You want to test the market first. Not too many—around 3,000 pieces.
The quote comes back. Higher than you expected. Lead time isn’t much better than a full production run either. You hesitate.
That hesitation isn’t unreasonable. For small batch PCBA manufacturing, the cost structure is different. Volume isn’t there to spread the setup overhead. Procurement still has to happen. Production still needs to be scheduled. Testing still needs to run. The per-unit cost looks high because the fixed costs aren’t amortized across tens of thousands of units.
What you see is a quote that gives you pause. What you don’t see is the work that goes into making small batch PCBA manufacturing work at all.
Below are three real cases from small batch production. Each one illustrates the same point: the challenges in small batch PCBA manufacturing aren’t about the quantity. They’re about whether the process is designed to handle small quantities at all.
A client needed their product encapsulated with hot-melt potting compound.
In the lab, it was simple. Apply the compound by hand, hold the board in place for 30 seconds until it cures, move to the next one. Fast, easy, no issues.
Then came the 3,500-unit production run.
The problem became obvious immediately: workers had to physically hold each board in place while the compound cured. Thirty seconds per unit doesn’t sound like much—until you multiply it by 3,500. That’s nearly 30 hours of workers standing there, hands occupied, waiting for glue to dry. No breaks, no other tasks, just holding.
What we did:
We designed a custom potting fixture—a mechanical clamping system that held the enclosure and the PCBA in position, with a lever-operated clamp and guide slots precision-milled to match the PCB and housing contours. Workers placed the board, pulled the lever, and moved to the next unit while the fixture held everything in place.
The result: 3,000 units completed in a single day. From 30 seconds of manual holding per unit to zero holding time per unit. One worker could manage multiple fixtures simultaneously.
The lesson:
What works in a lab doesn’t always work in small batch PCBA manufacturing. A production line doesn’t need something that “can be done.” It needs something that can be repeated, quantified, and controlled at a consistent rhythm.
A client’s design used an OSRAM optical sensor. Their regular order volume was 3,000 units per batch, with occasional 1,000-unit top-up orders.
The manufacturer’s MOQ for this sensor was 1,500 pieces. The client didn’t want to buy a full reel for a 1,000-unit order. But buying from a distributor who would break the reel meant paying triple the price. The math didn’t work.
This isn’t uncommon in small batch PCBA manufacturing. Procurement problems aren’t usually about “can’t get the part.” They’re about “can’t get the part at a sensible price.”
What we did:
We had an established relationship with the distributor. Every time the client placed their regular 3,000-unit order, we purchased an extra 1,500 sensors on their behalf—essentially buying 4,500 pieces in one go. The client paid for the extra 1,500 sensors as a buffer stock. That buffer covered all subsequent 1,000-unit top-up orders. No more emergency sourcing, no more distributor markups.
The client accepted the approach. The cost of holding 1,500 units of buffer stock was far lower than paying triple the price every time they needed a small top-up.
The lesson:
In small batch PCBA manufacturing, procurement problems are rarely about availability. They’re about paying the right price for the right quantity. A stable sourcing channel and smart inventory planning are more reliable than last-minute spot buying.
A client’s test firmware was written during the development phase. It was comprehensive—full coverage, every feature tested, every edge case considered. Five minutes per board.
During development, that made perfect sense. The goal was to confirm that the design worked. But in production, the goal shifts: confirm that every board matches the design. The logic is different. The speed requirement is different. The test strategy is different.
Five minutes per board. For thousands of boards, that’s hundreds of hours of test time. The production schedule couldn’t absorb it.
The client hadn’t realized that development firmware and production firmware need to be separate. In the development phase, their focus was on functionality validation—not on production test flow. This isn’t a mistake. It’s simply a different phase with different priorities. We received their production-specific test firmware, confirmed the version match, and adjusted the test schedule accordingly. No delays.
What we did:
We flagged the issue during our pre-production review. The client had time to adjust. They developed a streamlined test firmware specifically for production—focused on critical nodes, optimized for speed, designed to run on production test fixtures. The new test took 30 seconds per board.
The lesson:
Many clients don’t realize test firmware needs to be different for production until they’re already in it. The reason we get involved early in small batch PCBA manufacturing is to catch this kind of issue before it becomes a schedule problem. Some problems aren’t technical—they’re timing problems. Caught early, they’re manageable. Caught late, they’re expensive.
These three cases have one thing in common: none of them were “hard” problems.
The potting fixture wasn’t high-tech. The procurement strategy wasn’t rocket science. The firmware adjustment wasn’t complex engineering.
What made the difference was someone looking at every step of the process.
For context, industry standards like those defined by IPC provide the baseline for quality and reliability in electronics assembly. But standards alone don’t catch the kind of issues described above—they require a manufacturing partner who knows when to look beyond the checklist.
Our production line is the same production line used for high-volume runs. The equipment isn’t special. What’s different is the level of detail our project managers bring to each order—assembly, procurement, firmware versioning, production scheduling. They don’t just relay client requirements to the factory floor. They translate “what needs to be done” into language the production line can act on.
We’re not looking for one-off orders. We’re looking for long-term relationships. And long-term relationships start with a client trusting us on the first order. That trust doesn’t appear from nowhere. It comes from us having considered every detail—so the client is willing to give us their next project, and the one after that.
If you have a project and you’re unsure whether to start with a small batch, send us your BOM and Gerber files. We’ll give you an assessment within 48 hours—what needs to be prepared in advance, what pitfalls to avoid, and how to keep your first run on track.
