Every desk in the school was 700 mm, and the factory built exactly what the order asked for

The situation

A school group in northern France was fitting out fifteen rooms across two sites, fourteen classrooms from Reception through Year 6 plus a shared library and staff area, and put 480 student desk-and-chair sets on order with a factory in Foshan. The purchase order carried a single line for the desk: single pedestal, height adjustable 580 to 700 mm, desktop 600 × 400 mm, 18 mm particleboard faced with 0.8 mm melamine, frame in RAL 9010. One line for the chair: seat height 440 mm, fixed steel frame, matching RAL 9010.

The order was a repeat. The school group had bought the same model the previous year for a new secondary block, where every room was Years 7 to 11, and a 700 mm desk with a 440 mm seat is right for children of eleven to sixteen. Nobody re-read those two lines when the order changed from one secondary block to a primary school. The specification was not wrong about anything it said. It simply said nothing about who was going to sit at the desk.

Sixty per cent of the order value was unpaid when the goods were finished and packed.

The order was complete about the desk and silent about the children

Look at what the order did specify. The desktop was a size, a material and a facing. The frame was a colour. The adjustment was a capability with a range attached. What it did not specify was the allocation: which classrooms took which year group, which size mark each room was built for, how many sets of each size, and what the desk-and-chair pairing was in each room.

Every desk and chair sampled on site conformed to the order. The height at the top stop measured 700 to 702 mm against 700 mm ordered. The lowest stop measured 578 to 580 mm against 580 mm. The desktop measured 597 to 600 mm by 398 to 401 mm against 600 × 400 mm. The seat measured 438 to 441 mm against 440 mm. Not one of those readings was a fault, and not one of them was any use to the school.

The fact that mattered was the absence of a number, and no amount of measuring the goods would ever have produced it. It came out of a question the inspector asked: which year group goes in which room, and against which size. The school group sent a one-page room list within the hour, and it was the first time anyone had put the fifteen rooms and the 480 desks on the same page.

How the desk was measured, and what each number proves

Four measurements carry this case, and the discipline of a school desk inspection china is mostly in which of them to take and how, because each one has a method that decides whether the number means anything.

Height is measured from the finished floor to the top surface of the desktop, at the front edge and again at the rear edge, with the desk standing on a level floor and a gauge reading to 1 mm. Not to the frame, not to the underside of the top, not to the highest point of the mechanism. The rear-edge reading is not redundant: a top that is level front to back tells you nothing about a top that is level left to right, and a desk that rocks towards one corner will pass a single front-edge reading all afternoon.

Adjustment range is recorded as the lowest and highest stops, the pitch between them, and the travel between the two. A range without a stop count is a claim; a range with a stop count is a measurement, because it tells you what a child can actually be put at. This column had a stop every 10 mm from 580 to 700, so the desk could be put at 590 and 640 exactly and at nothing below 580. Those two facts settle the case before anyone reaches for a tape measure twice.

The desktop is measured at both long edges and on the diagonal, for squareness, and for the overhang beyond the frame. A 600 mm top on a 480 mm frame puts 60 mm of top outboard of the frame on each side, and that is the zone where a child's shins find the structure.

The chair is measured on its own terms: the seat height, and whether it adjusts at all. Then the desk-to-seat differential is worked out, because a desk and a chair are only a set when the difference between them is right, not when each one is right on its own.

Why 700 mm cannot serve a five-year-old

European school furniture is sized, not guessed. EN 1729-1 sets six size marks, each pairing a seat height with a table height across a range of child stature, and EN 1729-2 sets the strength and stability requirements that go with them. The desk and the chair carry the same mark, and the mark is the whole point: it is what tells the school which children a set is for, and what to reorder when the cohort moves up a year.

The size mark a room needs follows from the children in it. A Reception or Year 1 classroom is a size 2 room. A Year 2 or Year 3 classroom is size 3. A Year 4 to Year 6 classroom is size 4. A staff room is size 5. The fifteen rooms in this order were four different sizes, and the order contained one.

This is why nothing about the delivery looked wrong from the loading bay. At 700 mm and 440 mm the desk-to-seat differential is 260 mm, and 260 mm is exactly the difference between a size 4 desk and a size 4 chair. A buyer checking only the difference between desk and seat would have concluded the sets were right for Year 4 to Year 6, which is where most of them were going. But a differential check on its own is not enough, because the same 260 mm can be reached at 640 mm and 380 mm or at 700 mm and 440 mm, and the second of those is a size 5 pair scaled up by 60 mm. The proportion was right and the scale was wrong.

Scale is what a child actually feels. A size 2 chair puts a five-year-old's feet flat on the floor with the thighs roughly level. A 440 mm seat puts the same child 130 mm above where their feet can reach, so their legs hang, their weight comes forward onto the front of the seat, and their elbows end up below the desktop instead of on it. The desk is the same story from the other side: a 530 mm desktop for that child puts the surface level with their elbow, so the forearms lie flat on it and the shoulders stay down. At 700 mm the surface sits at the base of the child's chest, and the only way to write is to lean over the front edge with the forearms angled up and the shoulders lifted.

The range the school needed, and the range the desk had

Here is the part that decides the outcome. The school needed desktop heights from 530 mm in Reception up to 710 mm in the staff area, a span of 180 mm across four size bands. The desk on site had 120 mm of travel, from 580 mm to 700 mm.

That 60 mm shortfall is not a tolerance argument. It is the difference between the desk reaching a size 2 classroom and not reaching it, and no amount of adjustment closes it. The desk reaches 590 mm for Year 2 and Year 3 and 640 mm for Year 4 to Year 6, and it comes within 10 mm of the 710 mm an adult desk needs. It cannot get below 580, which is 50 mm above the 530 mm Reception and Year 1 require, so 80 of the 480 sets have no correct height at all.

The chair is the cheaper failure and the more widespread one. A 440 mm fixed seat is above every band the school needs except the staff area, and the error grows as the child gets smaller: 130 mm too high for Reception and Year 1, 90 mm for Year 2 and Year 3, 60 mm for Year 4 to Year 6. Four hundred of the 480 chairs are wrong, and every one of them looks identical to a chair that is right.

This is where the adjustable desk earns its place, because it changes what the remedy costs. If the desk could not be adjusted, the whole order would be scrap. Because it can, the 320 sets going to Year 2 and above need nothing but a chair of the correct seat height and a desk set to the right stop. Only the 80 sets for Reception and Year 1 need a different product entirely.

What the report recorded, and what it did not

The report was careful about who the finding belonged to. The goods conformed. Every measurement taken on site matched the purchase order, so there was no basis on which to report a manufacturing fault, and the report did not.

What it recorded instead was the requirement, taken from EN 1729-1 and from the school group's room list: fifteen rooms, four size marks, and 480 sets allocated as 40 Reception, 40 Year 1, 40 Year 2, 40 Year 3, 80 Year 4, 80 Year 5, 80 Year 6 and 80 for the library and staff area.

Then the arithmetic, which is where a report earns its place. Of the 480 sets, 80 are usable as delivered: the library and staff area sets, where the 700 mm desk and the 440 mm seat sit within 10 mm of the size 5 nominals. 320 need a chair of the correct seat height and nothing else, because the desk reaches 590 mm and 640 mm on its own stops. And 80 cannot be used in Reception or Year 1 at any setting, because the column stops 50 mm above the desk height those children need.

What the report did not do was decide what the school group should buy. Whether to take 400 replacement chairs and 80 replacement desks, to reallocate the Reception and Year 1 rooms into the Year 4 to Year 6 classrooms, or to ask the factory for a lower-travel column and re-ship is a commercial decision driven by the school opening date and the phased delivery programme. The report's job was to give the buyer the numbers that make the decision: 80 sets cannot go into a Reception room at any setting, 400 need a chair, and the desk travel is 60 mm short of the span the school needs.

It also did not forecast how the children would sit. Whether a Year 1 child at a 700 mm desk spends the year hunched over or merely uncomfortable is not something an inspection can determine from a desk standing in a factory. What an inspection can determine is the number, and the numbers were 50 mm and 130 mm.

The allocation table the order should have carried

One line in the purchase order; nine lines in the allocation table. This is what the order should have carried, and it is the same table the factory quotes against, the same table the report checks against, and the same table the school uses to reorder the cohort in three years.

What the specification should have said

Seven lines. Each one is free to write, and together they turn a purchase order that names one desk into one that names fifteen rooms.

• Name the standard and the parts. School furniture to EN 1729-1 for functional dimensions and EN 1729-2 for strength and stability, with test reports from an accredited laboratory issued at quotation stage. Designed to is not tested to, and the difference between the two is a document.
• Give the allocation, not a size. A table of room, year group, size mark, desktop height, seat height and quantity per band. One desk line covering fifteen rooms of different age groups is the defect, not the desk.
• Pair the desk and the chair in every line, and state both heights as absolute numbers. A desk and a chair of the right difference from each other can still be the wrong size for the child, which is exactly what happened here: the differential was a size 4 and the scale was a size 5.
• State the adjustment range per band, as a minimum and a maximum, with the stop count and the pitch. An adjustable desk whose lowest stop is 50 mm above the shortest child's height is not adjustable enough for the room, and the order should have said so.
• State the measurement method. Heights measured from the finished floor to the top surface of the desktop at the front edge, with a tolerance of plus or minus 5 mm on a nominal figure and no averaging across the batch. The gap between a nominal height and a measured height is where this entire argument lives.
• State the desktop size per band, from EN 1729-1, and the tolerance on squareness. A size 2 child needs a smaller working surface than a Year 6 child, and the two are not the same top.
• Put a size check before the run. Five sets per band presented for check before the batch is built, and no band shipped against an unchecked sample.