AI Consulting SEO / Positioning Strategy — Working Notes

Status: research phase. Not for publishing — working doc to drive keyword research and content planning for adambourg.com relaunch.

Business model under consideration

Primary idea: Retainer-based managed AI services for local Denver-area small businesses, starting with trades/field-service businesses (plumbers, electricians, contractors).

• Target: 10-20 retainer clients @ $200-$500/mo = ~$2,000-$10,000/mo recurring
• Lead offer: AI receptionist / call-answering (build on Vapi or Retell dev API, not a white-label resale product — need to own the product/margin/content story)
• Upsell path once trust is established: SEO, web, scheduling/dispatch, review management, marketing automation
• Secondary/parallel track: enterprise/compliance-focused AI consulting (RAG, HIPAA/PII/PCI-SOX), leveraging existing rag-saas-pitchdeck.html and HIPAA/security background — different funnel, different buyer, likely needs separate positioning or site section, not the same SEO target as the local-trades content.

Why AI receptionist as the lead product (not SEO/web/marketing bundle)

• Missed call = missed job — easy, visceral ROI story for a non-technical owner
• Trades workers can’t answer the phone while working (hands dirty, on a roof, driving) — solving a known pain, not creating a new concept to explain
• Trivially demoable live in a sales conversation (call the number)
• Weaker fit for offices w/ existing front desk staff (dental, legal, medical) — there it’s an overflow/after-hours pitch, less dramatic, but compliance angle (HIPAA) plays to existing strength for that vertical specifically

Content pillars (from earlier discussion)

1. Local problem/solution posts — “AI phone answering for plumbers: what it costs and does,” “How Denver [trade] businesses stop missing after-hours calls,” etc. Long-tail, low competition vs. big AI content mills.
2. Case studies / proof-of-work — highest-converting content type. Already have 2026-03-08-local-seo-rebuild.md and the vector-search post as a template. Need 2-3 more real before/after examples, even small ones.
3. Local-intent landing pages (not blog posts) — one per vertical: “AI Answering Service for Plumbers in Denver,” “AI Automation for Contractors,” etc. Structured for local SEO (NAP consistency, service area, embedded map, testimonials).
4. Plain-English AI explainer content — “What AI can actually do for your small business (no jargon),” objection-handling before a sales call.
5. Local presence / community — Google Business Profile, chamber of commerce, contractor trade groups, referral network (leverages the scrapping-hobby blue-collar network).

Tools / SEO research setup

• Semrush — already has an account for another project; add the Semrush MCP server to pull keyword volume/difficulty/questions directly into Claude Code sessions. Needs API key from Semrush account (Account Info → API).
• Ahrefs — alternative/supplement, official MCP server since July 2025, strong for backlink/competitive gap analysis.
• DataForSEO — pay-per-call alternative if avoiding a flat subscription, 79 tools across SERP/keyword/backlink/on-page.
• Google Search Console (MCP: mcp-server-gsc) — free, first-party, but only useful once the site is live and indexed. Set up + submit sitemap now regardless, so data starts accumulating during the content build-out.

Keyword validation to-do (once Semrush MCP is connected)

• Volume/difficulty for “AI answering service [trade] Denver” per vertical (plumber, electrician, HVAC, general contractor)
• Check how saturated the space already is locally — is a national player (ServiceTitan, Smith.ai, Podium) already ranking for Denver-local trades + AI answering terms?
• Volume for enterprise/compliance track terms — “HIPAA compliant RAG consultant,” “AI consultant [industry] compliance,” etc., to gauge whether that’s worth a dedicated site section
• “Questions” / People Also Ask data for FAQ-style content targeting voice/AI search

Case study to-dos (time-sensitive)

• Fab shop WordPress → static/Keystatic migration — “before” data captured. Full site audit of the old silverengineeringworks.com, dated Aug 4, 2026 (pre-crash). Site is now down, so this was the only surviving record of the “before” state — copied into notes/case-studies/silver-engineering-works-audit-before.pdf in this repo.

  **Old-site findings (the actual case-study material):**
  - 97 unique pages, avg load 0.77s, slowest pages (/equipment, /industries) 1.7–2.16s
  - Homepage alone loaded 46 scripts / 63 stylesheets — "typical of a heavily
    plugin-driven WordPress/page-builder setup," per the audit
  - 75% of pages (73/97) missing meta descriptions, including the homepage, About,
    Contact, Services
  - 73% of all 674 images sitewide missing alt text (ADA/WCAG exposure + lost
    image-search SEO)
  - **Headline hook: 31 of 97 pages (a third of the site) were unmodified WooCommerce
    demo-store content — still live and purchasable.** A visitor could add a fake
    "Space Helmet" ($220) to a cart on a metal fabrication company's site. Real
    liability, not just clutter — this is the strongest, most concrete story beat.
  - 52 duplicate URL variants (www/non-www/http/trailing-slash) all returning 200
    instead of redirecting
  - 2 dead outbound links pointing to a competitor's site (pendarvismanufacturing.com)
  
  **To do once the new static/Keystatic site is live:**
  - [ ] Run the same audit (or PageSpeed/Lighthouse) against the new site for a direct
        before/after comparison
  - [ ] Write the case study post — lead with the demo-shop/fake-purchase finding, then
        the performance and SEO-hygiene improvements
  - [ ] Confirm none of the fab-shop's real business content/copy was lost in migration
        (the audit's word counts per page, Sheet 01, are a checklist for that)
  

Vertical ideas — one file per vertical

Each candidate vertical gets its own working file once explored in depth, to keep this doc from becoming unwieldy as more get added:

• [[med-spa-idea1]] — Med Spa / Aesthetics (Denver metro). Selected over massage/wellness, personal injury law intake, dog boarding, dental, and real estate after a broad curiosity pass across ~50 Denver-metro business types. Decision: dedicated brand (not adambourg.com), agentic-delivery retainer model, inbound-only distribution (no cold calling).
• [[chiropractor-idea2]] — Chiropractic (Denver metro). Runner-up to med spa: bigger, suburb-durable local demand and a warm first-client lead (Dr. Joshua Fern), but the AI-answering-service angle is already contested by 3+ national AI-native competitors, unlike med spa’s open field.
• [[auto-repair-idea3]] — Auto repair (Denver metro), plus why fab shops/machine shops/welding/ CNC don’t work as a repeatable search-driven vertical (referral/RFQ-driven, not searched — keep Silver Engineering Works as a one-off case study only, not a vertical). Auto repair is a real open-field candidate like med spa, but lower ticket size/CPC than med spa or chiro.
• [[other-verticals-idea4]] — Personal injury law, dermatology, property management. Biggest dollar signals of the whole project (PI law CPC up to $235), but all three are heavily contested by entrenched national players (property management’s top results even include the AI voice platform vendor itself marketing into that niche) — passed on for now in favor of med spa/chiro/auto repair’s open fields. Full six-vertical comparison table at the bottom.

Open questions / other things to consider

Not yet decided — flagged for discussion:

• Vertical focus vs. broad “small business.” Sharper positioning (e.g. plumbers + electricians only, to start) may out-convert generic “AI for small business Denver.”
• Legal/liability. An AI system answering calls and booking jobs on a client’s behalf probably needs E&O / tech liability insurance and clear contract terms (what happens if the AI misquotes a job or mishandles an emergency call).
• Business entity/structure. LLC formation, contracts/MSA template for retainer clients, before signing paying customers.
• Pricing tiers. Flat retainer vs. usage-based (per-minute overages like the RAG SaaS pitch deck already models) — may want tiering (Starter/Pro/Enterprise) similar to that deck.
• Onboarding repeatability. The “low effort by client #10” assumption only holds if onboarding is templated (standard call-flow config, standard integrations) — worth designing the onboarding process before selling client #1, not after.
• Competitive landscape check. Local competitors already offering AI answering/automation to Denver trades — direct competitors, not just national platforms.
• Distribution beyond SEO. Existing YouTube channel (RobotGuide - AI) — could double as a content/lead channel; trade associations, local Chamber of Commerce, contractor referral networks (potentially higher-leverage than organic search for this buyer).
• Vapi vs. Retell decision. Still need a head-to-head on pricing/latency/integration ease before committing the build.
• Site architecture. Whether local-trades content and enterprise/compliance content live on the same site/nav or need separating to avoid diluting SEO signal (raised earlier, unresolved).
• “Built with Claude, deployed on Cloudflare” as its own angle. Currently rebuilding adambourg.com this way. Two distinct opportunities here, don’t conflate them:
  • Content asset: document the build itself as a live case study/proof-of-work post — “I rebuilt my site with Claude Code and deployed it to Cloudflare in [X hours].” Same Tier-1 proof-of-work category as the local-seo-rebuild and vector-search posts.
  • Possible service line: Squarespace/Wix dissatisfaction is a well-known, common pain (slow editors, generic templates, recurring fees, poor SEO/performance control) — “AI-built custom website migration off Squarespace” could be a real wedge offer for the same small-business audience: fast turnaround, custom (not templated), hosted on Cloudflare (cheap/fast/good Core Web Vitals). Could work as a one-time build fee that feeds into the AI-receptionist retainer relationship, or as its own lead-gen angle (“Squarespace alternative Denver,” “custom small business website AI built”).
  • Needs its own keyword check: volume/competition for “Squarespace alternative,” “leave Squarespace,” “custom website vs Squarespace small business,” etc.
  • Broaden beyond Squarespace to WordPress too. Also currently migrating a local fab shop’s site off WordPress to static HTML/CSS + Keystatic (git-based page editor) deployed at the edge (Cloudflare Pages). WordPress pain points (slow, plugin bloat, security/ maintenance burden, hosting costs) are at least as common as Squarespace pain points among small/industrial businesses — same pitch, different platform: “fast, secure, self-editable site with no WordPress plugin/maintenance overhead.”
  • Fab shop = strong case study candidate. Local manufacturing/metal fab is B2B, trades-adjacent, not yet covered by the plumber/electrician content pillar — a good second/third case study once live (before/after load speed, security surface, editing workflow via Keystatic vs. WordPress admin). Document as a Tier-1 proof-of-work post once shipped.
  • Needs its own keyword check too: “WordPress alternative small business,” “why is my WordPress site slow,” “static site vs WordPress,” “[industry] website redesign Denver.”