How to use this course

AI Agency — standalone deep-dive built from one already-researched playbook, the same relationship COVER has to Income Playbooks' Insurance module: where that source module compares one business model against seven others on a shared template, this course stays inside the AI-consulting mechanism alone and goes considerably further.

---

#What this course is

"AI agency" gets sold as one thing and is actually three, sharing the same infrastructure and credibility assets:

1. Niched-vertical implementation — building a narrow, working AI system for a client in one industry you understand, then expanding scope once trust is proven. The recommended core of this course.
2. Standalone paid AI-readiness audits — a fixed-fee diagnosis and roadmap, sold on assessment credibility rather than build risk.
3. Fractional / embedded AI-ops retainers — ongoing ownership of a client's AI systems: monitoring, retraining, new use-case scoping, ongoing accountability.

This course treats a fourth thing as inseparable from the first three rather than a separate business: AI-driven social and marketing operations — content generation, scheduling automation, AI avatars and UGC-style video, paid-social reporting — sold either as its own niched offer or bundled into a client's broader implementation. The underlying mechanism, buyer psychology, pricing logic, and compliance discipline are the same whether the workflow you're automating is a support inbox or a content calendar. A module of its own (module 2) treats it on its own terms rather than as a footnote, because it has its own tooling landscape, its own pricing benchmarks, and its own regulatory edge (AI-generated-content disclosure) that backend ops automation doesn't.

#Where this course comes from

Most of this course compiles one already-researched, confidence-tagged deep-research brief on exactly this business model — niched-vertical AI implementation, AI-readiness audits, and fractional AI-ops retainers, researched against 2026 market and regulatory conditions. That brief is also the source Income Playbooks cites for its own queued-but-unbuilt "AI consulting / agency" module, ranked first of the eight business models in that course's own shortlist — no US licensing wall at all, unlike the other seven. This course is the deeper standalone treatment of that same module, in the same way COVER is a deeper standalone treatment of Income Playbooks' Insurance module.

Module 2 (the social/marketing angle) is different: it's original supplementary research, not part of that source brief, added specifically because AI-driven marketing/content automation is a large and fast-moving enough sub-market to deserve its own sourced treatment rather than a guess. It's tagged the same way and held to the same sourcing bar — every claim in it traces to a named source in Sources & provenance, and that lesson states plainly which parts of this course come from which research pass.

#The confidence-tag system

Every material claim in this course carries one of three tags, the same system Income Playbooks and COVER use:

• [Established] — disclosed operator numbers, a primary regulatory document (a statute, an FTC rule, a named agency's own published figures), or data with a stated methodology. Treat as fact, but re-verify anything time-sensitive — a price, a rate, a rule's current status — before acting on it.
• [Directional] — a consistent pattern across multiple independent secondary sources (trade press, vendor pricing pages, practitioner writeups), but no single primary disclosure confirms it exactly. Treat as a strong planning input, not a guarantee.
• [Speculative] — a reasoned inference, a single-source marketing claim, or a number this research could not trace to anything more solid. Flagged specifically so you can weight it correctly, not so you'll ignore it.

Where a widely-repeated figure turned out to have no traceable source at all — the "95% of AI agencies will be dead by 2026" line that circulates in AI-agency content is the example this course keeps coming back to — that's said explicitly rather than repeated as fact. The underlying mechanism behind that claim (buyer fatigue and tooling commoditization compressing generic, unverticalized offers) is real and sourced; the specific number attached to it is not.

#What this course is not

It is not "how to get rich with AI." The margin in this business is actively decaying for anyone selling generic technical execution — wiring an LLM to a CRM is now a templated, largely no-code operation thousands of freelancers list as a skill. What this course argues, and sources, is that domain-context judgment — understanding a specific vertical's real bottleneck and who's liable when an agent gets it wrong — commoditizes far more slowly than the build itself, and is where the actual, defensible margin still lives in 2026. If you're looking for a course that says "just wrap an API call and charge $5K/mo," this isn't it, and the market described in module 1 will punish that offer within a sales cycle or two.

#Module map

1. Foundations — the root mechanism (why the margin exists and why it's decaying), the three sub-paths compared on unit economics, and the 2026 market and regulatory reality.
2. The social and marketing angle — the AI content/social tooling landscape, pricing benchmarks, and the AI-generated-content disclosure rules a backend-ops build never has to think about.
3. Launch — capital required, realistic timeline to first dollar, feasibility for an operator without US residency, and the week-by-week launch sequence.
4. Operate and scale — the tools and vendor stack, KPI/kill-switch gates, common failure modes, and the path to scale beyond client one.
5. Reference — every source this course draws on, and exactly which claims come from the original research brief versus this build's own supplementary research.

#How to use this course

Deciding if this is for you at all: read module 1 in full before anything else — the mechanism lesson explains why this margin exists and is decaying, and the market-reality lesson names which verticals are still defensible in 2026 versus already commoditized. If your instinct after reading it is "I'll just wrap an API and sell generic AI automation," that's the exact offer this course's own sourcing says is losing to commoditization — pick a vertical first.

Specifically interested in the social/marketing angle: read module 1's mechanism and market-reality lessons first (the underlying economics are shared), then go straight to module 2.

Ready to plan a launch: module 3 in order — capital, timeline, feasibility, then the phase-by-phase sequence.

Already building, want the operating discipline: module 4 — the tools table, the KPI gates that tell you when to keep going versus stop, and the failure modes named from real operator accounts and trade commentary, not hypotheticals.

#What this course assumes

It assumes you already have some operating history — a business you've run, a client base you've served, technical fluency with modern AI tooling (an LLM API, an orchestration platform like n8n or Make) — and that you're evaluating this as a real business decision, not a hobby. It assumes you are not a US citizen or resident with US work authorization, and treats that as the default case rather than an edge case, the same way Income Playbooks and COVER do. If you are US-based, most of this course still applies directly — the licensing wall the other Income Playbooks modules run into simply doesn't exist here, for anyone, resident or not — and the feasibility lesson tells you exactly which sections you can skip.

---

This course is business and market research, not legal, tax, or regulatory advice. Every dollar figure, pricing range, and tool cost in it is a snapshot as of the research date stated in each lesson — verify anything you intend to rely on before you spend against it, and re-verify anything regulatory (FTC guidance, state entity requirements, AI-content-disclosure rules) against its primary source before you act on it.