The methodology # How the AI Growth Marketing Team actually works. No black box. This page walks the engagement end to end: the AI marketing methodology behind it, what the strategist decides, what the agents produce, what you review, and what transfers to your account when it ends. [Book a 30-minute discovery call](/contact) [Start with the Strategy Sprint →](/services/strategy-sprint) On this page [01 An engagement, week by week](#weeks) [02 The operating model](#model) [03 Four ways compared](#compare) [04 The architecture, labeled](#figure) [05 What you own when it ends](#ownership) [06 How it works in practice](#faq) An engagement, week by week ## From intake to handover. The methodology is a documented 13-phase pipeline: it takes raw company materials in and puts SERP-justified, ranked pages out. Here is how that unfolds in calendar time. 1. 1 Week 1 ### Discovery An intake call and a materials audit: your analytics, your positioning documents, your past campaigns. In parallel, agents enrich the picture from the outside: competitor sites, review platforms, your own sitemap. The output is a company dossier, the fact-checking source every later deliverable traces back to. 2. 2 Weeks 2–4 ### Strategy Sprint The full pipeline runs: a semantic core collected across your geographies and clustered by commercial value; SERP analyses for each priority cluster; a site architecture mapped to that data; page templates with block orders justified by what actually ranks; and a strategy document that ties it together with priorities and a roadmap. The complete list lives in [the Strategy Sprint deliverables](/services/strategy-sprint). 3. 3 Retainer months · optional ### Production cycles Agents draft content from your templates, dossier, and semantic targets; a validation gate checks every draft against your dossier, guidelines, and tone of voice; the strategist reviews what passes before it reaches you; pages ship, get measured, get optimized. A monthly report shows movement against the baselines set in the Sprint. 4. 4 Final weeks ### Transfer Accounts move, documentation lands, and a handover session walks your team through running the system. The engagement ends; the system stays. The operating model ## What makes the model structurally different. ### One strategist orchestrates Every engagement is led by one senior strategist who defines priorities, reviews every deliverable, and holds the client relationship. You never explain your business twice. ### AI agents execute Semantic research, SERP analysis, content production, optimization, reporting: the work that traditionally requires a team. The agents run in parallel, not in sequential hand-offs; that is why the timeline reads in weeks. ### Delivered in weeks The Sprint produces the full strategy package in 3–4 weeks. Not quarters, and not a retainer minimum before you see anything substantive. ### Transferred at the end Every agent, every document, every script moves to your account. The section below spells out exactly what that means. The alternatives ## Four ways to staff marketing. Full-time CMO Agency Solo consultant / fractional CMO AI Growth Marketing Team Operating capacity One senior hire, team comes later 5–20 people, shared across accounts One person, 20–40 hrs/mo ✓One strategist + an AI agent team Time to first strategic deliverable 6 months to hire, 12 to impact Quarters Weeks for advice; output is on you ✓Full strategy in 3–4 weeks Execution depth Depends on team budget Deep, but vendor-operated Strategy plus light execution ✓Strategy plus full execution What remains when it ends The hire, if they stay The vendor leaves with the stack A deck and advice ✓The configured system, under your account Commitment Permanent payroll 6–12 month retainers, exit fees common Monthly renewals ✓Fixed Sprint; optional 6-month retainer ### Full-time CMO Operating capacity One senior hire, team comes later Time to first strategic deliverable 6 months to hire, 12 to impact Execution depth Depends on team budget What remains when it ends The hire, if they stay Commitment Permanent payroll ### Agency Operating capacity 5–20 people, shared across accounts Time to first strategic deliverable Quarters Execution depth Deep, but vendor-operated What remains when it ends The vendor leaves with the stack Commitment 6–12 month retainers, exit fees common ### Solo consultant / fractional CMO Operating capacity One person, 20–40 hrs/mo Time to first strategic deliverable Weeks for advice; output is on you Execution depth Strategy plus light execution What remains when it ends A deck and advice Commitment Monthly renewals ### AI Growth Marketing Team Operating capacity One strategist + an AI agent team Time to first strategic deliverable Full strategy in 3–4 weeks Execution depth Strategy plus full execution What remains when it ends The configured system, under your account Commitment Fixed Sprint; optional 6-month retainer All engagements are fixed-price and quoted at discovery, once scope is clear. The architecture, labeled Client materials flow through the agent topology; every draft passes the automated validation gate, then strategist review, before it reaches you. Two filters: one automated, one human. - ### Research agent Crawls your sitemap, competitors, and review platforms; collects and merges keyword data; clusters it by intent and commercial value. - ### Content production agent Drafts pages and articles from your templates, dossier, and semantic targets. Never from a generic prompt. - ### Optimization agent Compares each live page against the current SERP and produces specific, prioritized recommendations. - ### Reporting agent Assembles the monthly performance report: positions, traffic, conversions against baseline. Every box on this diagram transfers to you at engagement end. On the matter of ownership ## What you own when it ends. The transfer is contractual, not a courtesy. At engagement end you own: Stop paying and keep working. - The company dossier and every research and analysis document produced - The configured Claude Code setup, with the CLAUDE.md your team uses to run production - The semantic core: raw CSV data and the clustered, prioritized analysis - The automation scripts: keyword collection, data merging, reporting - All content produced during the engagement - The repository and drive folder holding all of it, under your account What makes the transfer worth having is not the code. It is the knowledge base compiled into the agents: your dossier, your proof points, your customers’ language, your semantic data, turned into system instructions. The same scripts without it are an empty editor. The transfer carries a 90-day support window: if an API change or a model update breaks a script or an agent in that window, we fix it at no cost. After that, fixes and extensions are fixed-scope [Automation Advisory](/services/automation-advisory) projects. Or renew, because output quality is higher with the original strategist involved. Your call. [How the retainer handles transfer →](/services/ai-growth-marketing-team) The origin ## Built out of necessity, then productized. This is not theory. Between 2022 and 2026 the practice’s lead operator held three to four concurrent Marketing Director and CMO mandates at once, across fintech, hosting, AI products, and infrastructure. Nobody runs four mandates the traditional way; the AI-augmented operating model was built because the work demanded it, then formalized into the system this page describes. THE·HINE is that system, productized. [About the practice →](/about) 1. 2022 First concurrent mandates: fintech and hosting. 2. 2023–2024 Three to four CMO-level mandates in parallel; the agent layer built out of necessity. 3. 2025 The pipeline codified: 13 phases, a skill artifact, the CLAUDE.md. 4. 2026 Productized as THE·HINE: fixed scope, fixed timeline, full transfer. Selected work ## Where it has run. ### B2B fintech: lending software A 16-month engagement. The pipeline built the organic channel from an underperforming baseline into the primary inbound funnel: semantic core, SERP-driven site architecture, template-based content production. ### B2C infrastructure: hosting Full funnel built from zero: positioning, site architecture, organic channel, analytics stack. Grew into the primary customer acquisition channel over multi-year operation. Specific metrics are shared under NDA during discovery. In practice ## How the engagement works in practice. What exactly do the AI agents produce, and what does the strategist do? Agents handle volume: keyword collection and clustering, SERP analysis drafts, content drafts from templates and the dossier, optimization recommendations, monthly reports. The strategist handles judgment: priorities, interpretation of the data, positioning, and a review of every deliverable before it reaches you. You only ever see strategist-approved work. What happens if the AI gets it wrong? The same thing that happens when any team member gets it wrong: the gates catch it. Agents work from your dossier and your semantic data, not from generic prompts, which removes most error at the source. Every draft then passes an automated validation check against your guidelines, your tone of voice, and the facts in the dossier, and the strategist reviews what survives. The automated gate never gets tired or careless; nothing ships to you unreviewed. How does an AI marketing team work day to day during the retainer? In production cycles. Agents draft against the month’s priorities, the strategist reviews and adjusts, pages ship, and the optimization agent measures them against the live SERP. You get a weekly sync and a monthly report with positions, traffic, and conversions against baseline. Do I need AI expertise to run the system after transfer? No. Nobody on your team needs to open a console or write code: the system takes tasks in plain language, through the tools your team already works in. The transfer includes the `CLAUDE.md` instruction set, written so a marketing generalist can operate the system: how to brief the agents, run production, and read the reports. If your team wants deeper capability, the [Automation Advisory](/services/automation-advisory) engagement covers training and extension. What does “transfer” include, in practice? The configured agents, the instruction set, the semantic core data, the automation scripts, every document produced, and the repository that holds them, all moved under your accounts. Contractually defined in the SOW, not a goodwill gesture. After transfer there are no usage fees and no dependency on us, and a 90-day support window covers any breakage from API or model updates at no cost. Is my data used for other clients? No. The system is configured under your accounts and your materials stay in your environment. Nothing from your engagement, from the dossier to the semantic core to the content, is reused or pooled across clients. Who pays for the AI usage, and what does running the system cost? The system is built in your own Claude Code environment from day one, so model usage runs on your subscription and your API keys, never through us. That is deliberate: no pass-through billing, no markup, no surprise invoice for tokens. We configure budget limits during setup, and in practice the model cost of a full production month is a small fraction of one marketing salary. By appointment ## See if this is the right fit. A 30-minute discovery call. We’ll either point you to the right Sprint or tell you it’s not a match. Honest either way. [Book a call](/contact) [Start with Strategy Sprint details →](/services/strategy-sprint) --- Source: https://thehine.ai/how-it-works Last updated: 2026-06-10