Insights / Manufacturing
What a design transfer package needs to contain, a step-by-step transfer process, and a worked example of the kind of gap that only shows up once a design leaves the hands that built it.
Design transfer is the formal handoff from "we have a validated design" to "someone else can build it correctly, repeatably, without the original design team in the room." For a regulated product, it's also a defined step in the design control process, referenced directly in frameworks like 21 CFR 820.30(h) and ISO 13485, which require that design outputs be verified as adequate for production before routine manufacturing begins.
In practice, design transfer covers more than shipping a folder of files. The BOM, schematics, layout, assembly drawings, and test procedures all need to travel, but so does the information that rarely makes it into a formal document: why a tolerance is tighter than it looks like it needs to be, why a specific supplier was chosen for one component over an equivalent, what almost went wrong during bring-up and how it was resolved. A transfer that only moves the paperwork and not that context tends to surface its gaps at the worst possible time, mid-production, not during review.
Design transfer happens more than once in a typical product's life: from R&D to a pilot line, from pilot to volume production, and again whenever a program moves between contract manufacturers or a supplier changes hands. Each of those transitions carries the same risk, and each deserves the same discipline, not just the first one.
| Situation | During prototype builds, the original design team hand-selected a tighter-tolerance variant of a passive component after finding the nominal part caused marginal timing on one board out of a small batch. |
| What happened | The substitution was made verbally and tracked only in the engineer's personal notes, not in the BOM or an ECO, since the nominal part number never technically changed on paper. |
| What the transfer package showed | The original, nominal-tolerance part number, with no note explaining the tighter spec that had actually been used and verified. |
| What happened at the new build site | The receiving contract manufacturer correctly sourced the part number as documented, at standard tolerance, and a small percentage of units failed the same marginal timing issue the original team had already found and quietly worked around. |
| What would have prevented it | A BOM note or ECO capturing the tolerance requirement explicitly, and a design transfer review step where the receiving team asks "why this specific part" for anything that looks like it could have a story behind it. |
This is a common shape for design transfer failures: nothing was hidden on purpose, the information simply lived in a person's memory instead of a document. A transfer process that treats "why did we choose this" as a real question, not just "what did we choose," catches this class of issue before it reaches a customer.
It's a formal, documented requirement for regulated products, but the underlying discipline is valuable for any hardware program moving between teams or manufacturing sites, regulated or not. The cost of a bad transfer shows up in production either way.
It depends heavily on design complexity and how well-documented the design already was going in, but a first article build and gap resolution cycle measured in weeks, not days, is typical for anything beyond a simple board.
Undocumented tribal knowledge, information the sending team knows so well they don't think to write it down. Obsolescence issues and supplier-specific quirks are the most common places this surfaces.
More from Insights