B2B Manufacturing
AI Assistant for Heating Service
Vaillant uses an AI assistant to answer technical questions, guide customers to resources and automate service processes across its heating solutions website.
Comprehensive expert knowledge via free-text search
The AI assistant serves as a central hub for all Vaillant knowledge. Thanks to state-of-the-art Agentic AI, the system understands complex queries entered in the free-text field and immediately provides precise answers based on extensive data sources. This gives customers direct access to expert information from subsidy details to technical specifications without having to search for it.

Smart Diagnostics & Troubleshooting
In the event of technical issues, the bot serves as the first point of contact for troubleshooting. By entering error codes, users immediately receive a reliable assessment and peace of mind. Since the AI recognizes when an expert is needed, the user is seamlessly guided to an online appointment booking page. This automates the support process and significantly reduces the workload on the service hotline.

Immediate assistance through digital guidance
The chatbot empowers customers to resolve simple maintenance tasks on their own through step-by-step instructions and deep links. Whether it’s bleeding the radiators or adjusting the room temperature, the AI guides the user to the solution before a support ticket is created. This boosts customer satisfaction by providing immediate assistance while simultaneously reducing after-sales process costs.

The Vaillant Group is one of the world’s leading companies for heating, ventilation and air-conditioning technology. Heat pumps and subsidy programmes increase the information needs of customers, installers and prospects. An AI assistant on the LoyJoy Platform is the central entry point on the website.
Results at a glance
- One entry point for product knowledge, subsidies, maintenance and error codes
- Straight into the process: deep links, contact forms and online appointment booking
- Pre-qualified cases for service instead of open follow-up questions
- Knowledge that grows: searches without results become templates for new articles
Starting point
Customers come to the website looking for concrete answers: technical details, maintenance tips, subsidies or help with an error message. Navigation and static FAQ pages are often not enough. Vaillant wanted to answer these questions instantly and pre-qualify service requests digitally.
How the AI assistant works
Visitors ask their question in their own words. The assistant answers from the knowledge base, product pages and FAQ and, where needed, leads directly into the right process:
- Error codes: users enter the code and receive a well-founded assessment. If a technician is needed, the assistant leads to online appointment booking.
- Self-help: for simple tasks such as bleeding the radiators, there are instructions and deep links before a ticket is even created.
- Handover to people: complex cases reach customer service prepared, via contact form or appointment.
Product finder for the right heat pump
Even before the AI assistant, Vaillant implemented a guided product finder with LoyJoy. It leads customers to the right product, as was previously only possible by talking to customer service. It is especially popular for heat pumps, where advice matters most.

Turning searches into knowledge
The AI search also shows what customers look for without success. The Vaillant team uses searches without results as templates for new articles, so the knowledge base grows along real questions. Besides the consumer offering, Vaillant also uses LoyJoy in FachpartnerNET for installers.
Videos about the case
More customer stories
Which requests will your agents resolve next month?
In 30 minutes, we show you which customer requests your agents can resolve starting next month, and how you measure the result.
New customer stories in our newsletter
Monthly product updates, practical insights and customer stories. Unsubscribe anytime.
Unsubscribe anytime. Details in our Privacy Policy.