The Imagination Gap – Why Imagination Is the Real Bottleneck for Successful AI Implementation

The problem with the “right problem” is the decisive problem

Whether one solves such a problem with AI, with a spreadsheet, or with an abacus is ultimately irrelevant. If solving the problem itself brings no benefit to the organisation, then no tool in the world can generate value from it.

BCG’s “Build for the Future” study from 2025 confirms the broader picture: only five per cent of all organisations generate substantial economic value from AI. Sixty per cent achieve no measurable benefit despite considerable investment [4]. McKinsey’s State of AI Report reaches a similar conclusion: 88 per cent of organisations deploy AI in at least one function, but only six per cent qualify as “high performers” [5].

Our interpretation: the gap between tool access and actual value creation – as demonstrated by the high performers – is enormous. It cannot be explained by technological deficits, but rather by a gap in strategic imagination when it comes to addressing the “right problems” with AI.

Invisible friction: problems that nobody recognises as problems

The Imagination Gap has yet another dimension. It is not only about selecting the “right problems” where AI deployment genuinely adds value. It is also about the challenge of failing to recognise actual problems in the first place – because they have long since been normalised as typical workload. Or in other words: that’s how we’ve always done it.

Research shows that employees spend an average of 1.8 hours per day – 9.3 hours per week – searching for information [8]. That is roughly one-fifth of total working time, bound up in invisible friction. Nobody would place this information search on the agenda as a “problem” – it simply belongs to everyday work. That is precisely why it appears on no list of AI use cases. And that is precisely why it exemplifies the Imagination Gap: a real bottleneck that remains invisible because it has never been questioned as one.

The counterpart to this is what researchers at the University of Leeds describe as “Efficient Inefficiency” [9]: organisations that deploy AI on tasks that were superfluous from the start – thereby merely accelerating idle running rather than eliminating it. This, too, is a consequence of the Imagination Gap. Those who fail to recognise which tasks genuinely create value risk using AI to solve the wrong problems more efficiently whilst still contributing nothing of value. 

AI as diagnostic instrument rather than accelerator

The strategically interesting question is surely this: what happens when an organisation deploys AI not merely as a tool for acceleration, but as an instrument for diagnosis?

AI can reveal patterns that remain invisible in day-to-day operations. It can challenge processes that nobody questions any longer because they have always worked that way. It can – to use an analogy – function as an X-ray machine that shows where friction arises in processes, or where energy is not being deployed productively because no substantial value is being generated for the organisation.

And it can do something else: systematically unlock the implicit knowledge held in employees’ minds. In many organisations, the most important processes, decision rules, and accumulated expertise have never been comprehensively documented and examined. They exist as internal expert knowledge – but not as codified knowledge. This represents not only a risk to the organisation, but also a fundamental obstacle to AI deployment: without context, AI can only create value to a limited degree.

Organisations that systematically document their processes and domain knowledge – and AI can serve as an ideal sparring partner in this endeavour – simultaneously lay the foundation for valuable AI use cases.

This requires, however, people who know where to look. And this is precisely where the circle closes.

The solution lies not in “more technology” but in “more imagination”

BCG’s widely cited 10-20-70 principle for successful AI implementation attributes the technology component to 10 per cent algorithms and 20 per cent data and infrastructure. The remaining 70 per cent falls to people and processes [10] – to enabling people, inspiring them, and unlocking their imagination for where AI could genuinely provide helpful support in their daily workflows. 

The Imagination Gap does not close through more knowledge about AI tools, nor through better infographics. It closes when organisations systematically address three things.

First: deliberately strengthen imagination. This sounds unusual in the context of technology implementation, yet it is the decisive first step. Organisations need structured formats in which teams examine their own work processes with fresh eyes. The starting question is not “What can AI do?” but rather “Where does friction arise in our operations that we no longer question?”

Second: leverage existing process knowledge as a strategic advantage. In a world where every organisation can access the same AI models, organisational context becomes the differentiating factor [11]. Employees know the workflows, the exceptions, the workarounds, and the recurring bottlenecks. This knowledge is the raw material from which value-adding AI use cases emerge – but only when it is systematically unlocked.

Third: enablement before technology. In a study of over 1,100 professionals, Prosci found that 38 per cent of all challenges in AI adoption are attributable to insufficient user competence – compared with only 16 per cent for technical implementation problems [12]. Enablement does not mean offering training courses on prompting techniques. It means giving employees the confidence to integrate AI as a tool within their own professional expertise – without the fear of being replaced, and with the freedom to experiment.

Asking the right question for strategic AI Implementation

The Imagination Gap is not a technical problem. It is a strategic one. And it affects SMEs and family businesses to a particular degree: 81 per cent of German companies regard AI as the most important future technology [13]. Yet only six per cent deploy it across multiple business areas [14]. The greatest barrier identified by the Institut der deutschen Wirtschaft is not missing technology, but the perceived difficulty of assessing AI’s business value – cited by 62.7 per cent as the most frequent obstacle [14].

That is the Imagination Gap in a single figure.

The encouraging news: SMEs and family businesses bring precisely the prerequisites needed to close this gap. Deep domain knowledge, process orientation, short decision-making paths, and employees who know their business. What is lacking is not AI competence. What is lacking is a systematic approach to connecting existing professional expertise with AI’s possibilities.