enzwa lab · chassis · enterprise operating model diagnostic

What to fix first, and what fixing it is worth.

This page runs Chassis across the four organisation types it ships with, then follows one of them through the assessment. Every figure is computed by re-running the tool’s own scoring.

A tool for scoring how well a property function is organised, and working out what to fix first.

The problem

Every improvement plan has eight or ten things on it and a budget that covers three. The items that sound most valuable are the visible ones: new systems, better reporting, better data. Which of them will actually pay is a judgement nobody can show their working for.

What it does

Chassis scores an operating model across eight components, asks what level of maturity you are steering toward, and returns a sequenced plan with a number against each phase, costed against leaving things alone.

Printed technical documents overlapping in a loose grid, seen square-on from above: organisation charts of empty boxes, process diagrams and blank matrices, in flat shadowless light. — an operating model is invisible, and this is the nearest thing to a photograph of one

01 · the whole assessment

Eight components, four organisations, one grid.

The entire assessment on one sheet: four organisation types down the side, the eight components of an operating model across, each cell a maturity level from 1, ad-hoc, to 5, leading. Every section from here takes a slice of this grid and redraws it larger. Nothing new is introduced.

one cell is marked: the industrial function scores Process at 4 and the model counts 3.5 · swipe to see all eight components

02 · the split

Four of the eight hold the other four up.

The eight components sit in two layers. Vision and strategy, organisational design, governance and the service delivery model are foundations. Process, digital enablement, performance management and continuous improvement are enablers, and the tool weights the foundations more heavily because everything else runs on top of them.

layer	component
F	Vision & Strategy	foundation
F	Organisational Design	foundation
F	Governance	foundation
F	Service Delivery Model	foundation
E	Process	enabler
E	Digital Enablement	enabler
E	Performance Management	enabler
E	Continuous Improvement	enabler

03 · the ceiling

An enabler cannot out-run the foundations under it.

Chassis caps every enabler at the average of the four foundations plus one level. Score digital enablement at 4 while governance sits at 2 and the model records what the foundations will actually let you have. The technology archetype below averages 3.00 across its foundations, so its ceiling is 4.00, and digital is already sitting on it.

the dashed outline is a maturity level that could be bought and would not register

04 · the two investments

A level bought in the wrong place returns nothing.

Take the technology organisation and spend a full maturity level on the component it is already strongest at. Digital goes from 4 to 5 and the model returns nothing at all, because the ceiling has not moved. Spend the same effort on governance instead, from 2 to 3, and leave digital exactly where it is: $0.51M a year appears.

value released each year against the same $200M portfolio

The same shape holds for the industrial function, where process is scored 4 and counts as 3.50. Raising it to 5 returns nothing. Lifting all four foundations by one level, without touching a single enabler, returns $2.26M a year and carries process’s effective score up to 4.00 on its own.

The same documents gathered into one squared-up stack seen from above, the sheets beneath showing as thin parallel edges. — the order the sheets go in is the whole argument

05 · the sequence and the money

Foundations first, and every phase carries its own number.

The value model is a leak. At maturity level 1 the tool assumes 9% of annual portfolio spend is lost to the way the function is run, falling to 1.5% at level 5. Each full level is worth 1.875% of spend, which is $3.75M a year on the $200M portfolio the tool uses by default.

level	maturity	leaking	on $200M
L1	Ad-hoc	9.00%	$18.0M a year
L2	Developing	7.12%	$14.2M a year
L3	Defined	5.25%	$10.5M a year
L4	Managed	3.38%	$6.8M a year
L5	Leading	1.50%	$3.0M a year

The roadmap always sequences foundations first. Run the post-merger archetype, the weakest of the four, and it is leaking $14.7M a year against a $200M portfolio. Reaching Managed across every component recovers $7.9M of that and leaves $6.8M on the table, which is the residual the model says no operating model removes.

phase	what moves	value released
1	Lift the four foundations to Managed	$4.1M
2	Then process, digital and performance	$2.6M
3	Then continuous improvement	$1.3M

06 · what it is measured against

The leak is the tool’s own assumption.

Corporate real estate is usually an organisation’s second-largest expense after payroll, and vendor commentary published in 2026 puts the occupancy-cost gap between data-driven and reactive property functions at 15 to 20%. Chassis’s own leak curve sits inside that range, and the tool labels itself an illustrative model to be swapped for a client baseline.

Those two figures are the weakest external anchors used anywhere in this series, and both come from vendor material rather than research. They are recorded with that status in FACTS.md beside this page. Nothing in the argument above rests on either of them: the ceiling, the two investments and the sequence are all computed from the tool’s own engine.

07 · open it

Score your own operating model.

Chassis runs in the browser. Frame the brief, score the eight components against written descriptors for every level, and read back the profile, the value at stake and the sequenced roadmap. The scores are yours; the ordering is the model’s.

Open the diagnostic →