PCB Design Review Order: Checking the Expensive Items First
Every design review finds something, and the value of the exercise depends on the order in which the items are checked. Reviewing the silkscreen before the stack-up wastes the meeting, because a change to the stack invalidates half of what was agreed.
Order the Review by the Cost of Change
The correct order is the reverse of the order in which the decisions were made. Stack-up and material come first, because they constrain everything else and because they are the most expensive to alter once the panel is ordered. Placement follows, then the critical routing, then the general routing, and finally the cosmetic items.
A review that begins with the silkscreen and reaches the stack-up at the end will spend its time on the cheap items and run out of attention before the expensive ones. Our design release checklist is ordered on that principle.
Stack-Up and Material Review
The stack-up review confirms the number of layers, the dielectric thicknesses, the copper weights, the symmetry and the impedance targets. Each of those has a consequence that is difficult to reverse, and each should be confirmed against the requirement rather than against the previous design.
The material review confirms the glass transition temperature, the dielectric constant at the frequency of interest, the thermal expansion and the availability. Availability is the item most often skipped and most often the cause of a schedule slip.
Placement Review
Placement is checked against the mechanical constraints first: the outline, the mounting features, the connector positions and the component heights. Anything that conflicts with the enclosure is a change that affects the whole layout, so it belongs early in the review.
Once the mechanical constraints are confirmed, the placement is checked against the electrical intent: the current loops of any switching supply, the decoupling capacitors close to the pins they serve, the separation of noisy and sensitive blocks and the thermal spreading of the dissipating parts.

Critical Routing Review
The critical nets are reviewed before the general routing, because the answer to whether they can be routed determines whether the layout is finished. A differential pair, a clock, a high current path and a sensitive analogue trace each have their own rules, and each is checked against its own reference.
The review should include the return path as well as the trace. A differential pair that is routed perfectly but whose reference plane is split underneath it satisfies the visible rules and fails the electrical one.
General Routing and Design Rules
The general routing is checked mechanically: minimum width and space, annular ring, via count and size, clearance from the outline, and the absence of unconnected nets. This part of the review is the easiest to automate, and the automation should be used so that the meeting can concentrate on the judgement calls.
The results of the design rule check should be reviewed as a list rather than accepted wholesale. A rule violation that has been deliberately waived is a decision, and it should be recorded rather than hidden in a suppression file.
Silkscreen, Mask and Marking
These are the cosmetic items, and they come last because they are the cheapest to change. The review confirms that the reference designators are legible and not over pads, that the polarity marks are present and correctly oriented, that the mask openings are correct and that the marking is present where it is needed.
The one exception is a marking that carries a functional requirement, such as a connector orientation that a production operator relies on. That belongs earlier, because it affects the assembly process rather than the appearance.

Manufacturability and Panelisation
The manufacturability review checks the design against the capability of the shop that will build it, including the minimum features, the aspect ratio, the finish and the panel arrangement. This is the point at which a design that is electrically correct is confirmed to be buildable at the quoted cost.
Where the design sits close to a process limit, the review should confirm that the limit applies to this shop rather than to a generic one. Capability lists differ, and a design that is routine for one supplier may be marginal for another.
Recording the Outcome
Every finding should have an owner and a disposition: fixed, accepted with a reason, or deferred with a date. Findings that are neither fixed nor explicitly accepted tend to reappear at the next review, which is a sign that the process is not closing.
The record also serves the next project. A list of the issues that recur across designs is the most efficient input to a design rule set, because each entry represents a mistake that has already been made once.
Process Control and Verification
The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed. A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance.
Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance.
Related reading: our fabrication notes, board quality and design release notes cover the same ground.
Checks Before Release
The checks that matter are the ones performed on the product rather than on a sample kept for the purpose, because a coupon that travels with the panel is the only evidence about that panel. Where a requirement can be measured, it should be measured at the point of manufacture and recorded against the board or the lot it applies to.
Verification and Records
A parameter that is set once and never re verified drifts, and the drift is usually discovered by a defect rather than by the record. The tooling, the material and the profile form one system, and a change to any of them should be assessed against the other two before it is released.
Where the process window is narrow, the measurement resolution has to be better than the window, or the data cannot distinguish a good part from a marginal one.
FAQ
How long should a design review take? Long enough to cover the stack, the placement and the critical routing properly, which for a moderately complex board is a couple of hours. A review that finishes in twenty minutes has usually not examined the expensive items.
Who should attend? The designer, someone who understands the manufacturing process and someone who owns the product requirement. Adding more people beyond that tends to lengthen the meeting without improving the outcome.
What does gopcb review before a panel is released? We review the stack-up, the material, the minimum features against our capability, the panelisation and the impedance targets, and we report anything that would affect the yield or the schedule. The review is recorded so that the customer can see what was checked.



