AI-run marketing

A whole marketing function, built on AI

An AI-run marketing function does the work of a marketing team (data, tracking, content, campaigns and reporting) with AI agents doing the engineering, analysis and drafting, and a senior marketer setting the direction and approving what goes live. I build them, and I run one myself.

What it covers

Six parts, each with checks and a way back

Each part is what I built in my current role, described in general terms.

Data

A marketing data warehouse

Analytics, Search Console, ad platforms, the CRM, forms and product analytics, loaded automatically into one place on a schedule.

Insight

A weekly briefing

An analysis layer that writes a weekly insights briefing for the leadership team, so decisions rest on the same numbers.

Tracking

Measurement you can trust

Consent-aware, server-side tagging, first- and last-touch attribution passed to the CRM, and duplicate conversions removed.

Content

A content engine

Article ideas ranked from tracked keywords, then research briefs, drafts, quality and risk checks, and publishing through review. Regular internal-linking rounds.

Paid search

Paid search with guardrails

Value-based bidding, with every change made through tested tooling that checks it first, a one-step rollback and daily guardrail checks.

AI visibility

How AI assistants see you

The questions your buyers ask ChatGPT, Claude, Gemini and Perplexity, tracked over time, with machine-written searches separated from real people's.

What stays with a person

AI does the work. A person decides.

Strategy, positioning, spend and final approval stay with me. Nothing that touches budget, the public website or customers goes live without a person approving it.

  • Small, reviewed, tested, reversible. Every change is small enough to check and can be undone.
  • Costs measured. Every AI and data call is metered. The system I run was designed around a running cost of about $1,000 a month.
  • Decisions written down. Each decision is logged with its reasoning, and a log of site changes records anything that affects how the data should be read.
  • Search and AI, both tracked. Buyers' research has been moving from search engines into AI assistants, so the system watches both.

How I build one

From measurement to engines

Measurement first

Tracking you can trust, consent handled properly, and one place for the data.

One view of the numbers

A weekly briefing, so the leadership team decides from the same facts.

Then the engines

Content and paid search, each with checks before anything goes live and a way back if it goes wrong.

Then AI visibility

Tracking how AI assistants answer the questions your buyers ask, and what to change when they get it wrong.

Proof

Six and a half months, one marketer

500+scoped AI-agent tasks
39automated workflows, 29 on a schedule
~14,800automated tests
15+data sources, loaded automatically

Read the case study

Want a marketing function that runs on AI?

Tell me about your team, what you sell and where it gets stuck. A few lines is enough. I'll reply by email.

Get in touch