A tag tells you what kind of image this is. Only a note tells you why you kept it

Six months from now, tag:lighting will still be true of an image, but the specific reason it earned that tag — the exact thing the light was doing, what you were planning to use it for — is gone unless you wrote it down at the time. Tags scale because they're short and reusable across hundreds of images. Notes exist for the opposite reason: the one thing about this particular image that a category can't hold.

This is worth stating plainly because the instinct, especially early on, is to try to make tags do both jobs — to invent an increasingly specific tag for every nuance until the tag list itself becomes unreadable. A tag called rim-light-from-window-camera-left is really a note wearing a tag's clothing: it will never apply to a second image, so it isn't reusable, and a list of five hundred tags like it is worse than useless, because it makes the tags that actually are reusable harder to find in the list. The discipline that keeps a tagging system working at scale is knowing which of the two — category or one-off fact — a given piece of information actually is, and writing it into the right one.

1. Write the note at the moment you'd otherwise trust your memory

This is the single habit that matters most on this page, more than any particular style of note or any rule about length. Everything else here is about making the habit sustainable once you have it; this step is about actually having it, and it's the one that's easiest to skip when you're in the middle of something else and just want the image saved.

The best time to write a note is right after you save the image, while the reason is still obvious to you. "Found this searching for something else, kept it for the rim light on the left side" takes ten seconds to write and is completely unrecoverable ten seconds after you've moved on to the next image and forgotten you had that thought.

2. Answer "why did I keep this" specifically, not generally

If you find yourself typing the same phrase into several notes in a row, that's usually a sign the information belongs in a tag instead — a note that repeats verbatim across five images isn't carrying image-specific context anymore, it's a category in disguise, and it'll serve you better as one you can actually filter by.

"Nice colours" is a note that will mean nothing in a year, because it's true of half your library. "The teal in the shadow next to the orange practical — this is the palette for the warehouse scene" is a note that still means something because it's specific to this image and this use. If a note could apply to twenty other images in your folder, it isn't doing its job.

3. Use notes for context a tag structure can't express

Some things genuinely don't fit a tag: a source you'd otherwise lose track of, a person's name if the reference is a specific pose, a measurement, a correction to something the image seems to show but doesn't. These are one-off facts, not categories, and forcing them into tags produces a tag list nobody would ever reuse.

has:notes no:tags              written up but not yet categorized
has:notes tag:reference-only    context added, filed properly
no:notes rating:5               your best images, still unexplained

4. Let no:notes rating:5 be a queue, not a guilt trip

That last query is worth running periodically on its own: your five-star images with nothing written about them are the ones most worth a note, because they're the ones you'll reach for again and most regret not remembering why. It's a short list by design — writing a note on every image is not the goal, writing one on the images that matter is.

5. Search by what a note says, not just whether one exists

A word typed into the query line that isn't a recognized key becomes free text, and free text matches against note content along with everything else searchable. If you wrote "warehouse scene" into a handful of notes, typing warehouse into the query line surfaces every image carrying that phrase, whether or not you remembered to also tag them consistently.

6. Keep notes with the image when you move folders around

Because a note lives in the workspace next to the folder rather than being written into the image file itself, it survives a rename or a move the same way a tag or a rating does — the watcher follows the file, and the note stays attached to it. Reorganizing a project's folder structure after a note was written doesn't cost you the note.

6.5 Write the note as if someone else will read it

Even if nobody else ever will, writing a note as though a collaborator or a future client is going to read it produces a better note than writing a private shorthand only you'd understand today. "Rim light, camera left, kept for the warehouse scene" survives being read cold in a year better than "rim l — warehouse," which will mean nothing to you either, once enough time has passed that you no longer remember what the abbreviation stood for. This matters more than it sounds like it should, because the whole value of a note is that it still explains something after the original context is gone — a note that only makes sense while the context is fresh isn't really doing the job a note exists for.

6.75 Don't note what a tag or a rating already says

A note that just restates the obvious — "good lighting" on an image already tagged lighting and rated five stars — is wasted effort and worse, it clutters has:notes with entries that don't actually explain anything a filter couldn't already tell you. Save the words for the part that isn't visible from the tags and the stars: the reasoning, the source, the plan for how you'll use it, the thing that made this specific image worth keeping over the dozen similar ones you didn't.

7. Write notes on canvases too, not just on individual images

A note doesn't have to live on a single file. If you've pulled a set of images onto a .coii-canvas for a project, a short line of text placed on the board itself — why this grouping, what decision it's arguing for — is the note for the collection rather than for any one image in it. Because a canvas is plain JSON in your own folder, that context sits right next to the images it's about rather than being scattered across each file's individual notes, and it's the first thing you or a collaborator sees on opening the board again months later.

8. Don't let a backlog of unwritten notes become a reason to stop

If you've been collecting for a while without writing any notes, resist the urge to go back and retroactively write one for every old image — you won't remember most of the reasons anymore, and forcing a note where there's nothing real to say produces the same noise as an over-eager tag. Start the habit from today forward, apply no:notes rating:5 as a light, ongoing queue for your best untouched images rather than a debt to pay off all at once, and let the older material stay untouched. A note written honestly on a fifth of your library is worth more than a fabricated one on all of it.

9. Use notes to leave a trail for a handoff

If you ever hand a folder of reference off to someone else — a collaborator, a client, a future version of a team that's grown past just you — notes are the difference between a folder of unexplained images and one that actually communicates the thinking behind it. A tag says "this is a lighting reference." A note says why this particular one, over every other lighting reference in the folder, was the one worth pulling. Someone picking up the work cold reads the notes first, not the tags, because the tags tell them what's there and the notes tell them what to do with it.

10. Revisit old notes when you rediscover an image, not on a schedule

There's no need to audit your notes periodically the way there might be for ratings. A note written honestly at the time it was written stays true — it's a record of what you thought then, not a claim about what's true now. If you rediscover an old image and the note no longer makes sense, or you have more to add, update it in passing. But going looking for notes to fix is generally wasted effort; the value was mostly captured the moment they were written, and re-reading them only when you actually need them is enough.

11. Keep notes short — a sentence beats a paragraph almost every time

The temptation, once you get into the habit, is to write more than you need to. A note that runs to a paragraph is harder to skim in a search result and usually contains one useful sentence buried in three that weren't necessary. Aim for the single sentence that captures why this image mattered, and trust that if you need more context than that, the image itself and its tags are doing most of the remaining work. A library of short, specific notes is more useful at a glance than one of occasional long essays, because you'll actually read all of the former and skip past the latter.

What to do once you have thousands of noted images and can't remember writing most of them

At scale, notes stop being something you read individually and start being something you search through. A folder with a thousand notes isn't a diary you reread — it's a layer the query line can match against, which means the value compounds the more consistently you write them, even if you never revisit any single one directly. The discipline that matters isn't writing a note on everything; it's writing one on the handful of images per week you'd genuinely regret forgetting the reason for, and trusting has:notes to hold the rest together as a searchable set rather than a pile you have to scroll.

If notes are mostly going to carry composition and layout thinking for boards you're building, organizing references by project covers the folder side of that same habit. For the case notes solve best — a reference you know you saved but can't recall where — finding a reference you saved two years ago goes further into search strategy generally. And if you're coming from a board tool where annotation is a drawn note on the canvas itself rather than metadata on the file, the comparison with PureRef covers that difference directly, since PureRef's notes live on the board rather than on the image.

Concept artists collecting reference under deadline tend to write the most useful notes of anyone, mostly because they don't have time to write bad ones — see reference for concept artists for how that workflow holds together end to end. The same habit shows up naturally when collecting film stills as references, where a note is often the only place to record which scene a frame actually came from.

Try it against a folder you already have opinions about — the trial runs free for thirty days with notes, tags and search all included, which is enough time to find out whether a habit of writing ten words per image actually pays off for you.

A month is roughly the right window to test a note-writing habit honestly, because the real question isn't whether you can write a good note once — it's whether you'll still be doing it on day twenty-five, after the novelty has worn off and it's competing with everything else you actually need to get done. If the habit survives a real month of real work, it'll survive indefinitely. If it doesn't, that's worth knowing before you've built a library of a thousand unexplained five-star images and no time left to fix it retroactively.