The AQ Score
Twelve architectural traits, five maturity levels, one score out of sixty — benchmarked against published industry averages.
- Length
- 12 questions · about 4 minutes
- Scored out of
- 60
- Benchmark
- 27
Where your organisation sits on the twelve properties that let a machine improve a million-fold in a decade.
AQ Score
Your company is not slow because your people are slow. It is slow because it is running 1980s CPU architecture in 2026.
Your company is not slow because your people are slow. It is slow because of how it is arranged. This instrument scores the arrangement — twelve traits, five levels each, sixty points total. It is the same instrument used in the Accelerated Organization Program, run on your own reading rather than an assessor's.
Answer as the organisation actually behaves, not as the policy says it should. Every score is worth exactly as much as the honesty of the answer.
The benchmark, and how it was built.
A score is only as good as the thing it is compared against. So the benchmark carries a grade and a method, exactly like every other figure on this site.
27/60[B]Grade B · ClaimedMean of published organisational maturity assessments across mid-to-large enterprises. composite of secondary sources, as at 26 August 2026.As at 26 August 2026
Verified
Traced to a primary source — a filing, an audited statement, a contract, a dataset held by us — and the source is named.
Claimed
Reported by a credible party, including by us, but not independently traceable to a primary document.
Estimate
Modelled, sampled or directionally derived. Method stated. Treat as a shape, not a number.
All 12, published in full.
Nothing is hidden behind the interaction. If you would rather read the instrument than run it — or run it on paper, in a room, with other people — here it is.
01 Parallel processing — When your organisation faces a decision that touches four functions, what happens?
Answer for the last real one, not the ideal one.
- 1It goes function to function, in sequence, and takes weeks
- 2It goes in sequence but someone chases it
- 3Two functions work in parallel; the rest wait
- 4All four work in parallel against a shared brief
- 5Parallel is the default and sequential needs a reason
02 Distributed intelligence — Where does the authority to decide sit?
Not where the org chart says. Where it actually sits.
- 1One person, and everything queues behind them
- 2A small executive group, meeting weekly
- 3Delegated by value threshold, reviewed after
- 4Delegated by domain, with published limits
- 5Decisions are made where the information is, by default
03 Scalable architecture — When volume doubles, what breaks first?
- 1Everything — we would rebuild
- 2The people; we would hire proportionally
- 3One or two known bottlenecks we have listed
- 4Little; we scale most things without redesign
- 5Nothing structural. We have already run at that volume
04 Data as infrastructure — If I asked for your last quarter’s operating numbers, where do they come from?
- 1Someone assembles them from spreadsheets
- 2A monthly pack, built by hand from several systems
- 3A dashboard, with known gaps people work around
- 4Systems of record, queried directly, mostly trusted
- 5A single governed layer everything reads from
05 Continuous learning — What happens after a project ends?
- 1Nothing formal. We move to the next one
- 2A retrospective document nobody reads again
- 3A retrospective, and sometimes a process change
- 4A retrospective that must produce a named change
- 5Changes are made mid-project and the standard updates itself
06 Pattern recognition — How does your organisation notice a recurring problem?
- 1When someone senior happens to notice it
- 2When it appears in a monthly report
- 3We track recurrence on the big categories
- 4Recurrence is instrumented and reviewed on a cadence
- 5Patterns surface automatically and trigger a route
07 Feedback loops — How long between an action and knowing whether it worked?
- 1A quarter or more, if at all
- 2A month
- 3A fortnight for most things
- 4Days, for the decisions that matter
- 5Hours. The loop is instrumented and closed
08 Transfer learning — When one unit solves something hard, what happens elsewhere?
- 1Nothing. Other units solve it again later
- 2It spreads if the two leaders happen to talk
- 3We circulate it; adoption is voluntary and low
- 4There is a mechanism to port it, and it is used
- 5Solutions are packaged for reuse as a matter of course
09 Network effects — Does each new customer, partner or unit make the others more valuable?
- 1No. Each is served independently
- 2Marginally, through brand
- 3In one area — data or channel — deliberately
- 4Yes, in several areas, and we can point at the mechanism
- 5Yes, and the mechanism is measured and managed
10 Self-optimisation — Who improves your processes?
- 1A consultant, occasionally
- 2A central improvement function, on a programme
- 3Team leads, when they have capacity
- 4The teams themselves, with a standing mandate
- 5The system adjusts within governed limits, and reports what it changed
11 Modular composability — How much of what you do could be recombined into a new offer in a quarter?
- 1None. Everything is bespoke and entangled
- 2A little, with significant rework
- 3Perhaps a third, in known combinations
- 4Most of it. Components are defined and documented
- 5Recombination is how we launch, and it takes weeks
12 Emergent intelligence — Does your organisation produce answers no individual in it held?
This is the hardest trait to score honestly.
- 1No. Output tracks whoever is most senior in the room
- 2Rarely, and by accident
- 3Sometimes, in the teams that work well
- 4Often. We have arranged for it deliberately
- 5Routinely, and we can describe the arrangement that produces it
Scored out of 60. Published benchmark 27. Method and grade above.
What each range means.
Published before you take the test, so the reading cannot be tuned to flatter afterwards.
Level 1–2 · Sequential
The organisation is arranged for sequential decisions. The good news is that at this level the first three moves are cheap and highly visible — usually one approval layer and one instrumented feedback loop. The gap to a peer at level four is not effort. It is arrangement.
Level 2–3 · Transitional
Some traits have moved and others have not, which is the most common and most frustrating position — because the accelerated traits are held to the pace of the traditional ones. A machine is held back by its weakest part. Fix the laggards, not the leaders.
Level 3–4 · Accelerating
The architecture is genuinely different from a traditional organisation, and the compounding has begun. The risk at this level is the opposite of the usual one: you will be tempted to add capability rather than remove the two remaining structural brakes.
Level 4–5 · Accelerated
This is a rare score, and the honest first question is whether it was scored generously. Re-run it with a sceptic and compare. If it holds, the useful work is now benchmarking decision velocity externally rather than improving traits internally.
Questions about The AQ Score
Answered plainly, with the real figures. If something here is wrong, tell us and it goes in the log.
Twelve architectural traits, five maturity levels each, scored one to five — twelve times five is sixty. The bands are published before you take the test so the reading cannot be tuned afterwards: 0–23 is Sequential, 24–35 Transitional, 36–47 Accelerating, 48–60 Accelerated. Scoring happens in the browser and nothing is transmitted unless you request the report.
The published benchmark is 27 out of 60 — the mean of published organisational maturity assessments across mid-to-large enterprises, graded [B] as at 26 August 2026. Anything from 24 to 35 is the Transitional band and is the most common position. Scores of 48 and above are rare, and the honest first question is whether the run was scored generously.
Parallel processing, distributed intelligence, scalable architecture, data as infrastructure, continuous learning, pattern recognition, feedback loops, transfer learning, network effects, self-optimisation, modular composability and emergent intelligence. Each is scored on the same five-level ladder, from the sequential behaviour at level one to the arrangement that produces the trait by default at level five.
The instrument is free and requires no account or email to start. The AQ Deep Report — trait-by-trait breakdown, the full benchmark set with sources and dates, and the three moves that shift the score fastest — is listed at ₹4,900. Payments are not yet connected; that figure is the intended band as at 26 August 2026.
The lowest-scoring Foundation or Quick-Win trait — not the most interesting one. A machine is held back by its weakest part, not its strongest, which is why a mixed profile is the most frustrating position: the accelerated traits are held to the pace of the traditional ones. The Accelerated Organization Program runs this as ninety days, then a twelve-month track, at ₹25L – ₹1.2cr banded by headcount.
Five more, one per altitude.
The Flywheel Locator
Where your AI capability actually sits on the powertrain — and which part is holding the wheel.
Do You Actually Have a Moonshot?
Whether the thing at the top of your ladder is an objective you could be held to, or a sentence that sounds good in a deck.
The Coherence Matrix, lite
Whether the things your group holds build anything the objective needs — or merely make money.