Product fact sheet
The assessment engine inside TheAX
AXAT
The AX Assessment Triangulation capability diagnostic

A diagnostic without evidence is an opinion.
AXAT is what makes it a measurement.

AXAT is the assessment engine inside TheAX, the assessment and diagnostic platform. It turns a consultancy’s expertise into an automated, repeatable diagnostic that improves how consultants engage with clients. It measures capability through three independent lenses, diagnoses what is constraining the outcome, advises from the firm’s own model, and tracks whether capability and business results move together.

Measure
Where the client actually stands

Capability read against your expertise, corroborated rather than self-reported.

Diagnose
What is holding the outcome back

Not just a score - the cause, and which capability constrains the outcome.

Advise
The next best step

Guidance drawn from your firm’s own model, specific to this client and their desired outcome.

Track
Whether it worked

Re-read over time, so capability movement and business results are tracked and shown together.

01 ·The problem it solves
CLIENT OUTCOME Revenue, margin, resilience, growth USUALLY MEASURED CAPABILITIES What the organisation must be able to do differently THE MISSING LINK CONSULTING WORK Workshops, analysis, recommendations ALWAYS VISIBLE NO BASELINE · NO TRACKING · NO EVIDENCE

Capability is not one route to a client outcome among several. It is the only one — an outcome cannot change unless the organisation's ability to produce it changes. The work is visible. The outcome is visible. The layer that connects them is not measured at all.

Capability is never measured

No baseline at the start, no tracking through delivery, and no evidence  the outcome is achieved.

The score cannot be defended

Measuring capability by asking is a claim, and a claim does not survive a challenge in front of a board.

The judgement cannot be delegated

Reasoning lives in senior people  cannot be repeated reliably at volume, and it leaves when they do.

01 · The problem it solves01 / 04
AXAT · Product fact sheet
How it works
02 ·How it works

Nothing can be automated while it lives in people's heads - that's why you create Playbooks. But for consultancy to be scalable expertise has to be  written down in a form a machine can reason over. Everything after that is generated from it.

STAGE 01
The Consulting Atomic Model
The base

An LLM by itself cannot reason with your expertise and make it repeatable and measurable. That can only be achieved when your expertise is structured. The Consulting Atomic Model is that structure: capabilities and levels, the evidence that proves them, the guidance that improves them, and the pathways connecting capability to outcome. It is curated and versioned in your firm’s Product Book. TheAX uses AI workflows to structure and store your expertise in the way that makes AXAT diagnostics possible. Authored once with your senior people, in your language. Everything downstream is generated from this — which is why the answer is your firm's judgement rather than a public model's. This is your IP that improves with every engagement.

STAGE 02
The assessment
Generated, not written

To make information collection repeatable and automated TheAX builds a question-based assessment from your model: Perception scoring per capability, the artefacts to request, and the symptom probes that catch what documents miss. Send to stakeholders as an automated interviewer or consultants to use the assessment in face-to-face interviews. Nobody writes a bespoke questionnaire per client, and every engagement asks in the same disciplined way.

STAGE 03
AXAT reads it
The algorithm

Consultants have more important work to do with clients than spending days crunching numbers and performing diagnostics that AXAT makes repeatable and deliverable instantly. The algorithm triangulates the three lenses to create meaningful and deep diagnostics that help consultants get the punchline faster, and with diagnostics that can stand-up to client scrutiny. It runs on every capability, for every client, at machine speed — and reports how confident it is in each reading.

01
Curate the expertise

Structured once as a Consulting Atomic Model.

02
Create and collect

Instruments generated from the model; bots or consultants gather the data.

03
Read it with AXAT

Three lenses in. A defensible read and the advice that follows.

04
Share with the client

Links or an AX Portal, in your brand.

05
Read across the base

Benchmarks, movement and the gaps that recur.

One curated model runs the assessment to collect the data, produces the outputs, and stays behind the client-facing AI. The consultancy's judgement is present at every step — including the ones it is not in the room for. So juniors to seniors can run these assessments with confidence

03 ·The two scoring rules
Rule 1 — capability level
The weaker lens wins.

A capability scores at the lower of evidence and symptoms, never at the claim. Evidence sets the floor; symptoms cap it. The claim is held separate and tested against both.

Rule 2 — firm level
Levels are cleared, not averaged.

A level is reached only when every capability has reached it. Strength in one area does not buy back weakness in another. A mean is used as context, not as the answer.

What it asks of you. AXAT costs no more to run than a self-rating survey, and reaches a meaningful diagnostic faster. It asks a little more only where you choose to close the evidence gaps it identifies, which is what lifts the confidence score — and that is the point, because the evidence layer is where the value sits. The effort is front-loaded into building your model: once that exists, every new assessment is configured in minutes.

02 · How it works02 / 04
AXAT · Product fact sheet
The three lenses and the two rules
04 ·Why three lenses
A score is only as strong as the evidence beneath it.
A doctor never diagnoses from one answer.

They ask how you feel, examine you, then send you for tests — three instruments, none trusted alone. A good consultant already works this way. What they cannot do is work that way on every capability, for every client, at speed. That is the part AXAT automates.

What people say
Perception

Leaders and teams rate the capability. The score and the spread of opinion stay visible instead of being averaged away.

What can be shown
Evidence

Named artefacts, marked in place, partial or absent. Not "do you have a process?" but "show me the one you used last month".

What is experienced
Symptoms

If this capability were weak, what pain would surface? Enough stakeholders reporting that pain counts as a signal, whatever the documents say.

ONE CAPABILITY, THREE READINGS Execution & Delivery L1L2L3L4L5 PERCEPTIONEVIDENCESYMPTOMS 1.9 · the claim1.7 · sets the floor2.1 · caps the score AXAT SCORE 1.7 capped by evidence

The stakeholders claims L1.9 and symptoms suggest L2.1, but the artefacts only support L1.7 — so the capability reads at 1.7. Had the artefacts been strong and the symptoms weak, symptoms would have capped it instead. Disagreement is reported as a finding, not averaged away.

05 ·What the rules do to a real number
Eight capabilities, one engagement (example model) ClaimedCorroborated
L1L2 L3L4L5 corroborated average 2.29 floor 1.67 sets the level Knowledge &Intelligence OutcomePathway Execution &Delivery Leadership &Operating Governance CommercialModel Technology &Platform ClientEngagement
Panel's own average
L2.90
Corroborated average
L2.29
Level actually cleared
L1

Four capabilities sit on the floor at 1.67, so nothing has cleared L2 and the firm reports at L1. Averaged instead, the same engagement returns L2.29 — more than a level above the position that exists; the panel's own average, L2.90, is nearly two. Anonymised engagement, consultancy growth model.

04 · Why three lenses03 / 04
AXAT · Product fact sheet
What the rules do, and what you get out
06 ·Diagnostic insight
Four kinds of gap. Four different fixes.

AXAT goes beyond a score. The AI reads across the assessment to surface cross-cutting themes and identify the root cause of what is holding each capability back. Documented evidence plotted against lived experience sorts every capability into one of four problems, each needing a different intervention. Two of the four are not fixed by writing anything down.

CARRIED BY PEOPLE EMBEDDED EXPOSED NOT LANDING DOCUMENTED EVIDENCE → LIVED EXPERIENCE →
Carried by people

No system behind it. Delivery feels acceptable because it lives in individuals, and works until those individuals are busy, absent or gone.

Exposed

No system, and delivery already shows it. Symptoms are visible to the client now, not later.

Embedded

System and practice aligned, and ahead of the rest of the firm. Protect these — do not spend scarce change capacity here.

Not landing

Artefacts exist and authoring is done, but embedding has not happened. This is an adoption problem.

07 ·What you get out

Not only faster consulting. A more scalable firm.

01
A position you can defend

Every capability scored, with the lens that set the level named and confidence reported separately.

02
A diagnosis, not a score

Which capability is constraining the outcome, and which kind of gap it is, so the fix matches the cause.

03
Repeatable advice

The next best step from your own model, so a mid-weight team gives the answer your best partner would.

04
Benchmarking that means something

Same model, same rules every time, so readings compare across clients, business units, suppliers and time.

05
Outputs in usable formats

Reports and board packs in your brand, produced from the assessment — plus exports for other workflows and LLMs.

06
Sharing that does work for you

Share document links or an AX Portal where the client explores the data and asks an AI grounded in your expertise.

More on TheAX

Every part of the platform has its own fact sheet, and a walkthrough on a completed assessment is the quickest way to judge the method for yourself. Both start at the website.
hello@theax.ai  ·  Marlborough, United Kingdom

Find everything at
theax.ai
Also in this set: Overview · Product Book · Assessments · Portals · Assessment Intelligence04 / 04
AXAT Download PDF