The key to making quantum real 

Airbus and Great Britain’s National Energy System Operator (NESO) are two leaders in applied quantum exploration. Airbus has been among the first to push quantum into real industry use – developing algorithms to model aerodynamics and materials chemistry. Great Britain’s NESO has focused on the other side of the problem – building first‑of‑their‑kind tools that translate quantum research into actionable risk models, helping protect national energy infrastructure from future quantum‑enabled cyber threats. But below the headlines, the foundation of their success is something deeply practical. And it’s the key to making quantum real. 

Why resource estimation matters now 

For the past several years, quantum computing has been in an exploratory phase. Organizations invested time in understanding the basics and experimenting with early ideas. That phase served its purpose: researchers built awareness, calibrated their expectations, and the limits of near‑term quantum hardware became clearer. 

We are now entering a different phase. The question is no longer what quantum computing might do one day, but what can realistically be built, tested, and evolved today. High‑level ideas of potential applications are no longer sufficient to guide serious technical or strategic decisions. The next phase of quantum adoption is about prototyping, feasibility, and engineering trade‑offs – about moving from promise to practice. 

Organizations that want to start implementing quantum solutions will need a much more explicit engagement with practical constraints. And that starts with resource estimation. 

The value of resource estimation 

Resource estimation is the mechanism that turns quantum ambition into action. It forces clarity on what is otherwise left implicit: how many qubits are needed, how deep circuits become, what error rates are tolerable, and how these requirements evolve as problem size or precision increases. 

Resource estimation also lets us identify where quantum computation fits within end‑to‑end workflows, and what it would actually cost to deploy it there. It informs algorithm design by making cost constraints explicit early on. It allows different hardware platforms to be evaluated on a like‑for‑like basis, grounded in workload‑specific requirements rather than generic performance claims. 

Ultimately, all of this information helps us determine feasibility. Can this project be done? And if so, by when, and for what price? This helps organizations avoid spending years refining ideas that will not scale – or waiting too long to act on those that will. 

How to develop a quantum application in practice 

Developing a meaningful quantum application is not a linear exercise, but there is a clear structure to the process. 

The starting point is selecting a high‑value problem – one that matters scientifically or operationally, not just one that appears “quantum‑friendly” on paper. Many candidate problems fail this test: they are interesting demonstrations, but solving them wouldn’t lead to clear benefits. 

The next step is to show that the problem is genuinely computational. Many real‑world challenges are constrained not by compute, but by data quality, modelling assumptions, or organizational processes. This step often involves sensitivity analyses: demonstrating that better solutions emerge as assumptions are relaxed, precision is increased, or model fidelity improves. Only then does additional computational power become relevant. 

From there, a hybrid quantum–classical workflow must be developed. Quantum computation rarely stands alone; its value lies in how it integrates with classical pre‑ and post‑processing. The question is not whether quantum can solve the whole problem, but where it adds leverage within the broader workflow. 

This leads to identifying the “quantum golden zone:” the space where the problem is large enough to challenge classical methods, yet still small and structured enough to be tractable on quantum hardware. Outside this zone, quantum either adds no value or becomes infeasible. 

However, at this stage jumping straight to quantum hardware may still be premature. The real progress comes from refining the problem and approach until running it is likely to deliver useful results. 

Finally, sustained collaboration with the quantum industry is essential. Hardware and software platforms are evolving rapidly, and assumptions made today will change. Working closely with technology providers allows algorithms, mappings, and error models to co‑evolve with hardware roadmaps, and ensures that feasibility assessments remain grounded in reality rather than static projections. 

Resource estimation as the common thread 

What ties all of these steps together is resource estimation. Resource estimation helps identify computationally viable problems. It reveals whether increased accuracy or reduced assumptions lead to manageable or explosive growth in resource requirements. It guides workflow design by showing where quantum components should sit, and where classical computation will remain dominant. These are central to the process of structuring a quantum project. 

Why implementation details – and expertise – matter 

Just as in traditional computing, the compiler (the last step before execution) is a moment of truth. But feasibility is not determined by the compiler alone. Choices across the full quantum stack have a decisive impact on circuit width, depth, and error tolerance. How problems are represented at the Hamiltonian level, how qubits are mapped, which algorithms are selected, and what error‑correction strategies are assumed can change resource requirements by orders of magnitude. 

There is a growing ecosystem of tools that support parts of this process – from problem formulation libraries and algorithm frameworks to dedicated resource estimators. These tools are powerful and increasingly mature. However, resource estimation remains highly context‑ and platform‑dependent. No single tool captures all relevant design choices, and results are only as meaningful as the assumptions behind them. Interpreting outcomes requires deep domain understanding and architectural insight, not just familiarity with quantum software. 

This is why resource estimation is as much an expertise problem as it is a tooling problem. 

Where Capgemini fits 

This is where Capgemini focuses our quantum work. 

Resource estimation sits at the intersection of domain science, algorithm design, and hardware reality. Capgemini brings these elements together through a platform‑agnostic approach, deep technical expertise across the quantum stack, and decades of experience in computational science and engineering. 

Because we are not tied to a single hardware or software platform, we can objectively assess different approaches against the same problem. Our background in high‑performance computing, simulation, and applied mathematics has taught us that domain details matter – how problems are formulated, which assumptions are made, and where precision truly creates value. Through our classical computing, cloud, and quantum labs, we combine this domain depth with hands‑on expertise in algorithms, workflows, and resource estimation. 

Our focus is not on demonstrations, but on decision‑grade insight: helping organizations understand what is feasible, when it may become viable, and where to invest. 

From theory to practice: NESO and Airbus 

This approach is already being applied in practice. 

In the energy domain, our work with the National Energy System Operator (NESO) uses resource estimation to assess when advanced computational approaches could meaningfully contribute to large‑scale system planning and optimization. Rather than starting from abstract algorithms, the focus is on feasibility, timelines, and the conditions under which new computational capabilities could create operational value. 

Similarly, in our collaboration with Airbus, resource estimation has been central to evaluating quantum approaches for scientifically and industrially relevant problems in materials science and chemistry. By quantifying algorithmic and hardware requirements early, this work has helped clarify not only what may become possible in the future, but also what is not viable today – enabling informed decisions on research direction, collaboration, and investment. 

In both cases, resource estimation acts as a bridge between ambition and execution. 

From exploration to execution 

The time for open‑ended quantum exploration is over. The time for prototyping, engineering realism, and practical trade‑offs has arrived. 

Organizations that build strong resource‑estimation capabilities now will be the first to invest wisely, avoid hype‑driven dead‑ends, and capture real quantum advantage when it becomes viable. Resource estimation is the key to making quantum ambition real.