Rethinking an Excel-Driven Industry
For most mid-sized industrial companies in Germany, commodity purchasing decisions — when to lock in a contract, when to wait — are still made using historical price spreadsheets and gut feel. At stake is often the entire annual result; the tool used to answer the question is frequently an Excel history that only shows what has already happened.
Datasphere Analytics, a GmbH registered in Münster and founded in 2025, wants to change that. The platform turns global events into commodity price forecasts that procurement, finance, and risk teams can actually defend. It's a direct attack on one of the most data-poor corners of industrial operations.
The Team Behind the Platform
Behind Datasphere Analytics stands a three-person founding team with long-running experience across AI, data analysis, and industrial markets. CEO Lukas Haemisch leads the company externally, while CDO Patrick Petter is the technical architect behind the AI infrastructure, and Prof. Dr. Reiner Kurzhals — an AI and industry expert — anchors the scientific methodology and customer network.
Together, the founders have developed data-intensive AI applications across different industrial contexts — from forecasting systems to decision models for complex markets — and are now applying that know-how directly to raw material and procurement decisions.
Notably, Haemisch has taken a deliberately lean approach to financing. While many founders chase VC millions, he chose to take just a single business angel on board — a bet that staying capital-light keeps the team focused on customer value over growth theatre.
How the Technology Works
Datasphere brings together expertise in data science, finance, procurement, and trading to build AI agents for commodity forecasting and decision intelligence. The system is built around what the company calls Event Intelligence — connecting geopolitical and market signals to commodity price movement in near real time.
The platform's architecture spans three layers:
- Forecasting: converting news and global events into directional price signals
- Scenario intelligence: modelling geopolitical and market shocks to surface price impacts early
- Workflow integration: embedding results directly into procurement, finance, and risk workflows for measurable business impact
At scale, multiple agents combine forecasting, event extraction, and scenario intelligence to automate decisions and augment teams.
This architecture matters because it sidesteps a common failure mode in industrial AI: building impressive models that never reach the people making the actual call. By designing for workflow integration from day one, Datasphere is targeting adoption, not just accuracy.
The Real Barrier: Organisational Inertia
Haemisch is candid about where the friction actually lies. The technology exists — the harder problem is cultural. Many procurement teams have operated the same way for decades, and AI-driven forecasting requires them to trust a model's output over years of manual instinct.
This is a pattern familiar across industrial B2B software. Companies like Celonis and SupplyOn spent years convincing similar audiences that process data could be turned into business decisions — and both found that the sales cycle was really a change management cycle in disguise. Datasphere faces the same dynamic, just one layer deeper in the supply chain.
The company also sees its Event Intelligence as relevant well beyond procurement — for media outlets and analyst firms that need to better anticipate market reactions to breaking news. That optionality suggests the core engine has broader applicability than the initial industrial niche, and gives the team a potential expansion path that doesn't depend entirely on convincing conservative procurement departments.
Ambitions and Road Ahead
Datasphere's stated goal is to become the leading Commodity AI Intelligence Platform for industry, procurement, and risk management in Germany. From there, the expansion path looks like this:
- Scale the Commodity Intelligence Platform to cover additional markets and commodity classes
- Move from point forecasting to full AI-assisted decision support
- Expand Event Intelligence into adjacent verticals like financial media and risk advisory
For founders and operators building in B2B or industrial AI, Datasphere is a useful case study in a recurring pattern: the largest incumbent advantage isn't a better competing product — it's the inertia of a workflow that's "good enough." Winning here means proving ROI in the language of procurement managers, not data scientists. The lean financing structure and deliberate focus on workflow integration suggest the team understands that — and is building accordingly.
