough to be the hunting ground. Tier the geography instead.

Tier 1, the core:

• Boulder (home turf, relationships, referral density, over-deliver)
• Denver metro (scale, deal value, transferable proof)
• Use the GEO/AI-whitespace angle as the win mechanism so Denver's heavy traditional-SEO competition doesn't slow case studies down. GEO competition is near zero everywhere, so Denver is raidable on the AI surface even though it's brutal on the map-pack surface.

Tier 2, fast-win complements:

• Longmont (~100k) and Broomfield (~75k). Real business bases, low competition, near-certain 60-day map-pack wins. Good for seeding the proof wall early.

Tier 3, opportunistic only:

• Lafayette, Louisville, Superior, Erie, Niwot. Take a candidate only if a standout high-value, code-fixable, AI-invisible business appears. Never force the count from here. (Louisville/Superior business base also disrupted by the Marshall Fire.)

Why this split: supply (small cities can't yield 25 high-value candidates), winnability (small = easy/fast wins), closeability (pilots easiest to sign locally), and proof transferability (Denver case studies convince the paying clients we want). No single geography wins all four, so we sequence: open with Boulder + Longmont/Broomfield (easy to close, fast to win), then deliberately seed 2-3 Denver-metro flagship case studies for transferable proof.

---

4. The AI search reality (what the evidence supports)

Honest boundary on the GEO bet, so we don't overplay it:

• Today, AI search is NOT a bigger business generator than traditional local SEO for these verticals on volume. AI is ~1% of web traffic; Google still sends ~80%+ of referrals.
• But search splits by funnel, and our verticals live on the AI side of the split:
• Ready-to-buy local queries ("med spa Boulder"): local map pack appears 90%+, AI ~15%. Maps still owns the booking (top pack result takes 44-58% of clicks; 28% of local searches convert; 76% visit within 24h).
• High-consideration research queries ("average cost of dental implants in Boulder," "is LASIK worth it"): AI Overviews appear 92-97% of the time.
• Our verticals (implants, LASIK, surgery, PI law, IVF, wealth management) are exactly the long-research, high-ticket decisions where buyers spend days researching, and that research is moving to AI. So GEO matters more for us than for local business in general.
• AI is small but growing ~165x faster than organic. The empty seat fills up; first mover compounds.

Strategic conclusion: it is not GEO vs local SEO. Sell and prove both. Lead every client case study with the Maps/organic volume win (near-term ROI, what convinces the owner the pilot worked). Layer the AI-citation win (absent to cited) as the ahead-of-the-curve differentiator. The GEO win wins us the client; the local SEO win pays the client's bills.

Caveat: most national stats come from SEO vendors and conversion multiples likely suffer selection bias. Direction is solid; precise multiples are marketing. No data exists at the city-vertical granularity, so nobody can prove GEO beats local SEO in Longmont med spas, in either direction.

---

5. AI-whitespace findings (live test, Boulder vs Denver, 2026-05-31)

Tested the sources AI engines cite for "best X" in three verticals. The test killed one vertical, which makes it trustworthy on the others.

Named whitespace candidates already surfaced (strong businesses invisible to what AI cites):

• RESTŌR Medical Spa (Boulder + Denver, 5 stars, 112+ reviews, founded 2011)
• Alber Dental Studio (Boulder, 575+ five-star reviews)
• Dr. Ania, Boulder Canyon Dental, Boulder Restorative Dentistry

Takeaway: Boulder's AI-citation ground is genuinely empty in dentistry and med spas. Cheap, fast, code-fixable wins (schema, structured content, a real local roundup or cost guide, llms.txt). PI law is out for now.

---

6. The 25-company case-study play

The case studies are the engine that gets us to Volume-Nine-level proof without 20 years. They are also how we out-prove the field in the verticals it doesn't own locally.

Target verticals (ranked, evidence-backed via whitespace tests 2026-06-01 , see decisions doc): (1) cosmetic/implant dentistry [implant-cost-page win], (2) orthodontics [braces-cost-page win, same shape], (3) med spas/aesthetics, (4) HRT/TRT local in-person clinics nationwide in franchise-light metros [Nashville + Boulder widest; portable play; telehealth is NOT the competitor, multi-location franchises are], (5) cosmetic surgery [opportunistic]. DROPPED as closed/poor bets: personal injury law, wealth management, LASIK/cataract, fertility/IVF.

Selection filters (a candidate must hit these):

1. The cusp / "almost there." On the cusp on at least two of three surfaces: map pack position 4-7 across a geo-grid, organic page-1-not-top-3, or AI-answer-absent-but-strong. Not already winning (no story), not hopeless (too slow). The cusp is where effort converts to a visible jump in 60-90 days.
2. Code-fixable primary blocker. Favor companies whose main blocker is something Claude Code fixes fast: schema, llms.txt, on-page, thin/unstructured content, missing service/location pages, crawlability. De-prioritize or disqualify if the main blocker is off-site authority, review volume, or unverified GBP (slow, manual, not our leverage). Platform-access gate: must be a site we can deploy code to or rebuild.
3. High deal value. One new customer worth thousands, so a ranking win pays the retainer many times over and they can pay $2K-$5K/mo after.
4. Foundation + motivated owner. Real site, claimed GBP, some reviews (something to optimize), owner-operated and responsive.

Why this is win/win: the cusp + code-fixable + AI-whitespace company is one whose problem is exactly what we fix cheaply and fast. Our near-zero-cost fix unlocks a win they were stuck on. And GEO wins are MORE code-fixable than traditional ranking wins (AI rewards structure/schema/clean content, not decades of backlinks), so this filter naturally pushes toward the on-brand, least-contested candidates.

What makes a DRAMATIC before/after (the photogenic test): at least two of these must visibly move in 90 days for a non-expert: map pack not-in-top-3 to top-3; AI answer absent to cited (screenshot, strongest GEO proof); a clean organic traffic/leads percentage.

The offer: discounted or free pilot in exchange for (1) permission to publish a named before/after, (2) baseline data access (Search Console, Analytics, GBP), (3) a Google review and LinkedIn recommendation at the end.

Sequencing (realism rail , revised 2026-05-31): Bandwidth is the binding constraint: ~15 hrs/week solo (Claude-Code-multiplied). The rail caps case-study COUNT but the real risk is CONCURRENCY. So serialize hard:

1. Case study zero = Rebuilding Seminars (already signed, live data, relationship now public so a named before/after is publishable). Fastest proof. Build the prediction scoreboard + baseline against it first.
2. Then one new Boulder anchor (med spa or dentistry, the proven-open verticals). Run it to a published before/after.
3. Gate anchor #2 on anchor #1's 30-day signal , same discipline Section 7 applies to speculative pages, applied to pilots.
4. Scale concurrency (more anchors, Denver flagships) only when revenue funds help or a partner (Annette is open to hiring once a salary is covered).

One complete published loop teaches the real delivery + owner-management cost. Five-in-parallel teaches five half-lessons and risks five half-results. Five great published before/afters beat 25 half-finished ones. A weak result you published is worse than one you never ran.

---

7. Site build priorities (to match and beat Volume Nine)

You cannot buy Volume Nine's 20 years of domain age and reviews. You can match and beat the rest:

• Ship llms.txt, llms-full.txt, an AI-crawler instructions page, and full schema on day one (closes their GEO-infrastructure lead in a week).
• Build one cheap-to-maintain free tool (a static "is your site AI-readable?" checker), not an API-cost-heavy one.
• Build the case-study engine (the 25-company play) to reach "25+ case studies with numbers" without waiting years.
• Add the move they don't have: a public prediction scoreboard per client (dated baseline, written 30/60/90-day predictions, actuals posted next to them including misses).
• Launch a tight spine, not 50 pages. Gate every speculative asset behind a 30-day traffic signal. (Full page-by-page spec in site-build-blueprint.md.)

---

8. Open items / in progress

• Map-pack / geo-grid heat-map tool evaluation (in progress): comparing SaaS geo-grid tools (Local Falcon, Places Scout, BrightLocal, etc.) vs DIY SERP APIs (SerpApi, DataForSEO) that Claude Code can drive programmatically to build heat maps across a 1-hour-drive radius. Priority: API access and cost at 25-business scale. Recommendation to be added on completion.
• Hub article "Your Next Customer Is Asking AI Before They Ask Anyone Else" (the AI-search-buyer-research piece): drafted and deploying to annettethompson.com/hub/.
• Ajijic case study: removed and deployed live (confirmed). About-page bio mentions of AjijicRelocationGuide intentionally kept.

---

9. The one-paragraph strategy

Most Colorado SEO firms argue that AI search matters. SEO Believer shows it. Win the client with the GEO differentiator (a public, dated, falsifiable AI-citation case study, in verticals and a city where the AI ground is empty), pay the client's bills with the traditional local-SEO volume win, and pick prospects on the cusp whose blockers Claude Code fixes cheaply, so the before/after is fast and the margin is high. Anchor in Boulder, raid Denver via the uncontested AI surface, and use Longmont and Broomfield for quick wins. Five great case studies, built in public, beat twenty years of someone else's domain age.