What can be automated
Connecting systems
When an order arrives, a document lands or a status changes, the next step runs by itself, across shop, accounting and email. Built with n8n, running on your infrastructure or hosted in the EU. If the goal is data that stays correct between shop, ERP and PIM, the e-commerce integrations page goes further.
Quotes & follow-up
Quote goes out, the reminder runs by itself. No quote is left unanswered because daily business got in the way.
Documents & invoices
Invoices and delivery notes go from the inbox straight into accounting. The AI reads the contents, you only review.
Product data
Supplier lists into the shop without Excel afternoons. Prices, stock and copy update themselves.
How Getting Started Works
Four steps, each with a clear price. You know what the next one costs before it begins.
-
Conversation
€0Free, 30 minutes, no obligation. You describe the daily routine, I tell you honestly whether automation is worth it.
-
AI Fitness Check
€890Half a day on site or remote, plus a written assessment. Credited in full against a sprint booked within 30 days.
-
Build Sprint
€4,900Two weeks, fixed price. At the end exactly one automation runs in your live operation.
-
Operation
monthlyMonitoring, adjustment and extension of the automations already running.
Why not just use AI?
Much of it doesn’t need AI at all, it needs a clean connection between systems that don’t talk to each other today. AI joins in where messy input has to become structure. I don’t sell anything I don’t run myself.
In conversations with business owners lately I hear the same wish almost every time: AI should write the quotes, AI should sort the receipts, AI should keep an eye on stock. Understandable. AI is a wonderful tool, and it lets people without a technical background do things that used to be out of reach.
But a language model is, at its core, a probability engine. It may interpret the same process a little differently next time. For a price list, a booking or a stock level, that is the wrong tool. Those processes need deterministic automation: a script or a workflow that does exactly the same thing today, tomorrow and a year from now.
Where AI is genuinely strong is elsewhere. It finds the connections in messy, human material: half an idea in a voice note, an email with three requests in it, a delivery note with a coffee stain. It turns unstructured thinking into something a system can work with. And the tooling around it, which people often mistake for AI itself, has taken a huge leap in a short time: n8n, APIs, MCP servers that connect models to real systems. This page, for what it’s worth, came about exactly that way, from spoken thoughts rather than a spec.
The craft is in the combination: AI where mess has to become structure, deterministic workflows where exactly the same thing has to happen every time. That is exactly what the AI Fitness Check settles: which of your routines needs AI, which needs a clean workflow, and which needs both. The Build Sprint then delivers what the check found.
Reports That Arrive on Their Own
A report is an automation like any other. The numbers sit in the shop, in the accounts and in the ad platforms, somebody copies them into a spreadsheet once a week, and by the end of the month no two versions agree. I collect them where they are produced and put them in one place.
What that becomes depends on the question: a dashboard, a report that lands in your inbox every Monday, or a single analysis for one decision. For Bodylab24 it was conversion analytics.
One Person, From the Question to the Code
I work out what is needed, write the requirement, organise whoever else works on it, and then build it myself. Usually that is three companies, and something is lost at every handover between them. Here there is nobody between your question and the running system.
What Work Comes Back Every Week?
Describe one routine that costs you time regularly. I will tell you whether it can be automated and roughly what that means.
this page as a PDF to pass along