Sample and Prototype Process for Custom Displays: What Happens at Each Stage

A custom display programme usually runs through three or four sample stages, and each one exists to answer a specific question. Programmes that treat the stages as deliveries rather than as decisions tend to repeat work, and programmes that skip a stage discover the same information later at higher cost. Knowing what each stage proves is what makes the schedule predictable.

Part of the Custom Displays, Sourcing & Manufacturing guide — 46 articles on tooling, sampling, MOQ and supply-chain risk.

Last updated: October 6, 2026

What each stage is for

Each stage closes a different uncertainty.

Sample stages and the question each answers

Stage	Question answered	Typical content
Evaluation sample	Does the concept function and fit?	Standard module, approximate mechanics, working interface
A-sample	Does the design work with the intended materials and format?	Custom front, first mechanical design, functional interface
B-sample	Does it meet the environmental and optical requirements?	Production-intent materials, bonding, coatings
Production sample	Can the line build it consistently?	Production tooling, process parameters, inspection standard

The distinctions matter commercially: an evaluation sample validates an idea cheaply, while a production sample validates a process expensively. Asking for production-level confidence from an evaluation sample is the most common way to spend money early and learn nothing.

What to check when a sample arrives

A short list, applied every time.

• Mechanical fit against the drawing, with the enclosure if available.
• Optical performance in the intended lighting condition, not on a bench.
• Interface and timing behaviour with the actual host.
• Touch behaviour with the gloves and conditions in use.
• Front appearance under the agreed illumination, against the boundary samples.
• Assembly and disassembly, as a service technician would perform them.

Recording the results with photographs and measurements, rather than reporting impressions, is what makes the next stage faster. The evaluation should also record what was not tested, so that the outstanding questions are visible rather than assumed.

What each stage costs and why

Cost rises with production intent.

Evaluation samples often use standard parts and can be inexpensive. A-samples introduce custom front tooling and hand assembly. B-samples use production-intent materials and processes, including bonding and coating, and carry process setup cost. Production samples use the actual tooling and process, and their cost is largely the setup and yield of that build. Where a programme wants to compress the schedule by skipping a stage, the saving is usually smaller than the cost of discovering the same issue during a production build.

The practical approach is to decide at the outset which stages are necessary for the product’s risk profile, and to state that decision in the programme plan with the reasons recorded. Workmanship expectations for the assemblies produced at each stage are published by bodies such as IPC, and environmental test capability by chamber suppliers such as Thermotron.

Timing expectations and what controls them

Two variables dominate: tooling and material availability.

A sample schedule is driven by the tooling lead time for any custom items and by the availability of the panel and materials. Where the design uses a standard panel and a custom front, the schedule is relatively short. Where a panel-level customisation is involved, the lead time extends to the panel maker’s own cycle. The programme’s most reliable control is to freeze the specification before tooling and to obtain the supplier’s assumptions about material lead time in writing.

Asking for an accelerated sample is possible when the constraint is production scheduling, and much less possible when it is a tooling or material lead time.

What Each Payment Gate Should Buy

A payment gate is only useful if it releases against a named deliverable and an inspection you can actually perform, otherwise it becomes a schedule milestone that transfers cash without transferring risk. The table links each stage to what arrives and what to check before paying.

The list is a programme-control model, not a price list: amounts, stage names and what counts as acceptance depend on the project and on the supplier’s own process. What transfers between projects is the structure — money moving against artefacts, with an inspection between the two.

Stage	Deliverable	What the gate normally releases	What to inspect before releasing it
Specification baseline agreed	Controlled specification + mechanical drawing	Engineering deposit	Drawing checked against the enclosure and connector space
A-sample (form and fit)	A-sample units + test report	Sample payment	Active area, thickness, FPC exit, connector and mounting hole positions
B-sample (production intent)	B-sample + optical/electrical data	Tooling payment	Gamma, white point, brightness, uniformity and touch performance on the exact stack
Golden sample approval	Signed golden sample + final drawings	Pre-production payment	Change-control levels agreed and recorded
Pilot / PVT	Pilot report + first-article inspection	Balance per schedule	Process capability and traceability records, not only passing units

Two failure patterns recur. Paying tooling before the B-sample optics are confirmed moves the risk of a stack change onto the buyer, because the tooling is already cut for the earlier geometry. And accepting a golden sample without recording the change-control level leaves the first production batch free to drift within “normal” tolerances that were never written down.

What to ask for as evidence. Request a stage plan with deliverables and acceptance criteria, the inspection report formats used at each gate, the change-control levels that apply after golden-sample approval, and the document set you will receive at handover. The quality assurance and factory compliance page lists the records that should accompany each stage.

Payment gates and what they should be attached to

Attach payments to outcomes, not dates.

1. Tooling payment against the release of the drawings, not against a project start.
2. Sample payment against the shipment of the defined stage.
3. Production payment against approval of the sample and the inspection standard.
4. Final payment against delivery and acceptance.

Attaching the tooling payment to drawing release is the most useful gate, because the drawings are what the buyer can review and approve before money is committed. CDTech’s notes on the sample process and tooling describe the deliverables to expect at each point.

Common programme failures and how to avoid them

Five patterns, all recognisable.

1. Tooling released before the specification is frozen, producing a specification change during tooling.
2. Environmental requirements stated after the sample stage, requiring new materials.
3. Optical acceptance judged by eye at the sample stage rather than against an agreed standard.
4. Interface constraints discovered during integration rather than at the A-sample.
5. Volume expectations not stated early, so tooling was chosen for a volume that will not materialise.

Each of these has a low-cost prevention: freeze the specification, state the environment early, agree the acceptance standard, confirm the host interface, and state the volume. The prevention is cheaper than the stage repeat in every case.

A stage-by-stage checklist

What to have in hand before moving on.

Gate criteria by stage

Gate	Before proceeding
Evaluation to A-sample	Concept validated; envelope and interface confirmed; volume band agreed
A-sample to B-sample	Mechanical design frozen; optical acceptance standard agreed; material list defined
B-sample to production sample	Environmental tests passed; bonding and coating processes defined
Production sample to release	Process capability shown; inspection standard with boundary samples signed; documentation package complete

Moving through these gates in order, with the evidence recorded at each, is what makes the final stage a confirmation rather than a surprise. Where a programme has to compress, it is usually better to compress the time spent at each gate than to remove a gate altogether.

Managing the relationship during the sample phase

Information flows both ways.

Suppliers can only meet a schedule if they know which requirements are genuinely fixed and which are preferences. Telling the supplier that the enclosure is fixed but the coating is negotiable, for example, allows them to choose a coating that is available immediately. The same principle applies to the acceptance standards: sharing the boundary samples and the way they were established helps the supplier match them rather than guess.

A programme review after each stage – short, with the evidence in front of both parties – is the mechanism that keeps the schedule and the expectations aligned.