Your marketing now has to convince two:
the person, and the AI that advises them.
I design, build and verify the marketing and sales systems behind your growth, and I make sure they work.
Plenty of tools, a few automations, but no idea which of them bring in customers …
Maybe this sounds familiar: The CRM is set up. Some automations run. Maybe you already use AI, maybe not yet.
And yet … the results look much like before.
With AI it is often no different: most companies now use AI in some form, but only a small minority see a real financial return from it.Source: McKinsey, The State of AI in 2026: about 6% of companies attribute at least 5% of their operating profit (EBIT) to AI.
The gap is rarely the tool. It sits in four places:
No system around the tools.
Automations were built one by one, with no view of how leads, offers, data and people connect.
No clear goal.
Nobody decided what the system should achieve, so nobody can tell whether it does.
No one watching.
Many decisions are made silently:
- Which ads an algorithm shows to whom.
- Which lead gets answered first.
- How Google's AI describes your business.
If nobody checks, you lose money without noticing.
Many inquiries, few customers.
If nobody measures lead quality, you pay for leads that never buy.
Setting up tools, with or without AI, is easy now. Making them pay off takes a system, and someone who makes sure it runs successfully.
Strategy → Build → Verify → Operate
One accountable person through the whole sequence. No handoffs, no lost context, no "that's not my part."
Strategy: the System Blueprint
Before anything is built, we decide what is worth building.
- Goals and success criteria, measured from where you stand today
- Customer journey and funnel architecture: every touchpoint tied to a clear conversion goal
- What makes a good lead for you: clear criteria that marketing, sales and every AI are measured against
- Data flows and decision logic, including the unwritten rules of your business. I also talk to the people who speak with your customers every day, because they know where leads get lost.
- Where AI agents act: what they may decide on their own, up to which amount, and how you can trace who did what
- Risk analysis: what can go wrong, and what you should therefore not build
Build
Implementation of the blueprint on the tools that fit your business, not the other way around.
- GoHighLevel, Make, n8n, HubSpot, Brevo, Kommo, own automation scripts and AI agents
- Funnels, landing pages, lead magnets, automated follow-ups with behavior-based sequences
- Measurement from day one: which signals go to your ad platforms, and whether they reflect real revenue
- Direct response copy: sales letters, email sequences, ad copy, written to move people to a decision
- Creative strategy for Meta and Google ads: which angles, hooks and audiences get tested and scaled, not just ad management
Verify
With or without AI: every system I build gets checked before it helps decide your revenue. The quiet failures nobody notices come first:
- Leads and follow-up: does every inquiry reach the right place, and does none get lost? Do your automations (for example email sequences, booking, lead routing) run the way they should?
- Signals to your ad platforms and lead quality: is the algorithm optimizing for inquiries that become customers, or for worthless clicks, form fills and vanity metrics?
- AI answers about you: how do Google's AI Overviews and AI Mode, then ChatGPT and Perplexity, describe your business, is it accurate, and does it cost you customers?
- If you run chatbots or voice agents, I test them under pressure (red teaming): can they be talked into discounts, do they leak internal data, or do they claim to be human? I test this across whole chats, as inventively as a determined customer or attacker, in German, English and Spanish. And I check whether they push hesitant customers with persuasion instead of answering honestly. Weaknesses often appear in the language nobody tested.
- If AI agents act on their own, such as an AI setter booking calls or an agent sending quotes or discounts: guardrails with budget limits, approval steps, rules and alerts
- Every finding in plain language, with its business impact and the fix
Since August 2, 2026, the EU AI Act (Article 50) requires that people can tell when they are dealing with an AI system. I check whether your chatbot still complies when someone deliberately pushes.
Operate
A system only pays off if someone keeps it on track. Offers change, prices change, tools and AI models get updated.
- Regular monitoring, evaluation and re-testing of your automations and AI decisions
- One fixed monthly report with the key KPIs: cost per new customer (CAC) and customer value per channel, plus lead quality, conversion and ROI, measured against the goals from your blueprint
- Continuous improvement based on what your data shows
- One accountable person, not a ticket queue
Not sure where to start? Start with a System Blueprint.
Before anything is built, we decide what is worth building. In the blueprint, I map how your business really works today, including the rules nobody wrote down, work backward from your revenue goal, check whether your goals are realistic and measurable, and decide what should be built and what shouldn't.
You receive:
- A written blueprint any specialist can build from
- Goals with a viability check: what the system can deliver, and what depends on you
- A check that your inquiries and sales are tracked at all
- A measured baseline of your current system and at least one tested finding
- A phase plan with dependencies and acceptance criteria
- A test plan for your AI components
- A binding quote for the build
How it works: free Blueprint call → short questionnaire → one mapping session → blueprint draft → one review round → handover.
The blueprint is yours to keep, whether or not we continue together.
Your next customer may ask an AI first.
Many buyers now see Google's AI answer before they ever visit your website, and some ask ChatGPT, Claude or a shopping assistant.
Since August 2026, ChatGPT shows ads in Germany, Austria and Spain.
I check how these systems present your business today, and I build for both audiences: the person who reads your offer, and the AI that decides whether to show it to them.
Why this combination is rare.
Most providers cover one part of the sequence and leave a gap somewhere else. The ad agency optimizes the ad account, the CRM consultant the CRM, the SEO agency the rankings. Each holds one part. Nobody owns the chain.
Doing it yourself with AI.
ChatGPT or the AI in your CRM puts a workflow together for you in minutes. Whether it fits your business, it cannot know, because it doesn't know your unwritten rules. And generic AI answers about tools sound good and right, but are often wrong.
AI agencies with fast setup.
A chatbot in a week, a voice agent at a flat price. Setting it up is quick. Testing, measuring and looking after it rarely follow.
“The platform optimizes itself.”
Meta and Google steer budget and audiences automatically, based on signals. If those signals are good, that works. But if the algorithm optimizes for worthless inquiries, it becomes very efficient at it. The result? More and more worthless inquiries.
Strategists who don't build, builders who skip strategy.
One delivers a plan nobody builds. The other a clean setup that doesn't convert.
Automated tests and certificates.
A scan checks a checklist. I test with my own, constantly growing library of pressure and persuasion techniques, tailored to your business. That finds the weaknesses no checklist anticipates.
What I bring instead:
- The whole system first. I see how strategy, tools, data and people connect before building any single part.
- Behavioral depth. 19 years of sales psychology in practice, and a published book on cognitive biases and how intelligent systems make predictable, exploitable errors.
- Verified work with AI. I use AI in my own work and check every claim before it reaches a client. From my own records and thousands of iterations with AI, I know where generic AI answers get it wrong. Ask an AI “are you sure?” and it often defends a wrong answer with even more numbers. That's why I check against sources, not in the chat.
- Three languages. Native German, fluent English and Spanish. Clients in DACH and Latin America. No blind spots in the languages your customers speak.
- Tested AI expertise. After passing 11+ competitive assessments, contracted by Mercor for LLM evaluation and red teaming.
Results that are specific, not approximate.
What a recent client says.
Working with Tobias has definitely changed our digital marketing efforts for the better. His comprehensive CRO/SEO audit delivered clear documentation and video walkthroughs that uncovered numerous opportunities. His digital marketing strategy and revenue projections aligned perfectly with our goals and constraints. … A true full-stack digital marketing partner.
Where is your system underperforming, and where is an AI deciding without anyone checking?
Let me know what you are running and what isn't working the way it should. On the Blueprint call, I'll give you 2 to 3 specific observations you can act on immediately, whether or not we work together.
Book a Free Blueprint Call30 minutes. No pitch deck. A direct look at what's broken and what to fix first.