Court utilisation is a fraction, and the denominator is the argument

Utilisation is occupied court-hours over available court-hours. The disagreement is never the division — it is which hours count as occupied, and which were ever available to occupy. Three numerators are defensible: hours with a booking, hours somebody walked into, and hours that produced money in the period reported. All 3 describe the same week, and they can sit 19 percentage points apart.

So the facility meeting stalls. Programming quotes booked, the duty manager quotes attended, the board paper quotes revenue-bearing. The same shape appears wherever one word covers several measurements — the separate things a club calls delay when a stream buffers or runs behind the pitch.

The three fractions, stated precisely

What belongs in the denominator

One test on every hour: could you have sold it with the staff, surface and licence you actually had that week? If yes it stays in, whatever you did with it. If no it comes out, and the deduction is written down.

• Out — the surface was unusable. Resurfacing, a flooded pitch, a lane closed for plant work. There was nothing to sell.
• Out — the floor was unstaffed. Those hours are unsaleable, and leaving them in makes any venue look worse than it manages.
• In — you chose to hold it. A committee block, a slot kept for walk-ups. Removing these deletes the decision the report exists to expose.
• In — it sold badly. Tuesday at 14:00 is genuinely available and genuinely empty. A denominator that drops unpopular hours can only go up.
• Decide first — the bookable unit. A pitch sold whole or as 2 halves has 2 denominators for the same grass, settled by modelling pitches, courts and lanes as bookable units.

One week, four courts, three headlines

That 45-hour gap is rarely all no-shows: some is waitlist spots released too late to use, traced in full class, empty floor. Attended is also only as good as the door data underneath it, and a turnstile admitting people it should not inflates the middle fraction — the fault in lapsed members still getting through the door.

Peak and off-peak: where the average stops meaning anything

The 65% is arithmetically correct and operationally useless. A venue at 93% in peak has a capacity problem another court would solve. A venue at 65% flat has a demand problem no court will fix. The whole-week average is the one figure that cannot tell you which venue you are running.

A utilisation percentage with no stated denominator is not a measurement. It is a position in an argument.

The reporting shape a venue actually needs

1. Per surface. Court 1 by the entrance and court 4 at the back never perform alike, and the average hides both.
2. Per hour band. Define bands once, in writing. A boundary moved by an hour changes peak utilisation with nothing happening in the building.
3. Per usage class. Member booking, block hire, coached programme, casual pay-and-play, internal use, comps and credits.
4. With deductions shown. Available hours, hours deducted and why, beside the percentage.
5. Against the same week last year. Term structure makes month-on-month movement mostly calendar noise.

Usage class is where the fractions pull apart hardest. A make-up is attended but was paid for earlier, lifting attended utilisation while adding nothing to this month's revenue-bearing figure — the scoping problem in a credit is an entitlement with a scope and an expiry. Block hire is the reverse: revenue-bearing whether or not the school turns up.

None of this needs a new platform. It needs booking, check-in and billing joined on a common occurrence key and reported once — ordinary internal tools and ops work. Capacity policy, allocation and registers sit under scheduling, attendance and the facility, within our sports, fitness and athletics practice.

• utilisation
• facility reporting
• metrics
• capacity