When a review runs for the second time against the same configuration, it does not start from scratch. It knows what happened last time.
That is what makes running reviews repeatedly bearable, and it is the difference between a list of problems and a picture of progress.
What happens on a later run
Issues you fixed are resolved automatically. Your team worked on the model for a week; the next run notices and closes them. Nobody has to tick anything off.
Issues you dismissed do not come back. Dismiss something once, with a reason, and it stays dismissed on every subsequent run. Without this, a weekly cadence would mean re-dismissing the same false positives every Monday.
New issues appear. Including things that became wrong because of what you changed — which is exactly the kind of problem that is hardest to catch by hand.
What this looks like over a milestone
Take one DD milestone reviewed weekly. Run one finds 300 open issues. A week later, run two finds 230 — 70 resolved. A week after that, 160.
What the person running QA/QC sees is not a list. It is a number going down, week over week, that they can take to a project manager.
One review per milestone produces no history. If you run a review once at the end of a phase, there is nothing to compare it against — you get a list of problems and no sense of movement. The cadence is what produces the progress report.
Looking at a specific run
Results default to showing everything currently open across the review. To see a particular point in time, filter by run — useful for answering "what did this look like three weeks ago?" or for checking what a specific week's work closed out.
When you want no history
Configuring a review as a fresh start gives you a result with no memory of previous runs. It will not auto-resolve anything and it will re-create issues you previously dismissed.
This is occasionally what you want — a genuinely clean look at a set, perhaps for a different audience. It is not what you want for ongoing production work.
Frequently Asked Questions
Q: Do I have to dismiss the same false positives every week?
A: No. A dismissal persists across every later run of that review. This is what makes a weekly cadence workable rather than exhausting.
Q: Are issues we fixed closed automatically?
A: Yes. The next run notices the fix and resolves them. Nobody has to tick them off by hand.
Q: Can I see what a review looked like at an earlier point?
A: Yes — filter by run. Results default to everything currently open, but you can look at any individual run in the history.
Q: What if I want a genuinely clean result with no history?
A: Configure the review as a fresh start. Be aware that it will re-create issues you previously dismissed, because it has no memory of them.
Q: We only run one review per milestone. What do we miss?
A: The burn-down. With a single run there is nothing to compare against, so you get a list of findings rather than a picture of progress. The cadence is what turns it into something a project manager can read.
Q: Will a later run find problems we introduced while fixing earlier ones?
A: Yes. New issues are surfaced on every run, including ones that became wrong because of what changed since the last one.