Per-Image Comments Beat Whole-Gallery Photo Feedback

Where Whole-Gallery Feedback Breaks Down

A comment box at the bottom of a gallery page is the easiest thing to build and the worst place to put client feedback, because it strips out the one piece of information that makes a note usable: which photo it's about. "The bride's dress needs to be brighter in a few of these" is a real sentence from a real client, and it describes zero photos on its own. Someone has to go back through the gallery, guess at "a few", and hope they guessed right before the retoucher spends time on the wrong frame.

Run the arithmetic on a single revision round. Say a 200-photo wedding gallery comes back with 25 notes, all left in one text box or one email. Before any actual editing starts, someone has to read each note, decide which frame it's describing, and open that frame to confirm the guess:

• Matching time: roughly 90 seconds per note to re-read the gallery, find the likely frame, and open it — 25 notes × 90 seconds = 37.5 minutes, spent before a single pixel is touched.
• Wrong guesses: even a low miss rate costs a follow-up message and a second look. At one wrong guess in five, that's 5 more round trips on top of the 37.5 minutes.
• Re-confirming later: when the retoucher delivers, someone has to re-read the same paragraph to check which of the 25 notes actually got addressed, because nothing marked them done individually.

None of that is retouching. It's the tax whole-gallery feedback charges for not knowing, structurally, which photo a sentence is about. Laid out against the same 25-note round, pinned comments remove each line item rather than shrinking it:

The 90-second and one-in-five figures are estimates for illustration, not measured averages — the arithmetic is the point, not the specific inputs. Swap in your own numbers for a real gallery and the shape of the comparison holds: every row on the whole-gallery side is work that a pinned comment never creates in the first place.

Per-Image Comments vs. Whole-Gallery Feedback

The difference isn't cosmetic. A whole-gallery comment box asks the client to do translation work — turn what they're looking at into a sentence that names it — and then asks whoever reads the note to translate it back. Pinning the comment to the frame removes both translation steps, because the frame the client was looking at when they typed the note is the note's context.

What a Good Per-Frame Note Actually Needs

Pinning a comment to a photo is necessary but not sufficient. A note that says "fix this" on the right frame is still a note nobody can act on. Three things separate a usable per-image comment from a pinned complaint:

• A location within the frame, not just the frame. "Fix this" on a full-body portrait still leaves the retoucher guessing between a stray hair, a wrinkle in the jacket, and the background. A note anchored to a specific area of the image — not the whole rectangle — removes that guess the same way pinning to the frame removed the gallery-wide guess. Retouching feedback that lands on the right pixel goes into what changes when a note points at a shape instead of an image.
• Which version it's about. A note left on round 1 and read against round 3 can point at something that's already fixed, or worse, something that's already been changed to look different than the note assumes. Notes need to be stamped to the version they were written about, not treated as timeless.
• One instruction, not a paragraph of them. "Brighten this, and also check the crop, and can you look at the background too" is three notes wearing one comment. Splitting them means each can be ticked off independently, instead of a whole comment staying open because one of its three asks hasn't been addressed.

Get those three right and a per-image comment does what a whole-gallery paragraph never can: it stays attached to exactly what it's about, from the moment it's written to the moment it's marked done.

Where the Note Ends Up

Pinning the comment to the frame on the client's side only pays off if it stays pinned on the way to whoever makes the change. Appruva's Photoshop panel is where that shows up for a retoucher working in the application: it knows which frame the open document is — an XMP stamp first, then the filename, then it asks if neither resolves it — and pulls in the same notes the client left, numbered to match the badges they saw in the gallery. Notes arrive live over a WebSocket where the network allows it, and on a short poll where it doesn't, so the numbered list next to the open document stays current without anyone refreshing a browser tab.

Ticking a note off happens in the panel, against the document that's actually open, instead of in a separate tracking sheet that has to be kept in sync by hand. That's the same idea as pinning the comment to the frame in the first place, carried one step further: the note stays attached to the thing it's about all the way through to the person fixing it.

Worth saying plainly: this is a Photoshop panel, not a family of plugins. There's no equivalent for Lightroom or Capture One today — if the retoucher isn't working in Photoshop, the notes still live in the gallery itself, pinned to their frames, readable without any extra software.

Getting Clients to Actually Use Per-Image Comments

The interface change is the easy part. The harder part is a client's habit of typing everything into one email, because that's what they've always done — a habit built over every previous vendor who only offered a comment box or an inbox. Three things move that habit:

1. Make the gallery the only place feedback goes. If email still works as a back channel, it will get used as one, and the whole point of pinning notes to frames disappears the moment a note arrives outside the gallery instead of inside it. That means resisting the urge to also accept feedback by email "just this once" — the exception becomes the default within a round or two.
2. Show, don't explain. A one-line instruction — "click the photo, then click where you want a change" — in the delivery message does more than a paragraph describing the review process. Clients who see a badge on a specific photo understand what it means faster than they'd understand a written explanation of the feature, because the badge is doing the explaining.
3. Reply inside the gallery, not around it. If a client's note is ambiguous, resolving it with a reply pinned to the same frame keeps the whole exchange attached to the photo. Answering by email starts the detachment problem all over again, just one step later in the process, and it teaches the client that email is a valid channel after all.

None of this requires new software on the client's side — they're already opening a link in a browser. The only behavior that has to change is where they type, and that's mostly a matter of not giving them anywhere else to type.

When Whole-Gallery Feedback Is Still Fine

Per-image comments aren't a rule for every gallery. A 6-photo headshot delivery where the only note is "these three are approved, skip the other three" doesn't need per-frame pins — the whole gallery is small enough that a single sentence carries all the context it needs. The failure mode is specifically a function of scale: it shows up once the gallery is big enough, or the notes are granular enough, that a reader can no longer hold "which photo does this describe" in their head while reading the feedback.

A rough line: if a note could be answered with "which one?" the first time someone reads it, that note needed to be pinned to a frame. If it couldn't — because there's only one photo it could possibly mean — a comment box under the gallery was never the problem in the first place. Stopping clients from marking up photos in PowerPoint covers the related habit of feedback leaving the gallery entirely, which per-image comments only solve if the gallery is actually where clients are typing.