Where most programmes quietly fail
- The bottleneck in most innovation programmes is not a shortage of ideas. It is the quality of the question those ideas are being asked to answer.
- Innovation fails at three predictable points: problem framing, the leap from discovery to definition, and the transition to scale. Most organisations only address one. The ones that break through address all three.
- The difference between an organisation that consistently innovates and one that does not is almost never talent or budget. It is whether innovation is treated as a discipline or an event.
Early in my career, I attended a product workshop where the keynote speaker opened with a simple exercise. He asked everyone in the room to take thirty seconds and sketch a design for a vase.
Almost everyone drew the same thing. A cylindrical container, wider at the top than the bottom, with something flower-like emerging from it. There were variations in proportion and decoration, but the core concept was remarkably consistent across a room of several hundred people.
Then he changed the question.
He asked us to design a new way for humans to enjoy and experience flowers indoors.
The room came alive. People stopped sketching cylinders and started asking entirely different questions. What does "experiencing flowers" actually mean? Is it visual? Is it scent? Is it the act of tending to something living? Is it seasonal, ephemeral, something to mark a moment? The ideas that emerged were genuinely varied: internal irrigation systems that fed plants year-round without manual care, diffusers that replicated the scent of fresh flowers across a living space, wall-mounted systems that combined plants with light and moisture sensors, subscription models built around living botanical installations rather than cut flowers at all. Some ideas were impractical. Some were genuinely interesting. A few made us laugh.
But the point was made, and it changed how I approach every problem I have worked on since. The quality of your solutions is directly determined by the quality of the question you started with. The vase was a pre-formed answer masquerading as a problem. The moment the question opened up, so did the entire solution space.
The Innovation Bottleneck Is Almost Never Ideas
I have spent a significant portion of my career driving innovation and product ideation across a wide range of industries. At Accenture and EY, working across financial services, telecommunications, retail, healthcare, public sector, and more, the problems changed with every engagement. The industries changed. The stakeholders changed. The technology changed. The competitive dynamics changed.
What stayed the same, with remarkable consistency, was this: the organisations that struggled with innovation were not struggling because they lacked ideas. Most had more ideas than they could ever execute. They were struggling because they could not distinguish between ideas that addressed a genuine, substantive problem and ideas that were simply responses to the question they happened to start with.
83%
of senior executives rank innovation as a top-three priority. Only 3% of companies qualify as innovation-ready to deliver on it.
BCG Most Innovative Companies Report, 2024
BCG's 2024 innovation research captures this gap precisely. 83% of senior executives rank innovation among their top three priorities. Only 3% of companies qualify as innovation-ready, a figure that has dropped sharply from 20% just two years earlier. The ambition is there. The capability to act on it is not.
The capability gap is not technical or financial. It is methodological. It is the absence of a structured, repeatable way to move from opportunity to outcome.
Where Innovation Actually Breaks Down
In my experience, innovation programmes fail at one of three predictable points.
The first is problem framing. The organisation identifies a domain it wants to innovate in, generates a set of ideas within that domain, and then invests in developing the ideas without ever rigorously examining whether the underlying problem was correctly defined. The result is a set of well-executed solutions to the wrong problem. This is the vase problem, operating at scale.
The second is the transition from discovery to definition. Many innovation programmes are good at generating insight and generating ideas, but struggle to make the structural decisions that turn an idea into a defined product opportunity. What is the target outcome? Who is it for, specifically? What does success look like at six months, eighteen months, five years? What are we not building? These questions are harder than generating ideas, and they are often avoided until the avoidance itself becomes the bottleneck.
The third is scaling. An idea works in a proof of concept. It works in a pilot. And then it fails to scale across the broader organisation, not because the idea is wrong, but because the operating model, change management, and governance required to scale it were never designed. The innovation team moves on to the next shiny thing and the organisation is left with a well-documented pilot that nobody is running.
A Framework That Works Across Industries
What I have found effective, across the varying complexity and context of industries I have worked in, is an approach that treats the innovation lifecycle as a distinct, structured discipline rather than a series of workshops and sprints.
It starts with genuine problem exploration, which means spending real time with the people who live closest to the problem before defining what the solution space should look like. This is the step that the vase exercise illustrates: the goal is to understand the underlying human or business need before the question is constrained by any particular answer.
It then moves through a structured discovery and definition phase, where insight is synthesised into a specific, testable opportunity definition. Not a broad direction. A specific articulation of the problem, the target user, the proposed value, and the hypotheses to be tested.
From there, development and testing happen in iteration cycles with deliberate learning objectives. Not sprints for the sake of velocity, but structured experiments designed to resolve the highest-risk unknowns first.
And then, critically, the scaling phase is designed from the start, not retrofitted at the end. Who owns this product as it moves from innovation team to business? What needs to change in the operating model for it to work at scale? What is the change management plan?
The consistency of this approach across different sectors has been one of the more validating things in my career. The problems differ. The solutions differ. The customers differ. The underlying discipline that connects insight to outcome is remarkably consistent.
Key Takeaways
- The bottleneck in most innovation programmes is not ideas. It is problem framing. Start by genuinely questioning whether the problem you are solving is the right one before any solution work begins.
- Narrow questions produce narrow answers. Reframe the question at the right level of abstraction and the solution space expands dramatically. The vase exercise makes this concrete.
- Innovation programmes fail at three predictable points: problem framing, the transition from discovery to definition, and scaling. Designing for all three from the start changes the outcome.
- A structured, repeatable approach to the innovation lifecycle consistently outperforms ad hoc ideation, regardless of industry or company size.
- Scaling needs to be designed before development begins, not retrofitted at the end. The organisations that crack this treat innovation as a discipline, not an event.