Assembly Service Models: Pure Placement or Full Turnkey
Underneath the price of an assembly order sits a decision that is rarely discussed explicitly: how the work is divided between the customer and the supplier. Two quotations for the same board may differ not because one factory is cheaper but because they describe different service models, with different amounts of the work included.
Choosing the model is a separate question from choosing the factory, and it is worth settling first. The model determines who buys the material, who owns the risk of a shortage, who designs the panel, who writes the test and who holds the records when the order is finished.
Pure Placement
In its simplest form, the customer supplies the bare boards and the components, and the factory performs the placement and soldering. The advantage is transparency: the material is chosen and paid for directly, and the quality of each part is known because it was selected by the customer.
The disadvantage is that the customer takes on the material work. Verifying that a delivered reel matches the specification, counting the quantity, ensuring the surplus exists and arranging the delivery is all real effort, and it is effort that a factory with an established purchasing route can often absorb more cheaply.
This model suits a project that already has a supply route, where the parts are unusual or where the customer’s own purchasing department already holds the material. It is less suitable for a first order where the availability of a single device can determine the schedule.
Partial and Full Turnkey
A partial turnkey arrangement sits between the two extremes. It is the common arrangement in which the factory buys the components while the customer supplies the boards, or the reverse. It exists for practical reasons: a customer with a long-standing relationship with a board fabricator may keep the board supply, while the component list is easier for the assembly house to handle.
Under a full turnkey arrangement the factory handles everything from the raw board to the assembled unit, and the customer’s involvement reduces to the data and the acceptance criteria. The benefit is a single point of responsibility. Where a board does not fit, a part does not work or a schedule slips, there is one party to reason with rather than a chain of them.
The trade-off is visibility. The customer no longer sees the material cost directly, and the arrangement depends on the quotation being itemised enough to show how the material, the surplus and the allowance have been treated.

Consigned Material and Where the Risk Sits
Where the customer supplies some or all of the parts, the question that decides the working relationship is what happens when the material is wrong, short or late.
A shortfall discovered at the machine stops the run, and the cost of that stop belongs with whoever accepted the responsibility. Where the consigned quantity was specified as exactly the number of boards multiplied by the number of positions, the shortfall is a design of the arrangement rather than an accident; an allowance for attrition is what prevents it, and it has to be agreed as a number rather than assumed.
The same applies to quality. Where a consigned part arrives oxidized, mis-packaged or in a package that does not match the drawing, the discovery should happen at goods-in rather than at the line, and the record of the check is what allows the question to be settled by evidence rather than by argument.
Choosing Between the Mixed Arrangements
Where the work is divided, the practical question is which half each party should hold. The board and the components have different characteristics: the board is a single item with a process behind it, while the component list is many items with a purchasing process behind it.
Dividing along that line usually makes sense. A customer who buys devices in volume for another product, or who has a design in which a particular part is critical, keeps the components; the boards and the assembly go to the supplier who will build them. A customer with a long relationship with a board fabricator and no purchasing route keeps the boards, and the assembly house buys the components from the bill of materials.
What should be avoided is a division created by accident. Where both parties assume the other is buying a particular item, the omission is discovered when the material list is assembled, which is later than it needed to be. A single statement of who supplies what, sent back with the quotation, removes the possibility at no cost.
The service model changes the schedule as well as the price. Under a turnkey arrangement, the material can be ordered as soon as the order is confirmed, and the factory can plan the whole sequence around a single commitment.
Where the customer supplies material, the schedule includes a dependency that the factory cannot control, and the placement cannot begin until the last part arrives. The same is true of the boards. Each element supplied by the customer adds a handover point at which the date can be lost.
That is worth weighing explicitly. The apparent saving of a supply arrangement is real, but it is purchased with an exposure, and the exposure is largest on the first order when nobody yet knows how reliable the supply will be.
<img src="https://www.gopcba.com/wp-content/uploads/2020/12/project_image_10.jpg" alt="consigned material prepared for assembly” />
Documentation and Responsibility
Whichever model is used, the division has to be visible in the records. A delivery that contains boards bought by the factory and components supplied by the customer needs to identify which is which, because a fault traced to a component has a different owner from one traced to a process step.
The traceability requirement is the same in every case: the board revision, the material lots, the programme version and the test result, linked to the units delivered. What changes is who is able to produce each part of that record.
Where the material is consigned, the customer already holds the lot information and the factory holds the process record. Where the order is turnkey, both are held in one place, which makes a later investigation faster but also removes the customer’s independent view of the material.
Choosing the Model
The decision usually follows from four questions. Does the customer have a purchasing route that is faster or cheaper than the factory’s? Are the parts unusual enough that the customer wants to control them directly? Is the schedule sensitive to a single device, and if so who is better placed to secure it? And how much administrative work is the customer willing to absorb per order?
For a first batch, the turnkey arrangement is often the more practical choice, because it removes the dependencies that are hardest to manage at a distance while the product is still unfamiliar. For a mature product built regularly, the balance may move, particularly where the customer already buys the devices in volume for another purpose.
What matters is that the choice is deliberate and written into the order, since the model determines the answers to the questions that arise later. The work itself is unchanged: SMT assembly, through-hole assembly where inserted parts are present, verification through PCBA testing and the controls under quality management.
FAQ
Which model is cheapest? Not necessarily the one with the lowest quotation. Consigned material reduces the factory’s price while adding purchasing work and schedule risk to the customer.
What should be agreed when material is consigned? The attrition allowance, the check performed at goods-in, and who bears the cost of a shortfall found at the machine.
Why choose turnkey for a first order? Because it removes handover points that are hardest to manage while the product and the supply route are still unfamiliar.



