If you work in a company and someone mentions quantum computing, the reaction is usually one of two extremes:
- “We need to jump on this immediately.”
- “This is too early — ignore it.”
Both are unhelpful.
A more balanced approach looks like this:
1️⃣ Don’t start with hardware. Start with your problem.
What are you actually trying to improve? Optimization? Modelling accuracy? Simulation fidelity?
If the problem isn’t clearly defined, quantum won’t fix it.
2️⃣ Map your bottlenecks honestly
Is your limitation:
- Data quality?
- Model assumptions?
- Compute cost?
- Scaling behaviour?
Most bottlenecks today are still classical.
3️⃣ Explore hybrid experiments
If quantum is relevant, it will almost always be as a small component inside an existing workflow.
Think:
- Proof-of-concept
- Limited-scope experiments
- Controlled comparisons
Not a full system replacement.
4️⃣ Define success before you begin
- Is a 5% improvement valuable?
- Is reduced modelling bias important?
- Is learning strategic positioning the real goal?
Without a clear success metric, experiments drift.
Quantum exploration today is about:
- Learning
- Careful evaluation
- Controlled experimentation
Not dramatic performance breakthroughs.
Companies that approach it calmly and methodically will be much better positioned if and when the technology matures.
Question:
If you’re in a company, what would make quantum worth even a small experiment for you?