Over 50% of Large Czech Organizations Are Holding Back AI Adoption Due to Unprepared Data

A November survey conducted by Adastra among 80 senior executives from major Czech enterprises shows that the primary obstacle to scaling AI is the poor readiness of their data. Respondents from financial services, telecommunications, retail, insurance, manufacturing and automotive, energy, utilities, and the public sector also confirm that high-quality data is critical to unlocking faster business growth through AI.

Beyond unprepared data, AI initiatives are also constrained by limited internal capacity and skills

According to 56% of respondents, the main limitation is that their data is not in a state that enables the development of reliable AI services. The second most common barrier is a lack of internal capacity and competencies (46%). Companies also frequently face unclear ownership of AI topics within the business (40%) and the need to identify a genuinely strong use case (36%). Lack of trust in AI appears as a barrier in only 14% of responses.

“The survey shows that Czech companies are ready to adopt AI and trust its value. The next step is to ensure high-quality, trustworthy data — only then can they build AI services that scale safely and effectively,” says David Kaláb, Vice President for Data Management, Adastra Czech.

Companies are looking to accelerate business impact and AI adoption

The main driver for investing in AI-ready data is direct business growth enabled by AI (48%). For one third of companies, the priority is to speed up overall AI adoption (35%), as well as to strengthen risk management and control the behaviour of AI agents (both 29%).

Half of companies are taking their first practical steps — but very few can scale AI across the organization

While 51% of organizations are already working on initial AI-ready use cases, only 4% report that their data is mature enough to scale AI or roll out AI projects company-wide. A third remain in the consideration phase (33%), and 12% have not discussed the topic at all. A lack of production-grade use cases is acknowledged by 18% of respondents.

“The survey confirms what we see in practice: without high-quality data and a clear data strategy, AI projects struggle to move beyond the pilot stage,” says David Kaláb. “But organizations already working on their first real use cases are seeing concrete impact — higher efficiency in customer service and faster decision-making in core processes.”