FRFabian Roser
AI TRANSFORMATION FOR COMPANIES & AGENCIES

Turning the skill of a few into a model the whole company carries.

I help companies, marketing organisations and agencies turn AI transformation into real strategic strength. From Hamburg, across DACH and beyond.

AI transformationCompanies & agenciesAdoptionHamburg · DACH
Fabian Roser in an enterprise workshop with a leadership team
Selected agency references
Jung von MattOgilvyDDBantoniDAMMANNWORKSPhilipp und Keuntjewob
Brands I've worked with
dm-drogerie marktDHLJägermeisterNetto Marken-DiscountGebr. HeinemannDeutsche TelekomAudiLufthansa
01The problem

Almost everyone uses AI. Almost no one has it under control.

In most agencies, a handful of people are genuinely good with AI. That's real and valuable. But it isn't an operating model. The know-how sits on islands, and islands don't scale.

What “advanced” actually means.

What usually runs today
  • Drafting copy. A chatbot delivers a first version, someone rewrites it.
  • Images, moodboards, motion. A model produces material for the presentation.
  • Shortening long documents. Studies, minutes and reports in brief.
  • Translating and proofreading. Faster than before. Otherwise as before.

Single tasks. Everyone for themselves, every time from scratch.

When whole processes are translated
  • Scattered communication becomes a brief. What sits across weeks of mails, attachments and notes turns into a clean brief. Whatever is missing comes back as a ready-to-send follow-up email.
  • The brief becomes campaign routes. Strategy, creative brief and several routes with insight, idea, headline and image emerge in one run. You pick and sharpen: the chain supplies the drafts, the judgement stays with you.
  • The approved idea becomes every asset. Every channel, every format, every language in final quality. What used to be weeks of adaptation runs through while you're already on the next thing.

Three steps, one chain. What comes out of one is what goes into the next.

And you sit at every handover: what moves on, what gets sharpened, what goes back to the start. The chain takes the busywork, not the judgement. The technical term is agentic workflows. In practice it's the difference between using AI and working with AI.

ISLAND KNOWLEDGEOPERATING MODEL
02The solution

Impact needs both: system and adoption.

System×Adoption=ImpactSystem built. Adoption anchored.

Most of the market covers only one half: some train without building a system, others build systems nobody uses. I do both — build the system and anchor it — until the whole organisation carries it. And because nobody has time for that alongside the day job, I do it with you.

What I'm working towards with you.

Island knowledge becomes an operating model.What depends on a handful of people today gets carried by the whole organisation. Repetitive work runs as a workflow instead of by hand.
You can offer what your clients are asking for.AI capability now sits at the top of what clients expect from their agencies. Sell it as a service, and you don't have to hand your own efficiency back as a discount in the next pitch.
People come along, and you can see it.No rollout against the team, but with it. Without fear, with more room for the idea. And adoption isn't a mood — it's something you can measure.

He is one of those who do, who kick things off. A creative leader with a digital heart, always fully committed.

Christoph Pietsch · Chief Growth Officer, Publicis Groupe Germany
03Working together

Let's begin.

Four building blocks that build on each other and continue if you want them to.

01
Block 1 · Entry

Day sprint

1 day

A lean way to get to know each other.

  • Where does the island knowledge sit, and where does adoption stall?
  • Which repetitive work eats margin and could be rebuilt as a workflow?
  • At the end: three operational levers.
1 dayLever 01Lever 02Lever 03Three operational levers, prioritised.
02
Block 2 · Core

The 30-day sprint

4 weeks

Away from island knowledge, toward an operating model.

  • Your key processes made tangible as AI workflows, with the first prototypes built by me
  • A use-case portfolio you can actually decide on, plus a governance baseline
  • A 90-day adoption plan with checkpoints that brings the whole team along, not just the few who can
Top 3–5FeasibilityMargin impact
03
Block 3 · Ongoing

Ongoing advisory

ongoing

An experienced pair of hands behind the business.

  • Fixed sessions each month, plus a quarterly look outward: what is new, what is ready, and what you can safely ignore
  • When it helps, I keep building and implementing myself
  • Keeps adoption moving, makes it measurable and keeps the team on board
2 sessions / monthQuarterly radarA steady rhythm over time.
04
Block 4 · Leadership

AI for executives

1:1 coaching

On your level, confidential, built around your calendar.

  • Build and extend your own AI capability, without having to work it all out on the side
  • Company-wide rollout: enablement and change management
  • Because adoption carries much further across the company when leadership goes first
1:1Your own AI capabilityRollout & enablementChange management

Common questions

Who is the AI consulting for?
For companies, marketing organisations and agencies, small to large: from the marketing department of a mid-sized company to advertising, media, digital and corporate-design agencies. Decision-makers at management and leadership level.
How is AI changing the agency business?
AI takes over much of the repetitive work many agencies build their utilisation on. The billable-hours model comes under pressure, job profiles and ways of working change. Value shifts from execution to idea, strategy and adoption. Those who shape it early gain room to move.
What if we already use AI?
I believe you, for a handful of people it's surely true. The question is whether that's an operating model or island knowledge held by individuals. That's exactly where my work begins: turning the skill of a few into something the whole agency carries.
Does this also apply to media agencies?
Yes. Media agencies in particular are under pressure, because platform AI like Performance Max and Advantage+ automates planning and buying. Value shifts to strategy, orchestration and control. A solid AI operating model can be built for that too.
What does AI consulting cost for a mid-sized company?
It depends on size, goals and depth. The entry is deliberately lean, a day sprint. We clarify the concrete frame in a first conversation, transparently and with no open end.
How do we start with AI: is a workshop day enough, or do we need a sprint?
Both have their place. A workshop day creates clarity about status and levers. The 30-day sprint turns that into a decidable portfolio with an adoption plan. That's why the path here starts with a day sprint: afterwards you know whether and how to continue.
Do you also offer AI training for companies?
Yes, as part of the path: enablement formats belong to the sprint and the ongoing advisory, and the AI-literacy requirement under Article 4 of the EU AI Act is covered as part of the groundwork. Blanket training with no system behind it is something I deliberately don't do.
How long does an AI transformation take?
The first tangible progress comes in the 30-day sprint: a decidable use-case portfolio, a governance baseline and a 90-day adoption plan. Real adoption across the organisation is a movement over months, which I accompany on an ongoing basis if you wish.
What about data protection and the EU AI Act?
Both belong in the groundwork. I build tool policies and governance to be GDPR-compliant and account for the AI-literacy requirement under Article 4 of the EU AI Act. In the DACH region, data sovereignty is an advantage when you shape it actively.
Does AI take away our creativity?
The opposite. AI takes the busywork. What remains is exactly what you are paid for: ideas, judgement and a point of view. I strengthen creativity instead of replacing it, and take the fear of change away from teams.
What are you not the right person for?
Prompting courses. I don't take it as given that everyone in the building needs to be able to prompt, and certainly not that it's the decisive future skill: it's a craft that shifts with every model release. And I'm not the one to push your creative tooling forward or tell you whether model X or Y is currently the better one. That question is answered differently six months later, and it was rarely the one things depended on. What interests me is what sits underneath: which processes you run again and again, how they should run in future, how roles should change for that, and how people and machines work better together.
What sets you apart from a strategy consultancy?
A consultancy writes up what I've done. I ran an agency with full P&L myself, owned the creative work and still build my own AI products today. Since 2024 I've been a Certified AI Officer of the German digital industry association BVDW. I don't deliver a slide-deck concept, but method and a path to adoption, operational and honest.

You don't have to know where you stand to talk. Talking is how you find out.

Book a call