You ordered an envelope, not an item
A few things on every project take longer to make than the design that specifies them takes to finish. A main transformer, a pressure vessel, a bespoke crane, a switchboard.
Wait for the design and the item arrives after it is needed. So the order goes out early, and it can't be placed against a drawing, because the drawing won't exist for another year. It is placed against parameters instead — a rating, a duty, a set of dimensions, a weight, connection positions.
Those parameters are the real purchase. From the day the order is signed they are constraints, and everything drawn afterwards has to fit inside them.
Nobody experiences it that way at the time. What is experienced is a procurement decision, taken to protect a date.
The direction of causation inverts
Ordinarily the design determines the equipment. A load is calculated, a duty established, a machine selected to meet it, and if nothing suitable exists the design changes until something does.
On a long-lead item that runs backwards. The machine is committed first, and the design is then developed to accommodate what is being built. The foundation fits the base plate that was ordered. The building fits the envelope. The access route fits the shipping dimensions.
Nothing announces this inversion. No document says from here the equipment governs, and there is no moment at which somebody signs to that effect. The order is a commercial act, filed in a procurement system, and its consequence is a design constraint that turns up months later as a discipline finding that something has to move.
The people who need to know are designers. The document that would tell them is a purchase order.
Who decided which items get this treatment
The list of long-lead items is a front-end deliverable. It is named as one of the things a gate expects to see, alongside the conceptual schedule and the budget estimate, at a stage well before award.
Which puts it in the same category as everything else in week 5: settled by people the delivery team never meets, on the basis of a design that was still conceptual.
The same source explains why the front end spends money getting further than a concept before it stops. A more developed basic design is what makes committing to these items possible at all, and reduces how much of the detailed design is still uncertain when they are ordered. So a front end that ran short doesn't only produce a vaguer estimate. It produces a longer list of things ordered on less information.
And it produces a failure that can't be repaired later. An item that should have been treated as long lead and wasn't identified as one gets ordered in the ordinary course, months after it needed to be. By the time anybody notices, the only remedies left are the expensive ones — expedited manufacture, air freight, or a different item that fits the calendar rather than the design.
What being wrong actually costs
This is the part that surprises people the first time and reads as obvious afterwards.
If the committed envelope turns out to be wrong, the cost isn't the item. The item can be modified, or in the worst case replaced, and a vendor will quote for that.
The cost is everything that was designed around it. A base plate a little larger than assumed is a foundation redesign, a reinforcement change, possibly a building dimension. A weight above what was assumed is a crane, a haul route, a floor slab. A connection in a different place is a run of pipework and the drawings that carry it, each of which goes back through the review loop from week 15, at 21 days a lap.
None of which appears as a change to the equipment order. It appears as engineering rework, distributed across disciplines, attributed to design development.
So the exposure isn't proportional to the value of the equipment. It is proportional to how much design was completed against the assumed envelope before the real one arrived.
Which has a scheduling consequence worth acting on. The gap between placing the order and receiving the vendor's certified drawings is the window in which an error is cheap, and it is a window that closes gradually as disciplines draw against the assumption. Anything that shortens it — and the drawing loop from week 15 is what governs its length — is worth more than it looks on a procurement tracker.
Early isn't a win
There is an instinct that an item arriving ahead of schedule is good news. For a long-lead item it is a transfer of problem rather than a solution to one.
Something delivered eight months before it can be installed is eight months of storage on a site that was never laid out for it, eight months of a preservation regime somebody has to perform — rotating shafts, changing desiccant, maintaining heaters, keeping records that a warranty may depend on — and eight months in which a vendor warranty measured from delivery runs against an item nobody is using.
The reporting makes it worse. A delivery achieved ahead of programme is a green line in the month it happens, and the storage, the preservation and the warranty consumption arrive in no report at all.
The construction extension notes the split that makes this awkward: an owner may buy the equipment while the contractor unloads, stores and preserves it until installation. Interfaces Week 14 takes that relationship apart properly, including having no privity with the vendor and no sight of their programme. What belongs here is narrower and lands on the same desk: the obligation to preserve arrives with the item, and it was priced, if at all, at tender.
System design
Row three is the whole week in a record, and its required quality is the unusual part: it has to capture what the order left out. An envelope with four parameters fixed and two unstated is a different risk from one with six fixed, and only the page makes that visible.
| Record | Produced by | Required quality | Verified against | Feeds |
|---|---|---|---|---|
| Long-lead list | A front-end gate, before award | States why each item is on it, so an omission is arguable | The lead times behind the judgement | What gets ordered against parameters |
| The order, and its parameters | Procurement, early | The committed envelope legible as a constraint, not buried in a schedule | The quotation it accepted | Every drawing produced afterwards |
| Envelope page | You, one per item | Records what the order left unstated as well as what it fixed | The order and the enquiry | Each drawing check · the vendor comparison |
| Vendor certified drawings | The vendor, after award | Compared to the envelope page line by line, on arrival | What was ordered | Whether the design still fits |
| Preservation regime | Whoever holds the item | Names the actions, their frequency, and who performs them | The vendor’s requirements | Warranty · condition at installation |
Row two asks for something orders are poor at. A commitment that constrains design is recorded in a commercial document that designers have no reason to open, which is the seventh time on this track that a fact has been filed where the people who need it won't look.
Practical insight
For every long-lead item on your project, write down the envelope. One page per item, and do it in your first month rather than when a clash appears.
Write down the parameters that were fixed when the order was placed: overall dimensions, weight, centre of gravity where it matters, base plate and bolt pattern, connection positions and sizes, clearances required for maintenance, and shipping dimensions. Where your order states none of these, write that it states none — a blank is information you will want later.
Then use the page twice. Check each new drawing that touches the item against it, which takes minutes and catches a divergence while it is still a drawing rather than a foundation. And when the vendor's certified drawings arrive, compare them to your page line by line, because that comparison is the earliest moment you can know whether the envelope you bought is the envelope you are getting.
That second use is the one that pays. It is the same failure week 1 followed through six hands, caught at the only point where catching it is cheap.
Key takeaways
- Some items take longer to make than their design takes to finish, so the order can't be placed against a drawing.
- It is placed against parameters, and those parameters are the real purchase.
- From the day of the order they are design constraints, and everything drawn afterwards has to fit inside them.
- The usual direction of causation inverts: the equipment begins determining the design.
- Nothing announces the inversion, and no document records that the equipment now governs.
- Which items get this treatment is a front-end gate deliverable, settled before award on a conceptual design.
- An item that should have been long lead and wasn't identified can't be repaired later, only paid for.
- Being wrong about the envelope doesn't cost the item. It costs everything designed around it.
- So the exposure is proportional to how much design was completed against the assumed envelope, not to the value of the equipment.
- Early delivery brings storage, a preservation regime somebody has to perform, and a vendor warranty running against nothing.
- The obligation to preserve arrives with the item and was priced, if at all, at tender.
Records born here. The long-lead list and the gate that produced it · the order, and the parameters fixed by it · the envelope page per item, including what the order left unstated · the vendor’s certified drawings checked against that page · the preservation regime, and who is performing it · the warranty start date and what it runs from.
What is coming next
Phase C ends here. Documents are being issued, orders are placed, and things are being made in factories.
None of it is on site yet. What follows is the arrival of the first people and the first plant, and the discovery that a site has to be built before anything can be built on it — roads, power, water, offices, a fence, a place to put things down. That work is priced as a lump sum in a bill, it is planned by almost nobody, and it determines how everything after it flows.
Enjoyed this lesson?
Join with Google to get each new lesson the moment it's published — and help me see which topics matter most to you. No spam, one email a week, unsubscribe anytime.
Already following on LinkedIn works too — this is just for the weekly email.