In Q1/2026, Deloitte Germany published a noteworthy study: “The ROI of AI: The paradox of rising investment and elusive returns – German Cut”, source below. The team identified a significant gap between overall AI adaptation rates among enterprises in Germany and the added strategic value to these early adopters. Most enterprises in this study focus on efficiency instead of strategy today.
The study found that enterprises in Germany are front runners in AI adoption, compared to peers from other countries. A bottleneck to further, more strategic use might be observed because management still delegates ownership to their IT to a very large extent. Only 2% of the CEOs keep AI initiatives as a priority on their own tableau. This is surprising, because the same study respondents expect strategic value gain in their business models by 2028, while not even actively pursuing this potential. As of today, AI is being used for operational efficiency instead of developing new business models. People are doing the same things as before, but now they are slightly faster with AI. As a result, operational effects occur, but there is hardly any strategic impact due to AI adoption in enterprises.
One reason could be the very long investment cycle which AI is demanding. According to Deloitte, real return requires 24-36 months runway. In Numbers: 31% of AI projects achieve ROI within two Years, 41% between two to three years, and 28% after three years. It goes without saying that smaller projects deliver a return earlier than large and deep projects. Many managers appear to get nervous too soon, stop initiatives after a few months already, and effectively kill the potential upside, never reach any strategic value, and lose the investment for good.
Another challenge is the underlying complexity of AI. The media noise is strong and “marketing AI” is promising perfect results while we enjoy our morning coffee. Reality hits very hard if one believed the advertisement blindly. Honestly, have you ever seen a fast-food ad on TV? How did the burger look like in that ad and how did it look like when you had it right in front of you? Why do so many people think advertisements are different in software?
To really use AI for deep transformative innovation within existing business models or new ones, a cultural shift must happen, and strategic integration is essential. However, this is still often overlooked. AI as a strategic lever belongs in the hands of those who are accountable for an organization’s strategy. It is not just another technical project. The data infrastructure must be developed to serve automation in scale. Everybody is responsible for data quality. And finally, we should focus on scalable use cases that are meaningful to our business once we learned how to apply AI in our organization.
I recommend reading this study: Deutschland im KI‑Paradox | Deloitte Deutschland