AI Empowers · Sign & large-format print supply (Benelux)
An AI support agent for Nautasign's technicians
Technicians ask a question in plain language. An AI agent searches the supplier manuals and proposes an analysis.
- Print & signage
- AI agents
- Document intelligence
- Private chatbot
The problem
Technicians lost time searching supplier manuals, FAQs and old emails to solve customer problems. The knowledge sat in many places and in people's heads.
The solution
A Copilot Studio agent with the service and user manuals indexed in Azure AI Search. It reads supplier, model and error code from a question, finds the right passages and writes a first analysis.
The outcome
Technicians find answers to technical problems faster. Technical and commercial users were onboarded, and the success led to a second phase.
The full story
Where it started
Nautasign supplies the sign and large-format print industry in the Benelux. Its technicians support customers on equipment from many suppliers, and the answers were spread over supplier manuals, internal FAQs and email threads.
Strategy first
In a Data & AI strategy session in March 2026, technicians, sales, purchasing and finance listed nine possible use cases, from order intake to churn analysis. The advice: automate repetitive work first, start with a few use cases with clear impact and give each an internal owner. The knowledge agent for technicians came out on top.
How it works
We compared NotebookLM, Copilot Studio with SharePoint, a custom Azure OpenAI RAG build and Claude, and chose Copilot Studio because it fits Nautasign's Microsoft environment. Nautasign gathered the manuals and example questions with the expected answers. The manuals are synced from SharePoint and indexed in Azure AI Search, which was added because the documents were large and complex. The agent first pulls supplier, model and error code from a question, so the search stays on the right manuals, then returns up to three relevant manuals and an analysis. The technician stays responsible for the answer that goes to the customer.
What's next
In the second phase the agent joins the Odoo helpdesk: questions from the support mailbox and the webshop become tickets with supplier, model, error code, manuals and a first analysis already attached. If the agent fails, a basic ticket is still created, so nothing gets lost.
Tech
Ask Pai
Ask Pai about this case
Pai is an AI. It only knows what is written on this page.
Pai is an AI. Answers come from this case page. Don't share sensitive data.
Chats are kept for 30 days unless you switch that off in Privacy and choices.