25+
Fragen
FAQ
Your Questions. Our Answers.
Everything you need to know about AI projects, costs, GDPR and implementation — before you take the first step.
25+
Questions answered
5
Categories
Ø 2 Min.
Read time per answer
Our AI projects start with a fixed price from €2,490 for the first AI pilot. This price covers complete development, integration and commissioning. More complex projects with multiple AI agents or extensive system integration are individually calculated — always as a fixed price, never as an open day rate. After the free initial consultation you receive a binding quote within 48 hours.
Because open day rates represent unacceptable budget risks for SMEs. We know the product we build — which is why we can calculate a fair fixed price. This creates trust, eliminates nasty surprises and enables you to plan your ROI clearly. If the scope changes during the project, we discuss it openly — no hidden scope creep at your expense.
That depends on the chosen infrastructure. External AI APIs (OpenAI, Anthropic etc.) are billed on a usage basis — typically €50-500 per month depending on volume. On-premise solutions have higher upfront investments but no ongoing API costs. For monitoring, maintenance and support we offer optional service packages from €299/month. We present all follow-on costs transparently in the quote.
Most of our projects pay for themselves within 4 to 6 months. The average ROI in the first year is 3.2×. This means: from €2,490 investment comes an average of €7,968 in value in 12 months. For projects with high manual effort (e.g. invoice processing, support), payback can even occur after 8-12 weeks.
That is exactly what we recommend. Our 5-day pilot is designed to start with a clear, defined task — and only expand the project after proven success. This approach minimises your risk, creates quickly visible results and gives your team the opportunity to gradually get used to AI.
We work with the leading AI providers: OpenAI (GPT-4), Anthropic (Claude), Microsoft Azure AI, AWS Bedrock and open-source models like Llama. For particularly data-sensitive applications we operate our own in-house GPU capacity where your data never leaves the company.
Yes — that is one of our core areas. We seamlessly integrate AI into your existing IT landscape: ERP systems (SAP, Microsoft Dynamics, Navision), CRM systems (Salesforce, HubSpot), document management systems, custom-built applications and more.
Classic automation follows fixed rules — if A happens, then do B. This works well for structured, uniform processes. AI, on the other hand, can handle unstructured data (emails, documents, speech), learn from examples and make decisions in variable situations.
Not necessarily. Modern pre-trained language models (LLMs) do not require company-owned training data — they are already trained on huge amounts of text and can be adapted to your domain through prompting and RAG systems. For specific fine-tuning we typically need 100-1,000 example data points, not millions.
Yes — we call these multi-agent systems or agentic AI. Specialised AI agents take over different subtasks and coordinate under the control of an orchestrator agent.
Yes — but it depends on correct implementation. GDPR compliance is not an afterthought at Globeria, but part of every project from the start. We work exclusively with European cloud providers or offer on-premise solutions.
By default in European data centres (Germany, Ireland, Netherlands) — never in the USA without explicit consent and appropriate guarantees. For maximum data control we offer on-premise solutions where all data remains on your own infrastructure.
The EU AI Act is the first comprehensive EU AI regulation (valid from 2024, full application from 2026). It classifies AI systems by risk categories. Most SME applications fall into the 'minimal' or 'limited' risk categories — with manageable requirements.
Your data belongs to you — always. After project completion you receive all developed systems, models and data fully transferred. We delete all project data on our systems within 30 days of project completion, unless an ongoing support contract exists.
Yes — with the right architecture. For sensitive data (financial data, personnel data, medical data, trade secrets) we recommend on-premise solutions or private cloud deployments where all data is processed in your infrastructure.
In four clear steps: First, analysis (week 1): free initial consultation, process analysis, AI readiness assessment. Second, strategy (weeks 1-2): clear roadmap, use case selection, fixed-price quote, GDPR check. Third, implementation (weeks 2-5): agile development, weekly reviews, integration. Fourth, go-live (from week 5): launch, team training, ongoing monitoring.
No — that is one of our most important advantages for SMEs. We handle all the technical work: development, integration, testing, deployment and ongoing operations. Your team only needs to operate the finished system — we train your employees in a 1-2 hour introduction session.
A first AI pilot goes live in 5 working days. That is our promise — and we keep it. More complex projects with full system integration, multiple AI agents or extensive custom development typically take 4-8 weeks with weekly status updates.
Go-live is the beginning, not the end. After commissioning we offer: ongoing monitoring of AI performance, monthly reports with KPIs, proactive optimisations, technical support and continuous further development.
Change management is part of our process. We involve your employees from the start: process interviews in the analysis phase, weekly reviews during development, training before go-live.
Before the project starts we jointly define clear, measurable KPIs — for example: processing time (before/after), error rate, cost savings in €, customer satisfaction, number of automated processes.
From our completed projects: Invoice processing: 3 full-time positions → 0.2 FTE, savings ~€120,000/year. Call centre: 12 agents → 4 agents + AI, savings ~€240,000/year. Email support: team of 5 → team of 2, savings ~€90,000/year.
We do not give blanket results guarantees — that would be unprofessional because every company is different. What we guarantee: first, that we deliver a working AI system meeting agreed specifications. Second, that we create a realistic ROI estimate before the project.
Four key differences: First, fixed prices instead of open day rates over months. Second, 5-day pilot instead of months-long concept phases. Third, GDPR as standard, not an option — with our own certified DPO in-house. Fourth, we are developers, not consultants — we deliver running systems, not PowerPoints.
We specialise in DACH SMEs and mid-sized companies with typically 10 to 500 employees. Smaller companies particularly benefit from the fixed-price model and fast implementation. Larger mid-sized companies appreciate our ability for deep system integration and complex multi-agent architectures.
Your question not listed?
💬
Free Consultation
30 minutes with an AI expert — we answer all your questions personally.
Book appointment →
Get Started
Ready for the first step?
Free 30-minute call — no obligation, no sales strategy, just honest answers.