AI that works with your data and systems, not just a chat window.
Price from €3,000 for a pilot project
In short
We build AI agents, chatbots on your documents (RAG) and connections between ChatGPT or Claude and your systems (MCP). A pilot project starts at €3,000 and usually takes three to six weeks. We start with one task that takes people’s time today, and measure it before expanding.
Most companies don’t need an “AI strategy”; they need one manual task fewer. So we look at the process first. Sometimes the answer is plain automation without AI, and then we tell you so.
What it includes
Answers customers around the clock using your prices, terms and stock. When a question is outside its data, it hands the conversation to a person and records the enquiry.
Good for
online stores, hotels, services with many repeated questions.
Staff ask in plain language and get an answer with a reference to the document it comes from. Contracts, procedures, technical documentation, internal rules.
Good for
companies with many documents, law firms, manufacturing, support teams.
We build MCP servers through which ChatGPT, Claude and your own agents work with your CRM, ERP, databases and ad accounts. With role-based permissions and a log of every action.
Good for
teams that already use ChatGPT or Claude and want it to see the real data.
They handle incoming enquiries, prepare quotes, read invoices and move data between systems. Important steps go through human approval.
Good for
offices with repetitive manual work on emails and documents.
If the task is the same every time, plain automation is cheaper and more reliable. AI makes sense when the input is different every time: free text, emails, documents, customer questions. Our rule is to start with the simplest solution that does the job.
AI agents and automation: pricing
from €3,000
The €3,000 starting price is for a pilot with one clearly defined task, for example a customer chatbot or AI on one folder of documents. Use of the AI model is paid separately, directly to the provider, like an ad budget. We give an estimate for it in the quote.
What affects the price
We talk to the people who do the task today and agree how success is measured: hours saved, response time, errors.
A working prototype on your real documents or data, which you try yourself.
Your team uses it in daily work. We measure accuracy and time saved.
We roll it out to everyone and add the next task if the first one has paid for itself.
Technologies
Describe it in two sentences. Within 24 hours we will tell you whether AI makes sense and what it would cost.
Other services
