Study on AI agents in German enterprises: Implementation lags behind hype
- Despite high expectations – 86% believe in the potential – only one in nine German companies has already reached an advanced stage of AI agent deployment.
- Money is not the problem: Only 14% of companies cite lack of funding as a barrier to future AI agent adoption.
- Successful companies create clarity: through a defined business case, a high degree of alignment between IT, compliance, and business departments, and close collaboration at board level.
Munich, February 26, 2026 – Why are some companies more successful than others when it comes to using AI agents? This is the question addressed by a recent Cloudflight study of 150 decision-makers from German enterprises. The study identifies the key prerequisites and criteria for successful AI agent deployment. Success is not primarily dependent on money – that much can be said upfront.
The majority of respondents (86%) agree: Agentic AI has the potential to transform their organization. Expectations regarding the degree of change vary slightly. Just under half (53%) are convinced that AI’s impact will be “significant,” while a third (33%) attribute even “transformative” qualities to Agentic AI.
Expectations are therefore high. But reality has not yet caught up. Only one in nine companies (11%) deploys agents successfully enough to be classified as “advanced.” 38% of surveyed companies are at least already in a scaling phase. Just under half, however, are still in the early stages of adoption. What explains this gap between potential and execution?
Budget plays a secondary role
Contrary to what might be expected, money is not the problem when it comes to launching AI agent projects. Only 14% of respondents cited insufficient budget as a barrier to AI agent adoption. Other obstacles were far more prominent. The most frequently cited barrier, at 51%, is lack of trust and fear of risk. Compliance issues are also seen as highly obstructive, at 39%, followed by unclear responsibilities (27%) and an unclear AI strategy (22%). 40% even conclude that their organization is not yet ready for AI agent deployment. Among companies looking back at failed AI agent initiatives, financial reasons also play only a minor role – just 8% conclude that their project failed due to budget.
Success criteria for Agentic AI deployment
Companies that manage to close the gap between conviction and execution share three characteristics: (1) a clearly defined business case, (2) strong alignment between compliance, IT, and business, and (3) close collaboration at board level. Only when these three elements come together does a company have the opportunity to deploy AI agents successfully.
A clearly defined business case is the foundation for evaluating and advancing a project. Without timelines, ROI targets, and other evaluation criteria, there is a risk that the project never progresses beyond its early stages – exploration, experimentation, or early adoption. The study shows that only 29% of respondents have a clearly defined business case for their AI projects. This means barely a third have even created the basic prerequisite for success.
A business case is the foundation, but it is not enough on its own. Companies also need to achieve strong cross-functional alignment. Who is responsible? How are decisions made? How does coordination work? The data shows that companies where compliance, IT, and business are fully aligned are able to move pilots forward more quickly and scale into additional business units. The presence of a business case also correlates with the degree of departmental alignment. Among fully aligned companies, 60% have a clearly defined business case, while only 8% of moderately aligned companies do.
Even when these prerequisites are in place, one further success-critical factor remains: close and coordinated collaboration at board level. The profiles and responsibilities of CEOs and CTOs are inherently different. CEOs bear responsibility for the company’s future viability and reputation. CTOs are accountable for technical outcomes and system performance. Against this backdrop, AI agents tend to be evaluated differently at the top. This is particularly evident in response to the question “What level of autonomy is acceptable for AI agents in your organization?” While CTOs show the highest acceptance of full autonomy at 33%, CEOs are at the other end of the spectrum at 19%. Blockages can only be avoided when there is open and structured dialogue between these perspectives and both sides are genuinely heard.
“Organizations that focus on the budget question to get Agentic AI projects off the ground are working around the problem. Successful companies in this space don’t spend more – they simply coordinate better. The question needs to shift: from ‘Can we afford this?’ to ‘Can we organize ourselves effectively around this?’ The second question is what determines success. But most companies haven’t yet created the foundations for it,” says Gernot Molin, Chief Technology Officer at Cloudflight | paiqo.
Methodology
The findings are based on a quantitative online survey of 150 C-level and senior executives from German enterprises, conducted between December 2025 and January 2026.
Participants included executives from management, IT, digitalization, and operations (including CEOs, CIOs, CTOs, and CDOs). Only companies with at least 250 employees and annual revenues of at least €50 million were included.
The sample covers companies from manufacturing/production, logistics, financial services, energy, retail, healthcare, and the public sector.
About Cloudflight
Cloudflight is a leading full-service partner for digital transformation. The company offers strategic consulting, custom software development, and cloud operations, with a focus on e-commerce, IoT, artificial intelligence, and cloud-native application development. Cloudflight helps clients across Europe translate innovation into practice and create measurable value.
