Rev C is approved — which Rev C?

Document control sends the transmittal letter: Rev C approved for release. Procurement opens the folder. Process opens the folder. They are not looking at the same Rev C.

Mechanical signed off on the vessel datasheets exported Monday night. Electrical reviewed the one-line diagram from Tuesday's export — after the process engineer swapped pump P-101A for P-101B in the native file but had not re-exported. The PDF in the submittal folder still shows the old tag. The approval stamp on that PDF is real. The content is stale.

How approvals get ahead of the files

Nobody is trying to cheat the system. Reviews happen in parallel. Exports happen when someone has a window. Email threads carry the real state: "Electrical is good on the set from the 14th, waiting on updated GA." The transmittal register says Rev C. The register is what the customer sees.

• Different disciplines approve exports taken on different days.
• A late edit in Word or Excel does not invalidate stamps already on last week's PDF.
• PLM or document control records status at the package level, not per file hash.
• Intermediate revisions never get a formal re-review — only the delta someone remembered to mention.

An approval without a pinned file revision is a opinion, not a record.

What auditors actually ask for

When something goes wrong on site — wrong valve trim, wrong relief setpoint — the post-mortem is never "who read the PDF." It is "show me exactly which revision each approver signed, on which date, for which transmittal." Teams that live in Bluebeam sessions and SharePoint folders spend days reconstructing that story from screenshots and forwarded emails.

Good release practice ties every approval to a specific revision of every file in the session — not a letter grade floating above the folder. That is the difference between document control (tracking status) and document-set change management (tracking what changed, who saw it, and what went out the door together).

Before you mark a package approved

• Freeze exports for the review session — one cut date, one set of files.
• List every file in scope; partial packages need partial sign-offs, not a blanket Rev C.
• If anything changes after the first export, restart review on affected files — do not amend in place.
• Store approvals against file version IDs, not transmittal numbers alone.
• When in doubt, say "approved pending re-export of X" in writing. Silence becomes Rev C by default.