A buyer who places a modest order and then hears nothing for a while tends to assume the order was forgotten, or that the factory prefers bigger customers. Neither is usually true. The order entered a queue that is not sorted by arrival time, because arrival time is the least useful thing a planner knows about a job.
A shop floor is not one line. It is a set of machines, each with its own setup, its own operator skill, and its own tolerance for interruption. A job that arrives on Monday may suit a machine that is booked until Thursday, while a job that arrived on Wednesday fits a gap on Tuesday afternoon. Planning strictly by date would waste that gap.
So the queue is sorted by fit. Fit has several parts: whether the material is in stock, whether the tooling exists, whether the drawing is released, and whether the operation can run without breaking a longer setup already in progress. A small order that scores badly on those four points waits, and it waits behind work that scores well, not behind work that is large.
In practice, small orders lose time in three places, and only one of them is the machine.
The first is engineering. A drawing that looks complete to a buyer may leave a chamfer, a thread class, or a weld finish undefined. The factory cannot guess, so the job sits in a query pile, and every day in that pile is a day the machine was never the bottleneck.
The second is setup. A setup takes a certain amount of time regardless of how many pieces follow it. On a long run that time is spread thin; on a short run it dominates. A planner with one free window will put the long run in it, because the same window produces more output. This is arithmetic, not preference.
The third is inspection. Small batches are often sampled at a higher rate relative to volume, and if a first article is required the job stops until someone signs it off. That stop is invisible in a schedule that only tracks machine hours.
Planners are measured on two things that pull in opposite directions: throughput and lateness. Push throughput and the schedule fills with long runs. Push lateness and it fragments into small jobs that each pay a setup. The visible behaviour of a factory is largely a picture of which pressure is winning that month.
This explains why the answer to a question about your start date changes between calls. It is not a promise being walked back. It is a re-sort, triggered by a new job entering the pool with a better fit.
A factory that plans honestly will admit the re-sort happens, and will tell you which lever moves your job forward. Common levers are a longer run that amortises the setup, a material grade already in stock, or a later slot in exchange for a firm one. Each is a trade worth discussing rather than resenting.
A question about shipping invites a guess. Better questions produce better answers, because they constrain the planner to facts.
A planner who is asked these will usually answer at length, because they replace a vague complaint with a solvable problem.
Most of the delay that buyers attribute to the factory was created before the purchase order was sent. A drawing pack with unresolved tolerances, a material specification that names a grade nobody stocks, a finish callout that conflicts with the base metal: these are specification failures, and they cost the same calendar time as a machine breakdown.
The remedy is unglamorous. Send a drawing pack reviewed by someone who will have to inspect the part. State the function of each critical dimension, so the factory knows which tolerances can be discussed and which cannot. Name acceptable substitutes and the ones that are not. Confirm the inspection level in writing.
A factory with more than 50 employees and four facilities in Shanghai can absorb a certain amount of ambiguity, but absorbing it costs days nobody invoices. Founded in 2009, the operation has seen enough drawing packs to know which will run cleanly within the first hour of review. That judgement is available for the price of asking before release, and the background sits on the company page.
The largest job in the building is sometimes deliberately held, precisely because it is large. A long run that starts a day late still finishes on time. A short run that starts a day late does not, because there is no slack inside it.
Planners use big jobs as ballast. They are placed where they fill a window completely, and moved when a small job would otherwise miss. From outside this looks like favouritism; from inside it is the large order absorbing disruption, which is what its size allows. Queue position is not a ranking of importance.
A first order is the worst case for fit, because nothing about it is known: no tooling history, no material history, no inspection history. That is not a reason to expect poor service, but it is a reason to expect the first order to move more slowly than the fifth, and to plan your own commitments accordingly.
Buyers who understand this send a smaller first order than they need, to establish the relationship on a job that can be checked quickly. They ask for the inspection report with the shipment rather than after it, and they keep the drawing pack stable between orders, so the second release needs no re-review.
A small order waits because it fits badly, not because it is small. The fixes are specification clarity, a realistic view of setup cost, and a conversation about which lever to pull. A factory that explains this is being more truthful than one that simply promises a date and hopes.
Buyers who want to test that claim can start with the parts listed under other parts, the customisation process, the stainless exhaust systems and titanium components, and the dealership terms, then ask the specific questions above on the next order.
