AI for marketing agencies, used by the whole team
In most agencies, a couple of people are great with AI and everyone else is watching. Here's how I helped one agency get the whole team on board.
The usual agency picture
A lot of agencies look like this. Two people have figured AI out and get a lot out of it. Their setup lives on their own laptops. Everyone else has tried it a couple of times and gone back to doing things by hand.
Meanwhile the agency has plenty of repeatable work: client reports, status updates, approvals, onboarding a new client. All of it done slightly differently by each person.
What I did at one agency
A 25-person marketing agency brought me in. Two teams ran on two different systems. Client reporting was pulled together by hand. The "shared" AI setup was really one person's personal setup, and teammates on Windows couldn't get it running at all.
I built four things:
- A client dashboard. It syncs every hour from the sheets the team already works in, so nobody had to change how they work. Each client sees only their own work, and a "waiting on your review" queue with email alerts stops approvals from sitting. The team adds a new client in under 10 seconds.
- A team AI workspace anyone can install. One command, Mac or Windows, from a fresh laptop to working. API keys get saved safely and never pasted into a chat.
- A safe way to share what works. Someone says "share this with the team". The change is checked automatically, first by fixed checks, then by an AI review. Safe changes go live on their own. Anything risky waits for a person.
- Training until it stuck. Onboarding per team, a weekly check-in, one-to-one help and written best practices.
What changed
- 22 teammates set up and working.
- 16 shared team processes, across sales, client delivery, reporting, content and operations.
- 5+ teammates besides the founder contributing their own.
- About 10 weeks to the first share that was checked and went live with no human involved.
Most of those 16 processes were written by the agency's own people, not by me. My job was making it possible for non-technical people to build their own, share them, and trust them.
We didn't measure hours saved, so I won't quote any.
The people side
Every team has someone who's uneasy about AI: worried about their job, burned by a tool that didn't work, or just tired of change. Alongside the AI work, I'm a mindset coach. I make time for those people, hear out what's actually bothering them, and help them see what AI takes off their plate rather than what it takes away. People who've been heard give a new way of working a real chance.
Where to start at your agency
Pick the piece of client work your team repeats most, write down how your best person does it, and build that first. Then give everyone a way to share what they build. More in How to get a whole team using AI.
Common questions
How can a marketing agency use AI?
Start with repeatable client work like reporting, status updates and approvals. Write down how it's done, build AI to run it, and train the whole team on it.
How do you get the whole agency using AI?
One setup everyone can install, a way to share what each person builds, and training until it sticks. Champions alone keep AI stuck with two people.
Do agency staff need to be technical?
No. At the agency above, most contributors weren't technical. They never saw the plumbing, just their work available to everyone.
How long did it take?
About 10 weeks to the first share that went live with no human involved. The client dashboard was live for the first client about seven weeks in.