Every year, Italian companies chase a prize worth roughly half a trillion euros, buried in PDFs. The country's public procurement system, a labyrinth of tenders known as gare d'appalto, is a massive economic engine. It is also a notorious time sink, requiring specialists to manually comb through thousands of complex documents to find the right opportunities and then assemble compliant bids. Cato AI, a Milan-based startup, is betting that the most valuable thing to sell in this market isn't a consulting service, but a filter.
Founded in 2025 by Andrea Zorzetto and Matteo Bossolini, Cato has raised €1.6 million in pre-seed funding to build an AI assistant for this exact process [Vestbee, 2026]. The platform scans public tender portals, filters opportunities based on a company's profile, extracts key requirements, and helps draft responses. In six months, the company reports it has been adopted by over thirty procurement and sales departments across healthcare, education, and infrastructure, with clients including Sol, Ivs, and Cns [F4.fund, 2026] [Ramin Afdjei - Ring Capital, 2026].
The Wedge: Automation as a Compliance Tool
The core of Cato's product is a workflow that turns a sprawling, unstructured information problem into a series of automated steps. Public tender documents are often hundreds of pages long, filled with legalese and specific technical requirements. Cato's AI aims to do that in minutes.
The process follows a clear cycle: filtering tenders, extracting requirements, drafting bids from historical data, and compiling the final paperwork [Lorenzo Franzi - Italian Founders Fund, 2026]. The company claims its systems have already filtered over 60,000 tenders, analyzed more than 2,000 in detail, and generated over 500 documents [Startupbusiness, 2026].
The Italian Laboratory
Cato's initial focus on Italy is a strategic choice. Italy's public procurement market is estimated at approximately 500 billion euros annually [Economyup, 2026]. Furthermore, the process is standardized across the country, with tenders published on official portals. This creates a uniform, if complex, dataset for an AI to learn from.
The Team and the Check
The founding duo brings a complementary skill set to the problem. The €1.6 million pre-seed round was led by Italian Founders Fund and included Vento, Moonstone Venture Capital, and a group of Italian angel networks and funds [Vestbee, 2026].
Where the Model Could Stumble
For all its early momentum, Cato's path is paved with specific, non-technical risks. Public procurement is a domain where accuracy is non-negotiable; a missed clause or a misinterpreted requirement can disqualify a bid. The platform's reliability in parsing complex, ever-changing regulations will be its most critical benchmark.
- The human-in-the-loop question: Can the AI sufficiently augment expert judgment, or will companies still require senior staff to validate every automated analysis?
- Scale beyond Italy: Replicating the model requires building new data pipelines and legal frameworks for each jurisdiction.
- Competitive response: Larger government software providers or enterprise resource planning (ERP) vendors could easily add similar functionality as a module.
The Unit Economics of a Tender
The real test for Cato won't be its ability to filter tenders, but to prove it changes the win-rate math for its customers. For Cato to become a lasting business, it must do more than just replace junior analysts. It must beat the incumbent workflow of dedicated consultants and internal legal teams who currently own the high-value, strategic end of the bidding process.