SEO Believer Launch Plan
Internal strategy reference. Enter hub password to continue.
That's not right. Try again.
SEO Believer: Launch Plan
SEO Believer: 90-Day Execution Plan (post-critique, 2026-06-01)
This plan turns the revised strategy (see research/seobeliever-strategy-master.md and decisions/2026-05-31-bandwidth-and-case-study-zero.md) into sequenced action. It assumes Annette's real constraints: ~15 hrs/week of HER time (a deliberate lifestyle cap, not a capacity ceiling), multiplied by Claude Code (code-fixable work) and a proven offshore SEO team model (she took Bone Voyage to DR 62 this way), funded by ~$3,500/mo investment income + retirement runway.
0. Guiding principles (the rails)
- Serialize on Annette's attention, parallelize on delegated labor. One client relationship + one quality bar in her head at a time. Claude Code and offshore staff carry volume.
- Lead every published case with the Maps/organic volume win; layer the AI-citation win as the differentiator. AI search is the consideration-phase + differentiation play, NOT proof it out-converts Maps today. Be honest about that boundary in all client material.
- Test before pitching. No vertical or metro gets pursued until a whitespace/openness check backs it.
- Cite only verifiable numbers. Bone Voyage = DR 0.9 to 62 (Ahrefs-verified, growth-to-peak, zero paid ads). Never cite "0 to 69." Replace vendor GEO conversion stats with SEO Believer's own data as soon as it exists.
- Every claim is falsifiable and dated (the brand promise: measure, don't guess).
1. Verticals (evidence-ranked, FINAL)
Lead: (1) Dentistry implant-cost, Boulder , displace the single cost authority (Boulder Dental Center "$3,150"). (2) Orthodontics braces-cost, Boulder , same single-source shape (dentistsranked). (3) Med spas, Boulder , proof vertical. (4) HRT/TRT local in-person clinics, nationwide, franchise-light metros , Annette's personal-credential moat, portable; first target St. Louis (open + daughter on the ground). (5) Cosmetic surgery , opportunistic.
Dropped as closed/poor bets: PI law, wealth management, LASIK/cataract, fertility/IVF.
2. Phase A (Weeks 1-3): Foundation + Case Study Zero
A1. Rebuilding Seminars = Case Study Zero. Already-signed live client; relationship now public so a named before/after is publishable. Capture baseline NOW: GSC, Analytics, GBP, current AI-citation state (screenshot Rebuilding's absence/presence across ChatGPT/Claude/Perplexity/Gemini/Grok/AI Overviews for its money queries). Write dated 30/60/90-day predictions. This is the prediction-scoreboard's first entry.
A2. Build the RankGridEngine (the DataForSEO heat-map engine). See Section 6 for the technical spec. Default to Standard queue mode ($0.0006/point) for sweeps, Live ($0.002) for spot-checks. Cost-estimate before every run.
A3. Site spine on seobeliever.com. Ship llms.txt, llms-full.txt, full schema (Organization, Person, LocalBusiness, Service, FAQPage, Article, BreadcrumbList), robots.txt allowing AI crawlers, sitemap. Publish the Bone Voyage proof graphs (DR 0.9 to 62, keywords 0 to 31,310, referring domains 0 to 1,187) as the hero case study. Gate speculative pages behind a 30-day traffic signal.
3. Phase B (Weeks 3-7): First New Anchor + Engine in Production
B1. National HRT openness scan (once DataForSEO funded, ~$30-50). Score top-100 US metros by Gameday Men's Health franchise map-pack density (the automatable "lock" signal) + AI-citation openness. Overlay warm-access cities (St. Louis, DC, Boston, Miami, Greenville, Boulder). Output ranked first-target list.
B2. Land anchor #1 in St. Louis HRT (warm intro via daughter). Confirm openness with a fuller read first (current sample was thin). Run cusp + code-fixable filter. Offer: discounted/free pilot for named before/after + data access + review.
B3. Boulder dentistry implant-cost anchor. Build the dated, schema-rich implant-cost page play. This is the highest-probability single AI-citation win identified.
B4. Gate anchor #3 on anchor #1's 30-day signal. Do not open more concurrency until one loop closes.
4. Phase C (Weeks 6-12): Differentiation Engine
C1. YouTube channel on local SEO + GEO for service businesses (the real moat: compounding audience + authority, hard to copy, feeds AI citation via transcripts). Use repurpose-niche / podcast-ops skills + offshore editing. Sequence so it does not eat case-study hours.
C2. Public prediction scoreboard live on site (proof layer, not the anchor): dated baselines, written predictions, actuals posted including misses.
C3. First offshore hire once a closed pilot pipeline justifies it (hire against booked revenue, fund from runway if needed). Train on the repeatable manual SEO labor; keep GEO-specific work close until proceduralized.
5. Weekly cadence (within 15 hrs)
- Annette's hours: sales calls, owner relationships, training, quality bar, strategic calls, YouTube on-camera. (The 1x work only she can do.)
- Claude Code: schema, llms.txt, content drafting, heat-map runs, reporting, scoreboard updates. (10x work.)
- Offshore (when funded): citations, directory submissions, outreach, content production. (Volume work.)
6. RankGridEngine technical spec
Purpose: business-agnostic local rank-grid heat-map engine; point at any business+keywords+city, get a heat map of where it ranks across a geographic grid. Sellable internal capability ("we built our own rank-grid engine").
Stack: Python. Modules already scaffolded in RankGridEngine/: creds.py (central-first credential load), grid.py (NxN grid generation, km-to-degrees), client.py (DataForSEO Maps SERP client, live + standard modes, batching, retry).
Pipeline:
- Grid: center lat/lng + radius (drive-time to km) -> NxN grid (start 7x7, scale to 11x11), fixed zoom 16-17 across sweeps for comparability.
- Query: task array (one per point per keyword), POST batches up to 100 to Maps SERP API. Standard queue for sweeps, Live for spot-checks. Respect 2,000 calls/min, retry with backoff.
- Parse: per grid point, scan items[] for target business (place_id match, fuzzy name+address fallback), record rank_absolute or 21+/not-found sentinel. Store rows (business, keyword, lat, lng, rank) to SQLite/CSV (Cloudflare D1 optional later).
- Render: folium HeatMap or colored CircleMarkers (green 1-3, yellow 4-10, red 11+/absent) -> drops into a hub page. Static PNG via matplotlib/plotly for reports.
- Cost guard: estimate (#points x #keywords x mode-rate) and print BEFORE firing; refuse runs over a configurable cap.
Franchise-density application: for the HRT national scan, run a light grid per metro for the core HRT keywords, count distinct Gameday Men's Health locations appearing in the map pack -> openness score.
7. Success metrics (90-day)
- SEO Believer cited by name in >=1 AI engine for "SEO consultant Boulder" / "AI search optimization Colorado."
- GBP in local 3-pack for >=3 target queries (geo-grid verified).
- Rebuilding Seminars case study published with dated predictions vs actuals.
- >=1 new anchor (St. Louis HRT or Boulder dentistry) with a published before/after.
- RankGridEngine in production, used for >=2 client baselines.
- Bone Voyage proof graphs + site spine + scoreboard live.
- DR climbing past 15, >=25 referring domains.
8. Known risks / open questions
- National HRT vs Boulder-local attention split: three lead verticals are Boulder-local, HRT is national. Does HRT run parallel or after a local proof? (Leaning: prove St. Louis HRT in parallel because warm-access lowers its cost, but it competes for the same 15 hrs.)
- DataForSEO balance must be funded before the national scan; cost guard mitigates overspend.
- Thin whitespace samples (St. Louis, Boston = 1 query each) need confirmation before pitching.
- Solo-to-team transition timing: hire against booked revenue, not hope.
- GEO not yet proceduralized for offshore delegation; keep close until repeatable.
- AI-citation magnitude unproven: direction solid, conversion multiples are vendor-sourced; replace with own data ASAP.
Internal strategy reference · Private · ← Hub