According to the ifo Institute (June 2026), 54.5 % of German companies are already using AI in their business processes – in 2023 that figure was still 13.3 %. For tradespeople and service providers in Upper Bavaria, this shift is now arriving in a very concrete form: not as an abstract software promise, but as a mobile app with an AI agent running in the background, keeping appointments, enquiries and customer data in sync in real time.
That might sound like a large IT project. It doesn't have to be – if you set it up the right way.
Most apps built for tradespeople solve exactly one problem: they display appointments or manage jobs. But as soon as a customer calls, an email comes in, the calendar changes and a field worker sends back a status update all at the same time – the system falls apart. Someone in the office has to glue everything together manually. Usually the owner themselves.
Softstorm observation: In 7 out of 9 of our trades projects, this was the very first bottleneck – not the technology, but the manual reconciliation between incoming channels, the calendar and the CRM. Automating that with an AI agent wins back between 4 and 8 hours per week.
Our view on this is clear: Most digitalisation consultants sell trades businesses a level of complexity that nobody running a team of 8 people actually needs. A lean AI agent that reliably does three things beats any bloated ERP system.
An AI agent solution intercepts incoming enquiries – whether via a contact form, email or WhatsApp – and classifies them straight away: emergency or standard request? Existing customer or new contact? Is there already a quote on file?
For this workflow, Softstorm uses n8n as the automation layer. Compared to Zapier or Make, n8n is considerably more flexible when it comes to complex branching logic and can be self-hosted – which matters for businesses that don't want their customer data sitting in US-based cloud systems. The AI agent in the background, built on Anthropic's Claude API, reads the enquiry, extracts the relevant information (type of job, urgency, address) and enters it directly into the mobile app.
Real-world example: A plumbing and heating company from Wolfratshausen used exactly this setup to double their daily enquiry throughput – from around 15 to 30 enquiries per day – without hiring any additional office staff. New customer enquiries are now automatically assigned to whichever technician is free in the nearest service area. The error rate from manual data entry has dropped to zero.
The second major problem: customer data lives in five different places at once. In the calendar, in the email thread, in a spreadsheet and somewhere on a notepad. The mobile app then shows an outdated address or the wrong contact person.
This is where Supabase comes in. Supabase is an open-source alternative to Firebase and acts as a central real-time database. All systems – the mobile app, the booking form, the email integration – read from and write to a single source of truth. Changes are immediately visible on all devices, with no manual reconciliation. Unlike proprietary database solutions, Supabase can be operated in a way that is in line with GDPR requirements and connects seamlessly with n8n workflows.
An AI chatbot with access to customer data from Supabase can also pre-fill follow-up enquiries automatically: when Mrs Maier from Holzkirchen calls for the third time, the system already knows which appliance she has, when a technician last visited and what was in the last quote.
The third approach is the one that causes the most day-to-day headaches: rescheduling. A job runs over, a team member calls in sick, a customer cancels at the last minute. That means three follow-on appointments need to be rearranged – manually, by phone, with a lot of back and forth.
An AI agent with access to the calendar that can check availability in real time takes over this task entirely. It automatically sends the customer new appointment suggestions (by SMS or email), waits for confirmation and updates everyone involved. In practice, this works reliably even for businesses with 5–6 field staff – as we demonstrated in our project on AI-powered quote generation in the trades sector.
We are increasingly extending this use case with Voice AI using tools like Vapi: customers can confirm or decline appointment changes via an automated phone call – no app, no form, just the telephone. Particularly for older customers in rural areas around Bad Tölz or Miesbach, that's a genuine advantage.
Here's the honest part: AI agents running behind a mobile app are not a magic bullet. They reliably fall short when enquiries become legally or contractually complex – for example, in disputes about warranty obligations, when job scope is unclear or when a customer later questions what was agreed. In those cases, you still need a person to make and document the decision.
There's also this: the setup only works as well as the data that goes into it. If you've been working with a messy spreadsheet for years, you need a clean data foundation first – before an agent can automate anything in a meaningful way. That's not a technical problem, but it's a real stumbling block at the start that is often underestimated during planning.
According to a case study by Salewski.IT (2026), a trades business with 12 employees managed to double enquiry throughput from 20 to 40 per day and increase new customer acquisition by 25 % using a structured lead-to-offer workflow agent – but only after an initial data migration and a clear definition of processes. Without that groundwork, the agent would have been running on empty.
If you want to know which of these three approaches makes sense for your business in Upper Bavaria first – and what the realistic effort involved looks like: Book a free initial consultation. We're based in Bad Tölz-Wolfratshausen, we know the region, and we'll give you a straight answer about what's feasible and what isn't.
We analyze your processes, identify automation potential and show you which digital systems really make sense.
