The shoot wraps at six. The selects take a week. Nobody is slow. The picks just come back in every format except the one Capture One can read.
"Love the one where she's looking left." "My picks are 3, 7, 14, and maybe 22." A screenshot with a circle drawn on it. A Slack message about the third one from that series. Every one of those is a decision. None of them is a filename.
So you translate. You open the session, scroll to the second setup, guess which frame is "looking left," and star it. Then you do it again for the next 149 picks. It is the most expensive hour of the job, and nobody pays for it.
Capture One can select a whole list of images in one move. Most photographers never find it. In your session, go to Select > Select by > File name list. Paste the names, tick Ignore file extension so a JPG proof still finds its RAW, and Capture One selects every match. Rate them, tag them, drop them into a Selects album. Seconds, not an hour.
The command is fine. The input is the problem. It only works with an exact list of your original filenames, and clients never send you that.
Proofing galleries help, and many will export a list. But the list carries whatever name the proof had when it went up. If the gallery renamed it, resized it, or cleaned it up for the web, the list points at the proof, not your capture. You fix it by hand, which is the job you were trying to skip.
I built AirSelects to stop doing this translation. The client makes the decision. The filename carries it back.
Either way, your client's selects are in your session. No retyping. No guessing which one was "looking left." No thread to reconcile.
The photography is the part you are good at. The week after it is where the work loses value: decisions stuck in texts, a retoucher waiting on a list, a client wondering why the files are late. Shorten that week and everything downstream moves.
One link out. Selects back in Capture One.