Home › Blog › The resubmittal with nothing marked
Plan Review

The resubmittal with nothing marked

A clean PDF makes the reviewer read every sheet again.

InstaPreView · Published November 12, 2026 · Updated November 12, 2026
Line drawing of a construction drawing sheet with a title block and a revision cloud, titled “The resubmittal with nothing marked”. InstaPreView by Byts & Bytes.

You answered every comment in a letter and sent a new PDF with no marks. The next letter says the reviewer cannot tell what changed, and the review clock starts over. That feels bureaucratic. It is also the only way a reviewer can avoid re-reading a hundred sheets to find one door.

The model code does not mention clouds. The 2021 IBC, Section 107.4, says the work is installed per the approved construction documents, and changes that do not comply are resubmitted as an amended set. The section is in Chapter 1. Clouds, revision triangles, and a narrative keyed to comment numbers are how most departments want that amended set identified. The rule lives on the intake checklist, not in Chapter 10. Ask. If the checklist says cloud the changes and add a revision date to the title block, a clean reprint fails the checklist even when the door is now correct.

A resubmittal that clears has three layers:

  • The revised drawing, with the change marked the way that department asked.
  • A response that points at the sheet and the detail, not only at the code. Comment number, sheet, adopted section. The comment-response note is the list of comments. The letter still has to point at a sheet.
  • A set in which the code sheet still matches the plans. Fixing a door and leaving the occupant load stale creates a new comment. See when the code sheet and the plans disagree.

Do not cloud the whole sheet. A cloud around everything is the same as a cloud around nothing. Cloud the door, the note, the dimension that moved. If a comment required no change because the drawing already complied, say so and give the detail number. Do not cloud a detail you did not touch.

The mistake is revising the model in the office and exporting "the latest" without a revision record. The reviewer does not have your model. They have two PDFs. Show them the delta.

Upload the marked set and run a free scan before you send it. Confirm the jurisdiction on the jurisdictions page so the edition in the title block is still the one they enforce.

This is advisory. It is not a determination by the authority having jurisdiction. Confirm the adopted edition, the resubmittal checklist, and local amendments before you change a drawing.