Home / Risk / Requisites catalogue
The hardest part of a design, and nobody has a tool for it.
Every method downstream — value engineering, FMEA, risk analysis, verification — reads from the requisites catalogue. Get it wrong and everything built on it inherits the error. Yet it is almost always written as prose in a document, where nothing can be checked and nothing can be counted. This one refuses a requirement that cannot be tested, and it is free.
Write the requirement
Tick what applies. For effort and readiness, tick what is large.
Tick what applies. For effort and readiness, tick what is large.
Six worth and exposure ticks set the letter; effort and readiness order the requirements inside it.
FUNCTION DEFINITION
CATALOGUE
Nothing is sent anywhere — this runs in your browser and closing the tab clears it. Use Copy catalogue to keep the result.
A function is a verb and a noun. Nothing else.
Value engineering has insisted on two words since the 1940s, and the discipline is the point: anything longer is a description, and descriptions cannot be measured.
Function is not action
A function is an abstract, solution-independent statement of what must be accomplished. It describes intent, never mechanism. An action is a physical doing — a lever rotating, a spring compressing — tied to one embodiment. State the action and you have fixed the solution before the analysis starts; every failure mode you then find is a failure of that mechanism, and the alternatives never get examined.
The noun carries the unit
Transmit force, retain pressure, resist deflection. A noun you can put a unit on gives a function that can be tested. A noun naming a thing rather than a quantity cannot: retain water has no unit, retain water pressure does. This tool refuses a requirement without one, which is the whole reason it is worth using instead of a table in a document.
Straight into a FAST diagram
Two-word functions are what a Function Analysis System Technique diagram is built from. Once the catalogue holds verb-noun pairs rather than prose, the how-why chain can be laid out directly — each function asking how it is achieved to its right and why it exists to its left. A prose requirement cannot enter a FAST diagram at all.
The same catalogue then feeds the risk analysis. A failure mode is a failure of a stated function, its impact is rated against the stated success criteria, and the test conditions say what would prove it. Three methods, one catalogue — which is why getting it right is worth this much trouble.
On importance
The tool offers MoSCoW because it is what people use. It is worth knowing what it does not do.
MoSCoW sorts requirements into four bands and stops. It provides no way to rank one Must-have against another, and the criteria separating Must from Should are a matter of judgement — both are documented criticisms, not opinions of ours. Forty Must-haves arrive unordered, which is the same defect a categorical risk priority has: it tells you that action is required and not in what order.
The alternatives trade one problem for another. Kano classifies by effect on satisfaction — basic, performance, delighter — which explains why something matters but needs survey data. AHP gives a genuine ranking from pairwise comparison, at N(N−1)/2 judgements: twenty requirements is a hundred and ninety, fifty is over twelve hundred. Weighted scoring gives an order, but the weights are themselves a judgement and can be argued with.
We think the answer is the one that worked for risk scoring: keep the band, then place an ordinal underneath it so the band no longer hides an unordered list. What that ordinal should be is not settled, and we would rather say so than ship a number nobody can defend.
That ordinal now exists: RIQ+ — Requirement Importance Quantified. Eleven ticks — six for what the requirement is worth and how far that worth reaches, three for what it will cost, two for whether the team can start — placed as 100×Value + 10×Effort + Readiness. The count of the first six sets the MoSCoW letter, so the band is derived rather than argued; the other two digits order the requirements inside it, cheaper first, readier next. 64 distinct scores, no ties, and a Could never outranks a Must. Press Work it out with RIQ+ under Importance above: it fills the letter in and keeps the score with the requirement.