Which industries lead quantum adoption in 2026?

Status as of: August 2026. Three industries top every 2026 adoption survey: chemicals and life sciences, travel and logistics, and financial services (McKinsey Quantum Technology Monitor 2026, Apr 2026). But "adoption" is an elastic word. What the surveys actually count is spending in anticipation — 300+ companies working with quantum vendors, a third of the 162 analyzed spending more than $10M a year, 7% more than $50M. What no survey shows, in any industry, is the thing the spending anticipates: a production deployment with a measured advantage over a strong classical baseline. That rung of the ladder is empty everywhere. Leading the adoption race and obtaining an advantage are, so far, two different things.

What do surveys measure when they say "adoption"?

Different counters count different pipes. McKinsey's Quantum Technology Monitor 2026 (Apr 20, 2026) counts engagement and budget: over 300 companies actively collaborating with quantum vendors, 162 analyzed in detail, one third allocating more than $10M annually, 7% more than $50M, with a maximum individual budget of $200M. It also notes that 72% of quantum computing use now occurs at privately-owned entities, and that European companies lead. QED-C's State of the Global Quantum Industry 2026 (via The Quantum Insider, Apr 14, 2026) counts vendor revenue: a $1.4B quantum computing market in 2025, projected at $3B by 2028, across 7,418 quantum-engaged organizations worldwide. And the IQM-published State of Quantum 2026 (research by The Quantum Insider, Jun 18, 2026; n=107 senior practitioners plus 19 interviews) counts self-reported activity: 89% report hands-on quantum work.

None of these numbers is a measurement of deployed value. Budgets measure conviction; vendor revenue measures what adopters pay in; self-reports measure what practitioners say. All three are honest metrics of input. The question the headline word "adoption" smuggles in — is it working? — is answered by none of them. We made the same point about government spending: input metrics are not outcome metrics.

Which industries lead — and what do they spend on?

The composition is consistent across sources. McKinsey names the same three verticals it has tracked for years, each mapped to a dominant problem class, and names adopters including Airbus, Boehringer Ingelheim, E.ON, JPMorgan Chase, and Liberty Mutual.

One detail in the finance row deserves its own sentence: the most concrete "quantum adoption" in financial services — post-quantum cryptography migration — does not use a quantum computer at all. It is a classical software migration driven by a regulatory clock (the one clock that is actually running). An industry can lead the adoption tables while its most tangible quantum line item is defensive.

Which rung of the ladder is each leader actually on?

"Adoption" flattens a ladder into a word. Unfolded, the ladder has four rungs: exploration (hands-on work, no commitment), pilot/hybrid (operational tests, classical systems doing the heavy lifting), production (quantum in a live workflow, on any terms), and production with measured advantage (beating your strong classical baseline on your instance, measured). The 2026 numbers place almost everyone on the bottom two rungs.

The 13% deserves scrutiny rather than celebration. It comes from a vendor-published survey (IQM published; The Quantum Insider conducted the research) of 107 practitioners, splits into 10% "limited production use" and 3% "scaled deployment", and ships with no published criterion for what counts as production. That is not an accusation — self-reported surveys are how young fields count themselves — but it means the number cannot be audited, and it is the only number anyone has for the third rung. The fourth rung needs no survey: a production deployment with measured advantage would be the most publicized event in the industry, and the scoreboard of measured end-to-end wins on useful problems stands at zero.

What does the gap between expectation and measurement look like?

A D-Wave-commissioned Wakefield Research survey (400 business leaders, fielded May 2–15, 2025) found that 81% of respondents believe they have "reached the limit" of what classical optimization can do for them, and 27% expect quantum optimization to return over $5M within the first 12 months of adoption. Set those two numbers next to the measured record: strong classical baselines remain unbeaten on same-instance comparisons in routing and portfolio optimization, where our own sealed runs show a provably-optimal classical solver winning 20 out of 20 against QAOA even when the quantum side is simulated without noise. The 81% is a sentiment; the 20/20 is a measurement. Neither number is dishonest — a commissioned survey measures expectations and says so — but reading sentiment as evidence is exactly the substitution this post's headline question invites.

Our own data adds nothing at the industry level, and we say so: Rosetta Q measures instances, not industries. Our sealed series cover small optimization and quantum-walk classes where the classical baseline is unbeaten; we have no survey data of our own and no measurements in any of the leading verticals' production stacks, and we claim none.

What we know / what we don't know

What we know. The 2026 leaders by spending are chemicals/life sciences, travel/logistics, and financial services (McKinsey, Apr 2026). A third of 162 analyzed companies spend over $10M a year; 7% over $50M. Most deployments are described by McKinsey itself as "experimental or hybrid". The best-documented executed workflow in a leading vertical — Algorithmiq's 50+ qubit cancer-drug pipeline with IBM and Cleveland Clinic — won a $2M realization prize (Apr 2026), while the $5M prize for demonstrated advantage has no announced winner. Zero industries have a published production deployment with measured advantage over a strong classical baseline.

What we don't know. Whether any of the self-reported 13% would survive a defined criterion — no auditor has looked, in either direction. What the selection biases of small samples (n=107; 162 analyzed; 400 surveyed) do to every percentage above. Whether Europe's lead reflects deployment or disclosure culture. Whether spending is a leading indicator of advantage or just of conviction — the historical record on that conversion does not exist yet, in any industry. And whether the first measured advantage will appear in the vertical that spent the most; nothing in the data guarantees that the adoption table and the results table will ever rank the same.

Rosetta Q publishes verdicts with reproducible raw data. This is educational content, not a product claim.