Grundstein AI

Inspiration

Small trade businesses in Germany are the backbone of the economy — over 567,000 craft firms, many with just 3 to 15 employees. Yet these businesses are trapped in an operational nightmare. Their daily workflows run across Excel spreadsheets, WhatsApp groups, paper time sheets, phone calls, and disconnected invoicing tools. Critical information lives in the owner's head, and 5 to 10 hours per week are lost to manual admin work alone.

We conducted 12 in-depth interviews with electrical craft business owners and office admins in South Germany, and the pattern was striking: they don't reject software — they reject rigid systems with too many fields, too much setup, and workflows that don't reflect how they actually operate. 70% of small craft businesses still run on Excel and paper because no tailored ERP solution exists for them.

Traditional ERP systems like SAP or Microsoft Dynamics are built for enterprises — they cost thousands, take months to implement, and require consultants. Low-code platforms like Ninox offer flexibility but still demand technical configuration skills that a hands-on electrician simply doesn't have. We saw a clear gap: high personalization at low cost, with zero technical barrier. That's what inspired Grundstein AI.

What it does

Grundstein AI is an AI-native, prompt-to-ERP platform that lets small craft business owners set up a fully personalized ERP system through a simple natural language conversation — live in minutes, not months.

The owner describes their business (team size, workflows, customer types, tools they use), and the AI configures a complete operational system tailored to their specific needs. No consultants, no IT team, no coding, no training required.

The platform covers five interconnected modules:

Jobs, Customers & Quotes — managing the full lifecycle from customer inquiry to completed project
Team Planning & Scheduling — replacing the owner's mental calendar and WhatsApp coordination
Time Tracking & Field Reports — crews log hours, materials, and job status directly from their phones on site
Invoicing & DATEV Export — auto-generate invoices from tracked work and export directly to the German tax advisor
Dashboard — a single morning view of job status, revenue, team availability, and alerts

Everything connects in one operational loop: customer calls in, a quote is created, the crew is scheduled, hours and materials are logged on site, an invoice is generated automatically, and the owner sees it all in one place. No more Excel. No more WhatsApp chaos. No more paper.

How we built it

Our process was research-first, build-second. We started with 12 structured customer discovery interviews with owners and office admins of small electrical businesses (3–14 employees) across South Germany. We mapped their end-to-end workflows from lead to payment, identified where information breaks across tools, and prioritized pain points by frequency and severity.

From these insights, we defined our Ideal Customer Profile (owner-managed electrical craft businesses, 3–15 employees, DACH region) and scoped our MVP feature set around the core operational loop that every interview confirmed: jobs, scheduling, time tracking, and invoicing.

On the technical side, we built a prompt-based onboarding system where the AI asks targeted questions about the business and configures the ERP accordingly. The tech stack includes a database schema with core system objects (jobs, customers, teams, invoices), an authentication layer, and an n8n workflow engine for automation. We developed a natural language to SQL pipeline so users can interact with their data conversationally, and built a working demo that showcases the full onboarding chat flow with a live dashboard and automated invoice generation with DATEV export capability.

We also conducted a thorough competitive analysis, mapping the market across three tiers — generalist cloud ERPs (Weclapp, Billbee), legacy vertical software (KWP, Streit, PDS), and emerging AI/low-code tools (Ninox, ianaERP) — to validate that no established prompt-to-ERP solution exists for craft businesses.

Challenges we ran into

The biggest challenge was scoping. Craft businesses have an enormous range of operational needs — from job management to material tracking to payroll to fleet management. Every interview surfaced new pain points. We had to discipline ourselves to focus on the core operational loop (job to invoice) rather than trying to solve everything at once.

Understanding the real user was another challenge. Our target customers are not tech-savvy founders or developers — they are electricians, plumbers, and construction workers who spend their days on job sites. Designing an interface and onboarding flow that feels natural to someone who has never used an ERP system required us to rethink assumptions we brought from our tech and consulting backgrounds.

The German compliance landscape added complexity. DATEV integration, XRechnung format for public sector invoicing, and audit-ready data exports are not optional features for our market — they are baseline requirements. Building with compliance in mind from day one shaped our architecture decisions significantly.

Finally, validating willingness to pay was tricky. Interview responses ranged from €40 to €250 per month — a wide spread. We learned that value perception depends heavily on how many manual hours the tool replaces, which pushed us to lead with time savings rather than feature lists in our positioning.

Accomplishments that we're proud of

We validated a real market gap with primary research. 12 interviews with real business owners confirmed that the problem is not awareness of software — it is the absence of software that fits. That distinction shaped everything.

We built a working demo of the prompt-to-ERP onboarding flow. A user can describe their business in a conversation and see a configured system come to life. This is not a mockup — it is a functional prototype that demonstrates the core value proposition live.

We identified a clear competitive whitespace. Our market positioning analysis showed that existing solutions force a trade-off between affordability and personalization. Grundstein AI is the only concept that combines both — high personalization at low cost through AI-driven self-service setup.

We assembled a team with genuine domain expertise. One of our five team members is a trained electrician with three years of hands-on industry experience. This is not a team guessing about user needs from a business school classroom — we have someone who has lived the problem.

What we learned

The most important lesson: small business owners do not want more features — they want less complexity. Every interview reinforced this. The winning product is not the one with the longest feature list but the one that feels like it was built specifically for their 12-person electrical shop.

We learned that the buyer and the user are often different people. The owner makes the purchase decision, but the office admin champions it daily, and the field technicians need to actually use it on site. A product that only appeals to one of these personas will fail.

We discovered that "no-code" is not enough — our market needs "no-UI." Platforms like Ninox eliminated coding but still require users to understand data models and configuration logic. For our target customer, even that is too much. The real innovation is conversational setup where the system asks the questions and the user just answers.

We also learned that the German craft market, despite being traditional and slow to adopt technology, is at an inflection point. 13% already use AI for automation and 38% plan to. The timing is right — but the solution needs to meet them where they are.

What's next for Grundstein AI

Our roadmap follows three clear steps:

Sharpen our ICP — identify which craft segments convert fastest and map decision-makers, buying triggers, and deal-breakers
Build out the MVP — complete scheduling, invoicing, time tracking, mobile access for field teams, and full DATEV export integration
Onboard pilot customers — 3 to 5 businesses from our interview contacts, measuring onboarding time, daily usage, and time saved

Beyond the university project, we see Grundstein AI as a viable business. The TAM for ERP solutions for small craft businesses in the DACH region is €1.2 billion, with a serviceable market of €100 million focused on small electrical craft businesses alone. Our year-one target is 750 customers at €4,000 ARR, reaching a €3 million SOM. The path from university project to launched product is clear — and we intend to walk it.