Priority Rules Written in Advance
Why a priority rule has to exist before the shortage, the four rules in common use, and what makes one defensible rather than merely stated.
A priority rule decides which claim gets capacity when not all of them can. Its only useful property is that it existed before the shortage, because a rule invented during one is indistinguishable from a decision about the people involved.
The allocation decision in “Priority Rules Written in Advance” should be made from visible constraints rather than an assumed headcount. Used carefully for task switching cost, explore the product context here can contribute recent workload context without replacing the documented priority rule or the manager accountable for the trade-off.
Most organisations have no rule and believe they do, which is a specific and recognisable state: everybody can describe the principle and nobody can predict the outcome.
The assumptions behind “Priority Rules Written in Advance” can also be tested against Bain organisation insights. Record which definitions transfer to the local plan and which do not.
The four rules in common use
Contractual and regulatory first, then value, then date of request, then seniority of claimant. Most organisations use a mixture, applied inconsistently, and the inconsistency is where the damage is.
Any of the four is defensible. Switching between them case by case is not, and it is what produces the belief that allocation is political.
Contractual and regulatory first
The only rule that is genuinely forced: obligations with external consequences come before internal preferences.
It is also the rule most often honoured in principle and breached in practice, because regulatory work is dull, has no sponsor present, and loses to whoever is in the room.
Value, and why it is hard
Ranking by value requires comparable value estimates, which almost no organisation has. Where they exist they are produced by the claimants, which makes the ranking a function of who writes the best business case.
Value is the right rule and the hardest to apply honestly. Where it is used, the estimates should come from one place rather than from each sponsor.
Date of request
Crude, transparent and surprisingly defensible. First come, first served has no judgement in it, which is its weakness and its strength.
It works best as a tie-breaker inside a band rather than as the whole rule, because used alone it rewards early speculative claims and punishes work that becomes necessary.
Seniority of claimant
The rule nobody writes down and many organisations use. It is predictable, which is its only merit, and it reliably produces a portfolio that reflects the hierarchy rather than the work.
Where this is the operating rule, saying so is better than pretending otherwise: at least then the people whose claims lose know why.
What makes a rule defensible
That it was written before the case, applies to all claims equally, and produces the same answer when applied by somebody else.
The third test is the real one. A rule that gives different answers depending on who applies it is a description of a judgement, not a rule.
The band approach
Most workable in practice: three bands — must, should, could — with contractual and regulatory work in the first, and a simple tie-breaker inside each band.
Banding is coarse enough to be agreed in advance and fine enough to decide most cases, and it avoids the false precision of a ranked list of forty claims.
Who sets it
The level that can cancel a claim, once, in advance, and in writing. Not the planner, who would then be deciding, and not the sponsors collectively, who cannot agree a rule that disadvantages any of them.
A rule set by the portfolio owner and published is the normal arrangement and takes one meeting.
Reviewing the rule, not the case
Annually, away from any live shortage. The question is whether the rule produced outcomes the organisation would defend, not whether any particular claim should have won.
Reviewing a rule during a shortage is how it gets amended into irrelevance, which is the end state of most priority frameworks.
The bands as they should be worded
| Band | What goes in it | Tie-breaker |
|---|---|---|
| Must | Contractual, regulatory, safety | Earliest deadline |
| Should | Committed work with a sponsor and a date | Date of confirmation |
| Could | Provisional and assumed claims | Value estimate, where one exists |
Three rows. They fit on the same page as the account, they can be agreed in one meeting, and they decide perhaps nine shortfalls in ten without anybody having to arbitrate.
The claim that sits outside every band
There is usually one: urgent, unsourced, from somebody senior, arriving mid-period. The rule exists precisely for it, and the rule will be tested by it.
Honouring the rule once, visibly, in that case is what establishes it. Making an exception the first time it is inconvenient means the rule will never be used again, and everybody will know that within a week.
What a rule does not do
It does not make the shortage less. It makes the allocation explicable, repeatable and survivable, which is all a rule can do.
Organisations expecting a priority framework to resolve overcommitment are using it as a substitute for the subtraction, and the subtraction is what actually tells them the work does not fit.