I have written down my process a few times over the years; it's interesting to have as a record as it has changed. Perhaps of interest to others, too; but as the specifics are probably TL;DR, I do think what is interesting is at the end, with thinking about the philosophy of how you find what's important to you.
Current workflow, in three parts: geotagging, RAW rendering and culling, and archiving
First, geotagging. Where I took a picture I know it is not at all important to other people, but it is important to me. Many of my projects are geography-based, and when I browse old photos I really like to know where I was. Back in the “good old days” of Aperture I would apply geographic data, like all of my other metadata, after my cull. Now, in the bad, post-Aperture days, I have to geotag everything before I even cull, because I need that data written directly into the file and not some program (like Aperture’s) database. This means the geotag info is searchable even via the Finder; and even though I end up geotagging duds the process below is mostly automatic so it doesn’t really matter.
1. From the various cameras, import the RAW or HEIC files into one folder on my desktop, called DNG
2. Upload from my iPhone the day’s geotag track. I use “GPX Tracker”, but there are lots of alternatives. Within GPX Tracker I save the track and upload it to a dropbox folder called GPX
3. Open up “Houdah Geo” on my Mac. Add all of RAW or HEIC files into its browser, add the GPX track. I add the photos from each camera in separate sessions, as minute differences in each camera’s clocks can affect the geotagging
4. I can fix time zone errors or clock differences now before adding the geotags. Adjust the tagged track by fixing any obviously wrong outliers and odd altitude recordings (Houdah Geo lets you substitute altitude from a database)
5. Houdah Geo can then write all of the geotag data to each file
6. Save the GPX track into a finder folder organized by month
Now I render / cull. I use Capture One’s sessions. Each session is made up of a month or a portion of a month, so named; for example, I’ll start a month as 2025 03 01 and keep dumping pictures into that session. If there’s a break in my routine — say a trip — then I start a new session with a name: e.g., 2025 03 16 Paris. Sessions are also small enough so that the entirety of each session can live on my laptop.
Within the structure of each session I import my geotagged files. Within Capture One’s import browser obvious duds can be flagged not to import. The RAW or HEIC files get renamed as they’re imported, to have the photo’s capture date at the start, then the out-of-the-camera file name, so e.g., “_R5_6859.CR3” becomes “2025-03-23 ~ _R5_6859.CR3”. After importing Capture One gives me an initial render, and I go through and “fix” them with simple edits for straightening, exposure and even simple cropping. Only now can I see what I have, and rate my photos. I color rejects with a “red” tag, and at the end of my sort sweep them into Capture One’s session trash folder.
Rating my photos lets me know what I have. When I look back later, I can see which photos made the grade for further work or inclusion in a project. I don’t toss lower-rated pictures because every now and then you need a lesser picture for its content: say, including a person who is not often in photos, or some other specific subject
Archiving. Capture One also allows for the use of catalogs, which is important so that you can see photos together over a span of time, not just restricted to a session. In my case, I have a catalog going of all my personal pictures in each three year period. So, at the end of the month, I do a little dance.
1. In Finder, I copy the “Capture” folder of my current session to its permanent home on a big external RAID. This means the “trash” folder, and all of its rejects, are not coming along
2. The external RAID is organized with the same naming system — year month day [event] — so that it can be understood even within the Finder.
3. I rename the imported Capture folder with the session’s name. Within the capture folder now is a small folder containing all of my edits and adjustments that I made in the Capture One session
4. I open my ongoing catalog, and import the session into the catalog. Importantly I tick the “use existing adjustments” box, and I don’t import them “into” the catalog, but leave them as referenced; so the catalog is just the database of my picture’s metadata and jpeg renders. Referenced is important; if I wanted to switch to Lightroom or some other future program all of my RAWs are already organized in a way to do it, just leaving the database to be translated. All of my session work now is consolidated into a larger, referenced catalog: the catalog lives on my laptop, the RAW files on my external RAID
5. I archive the old session to a different external RAID, and when my laptop gets too full I cull the old sessions from it.
6. At the end of the month I make a Carbon Copy Clone of my whole media catalog (my RAID) that I take to the office where I have another RAID, for off-site backup.
Within my catalog, I can do my further photo projects. From it, I will export “finished” photographs as jpgs or tiffs and then fold those into a book or whatever collection of photographs that live in a different place only my external RAID. Yes, there’s some duplication, but I do want a specific, tiny subset of my work to just exist. After all, when I’m gone, no one is going to bother with these giant databases; they’ll just want a few finished, treasured, readable with whatever photos.
***
Now I will say I have experimented with Photo Mechanic because I really like its ability to add keywords and captions. Back when we were using Aperture, I was religious about this, and I could make albums with my photos sliced and diced every which way; it was really helpful for further photo projects. But my preferred way of dealing with metadata is to apply it /after/ I have looked through and rated my pictures, and I can’t really rate them without seeing how Capture One (or whatever RAW converter) renders the image and I perform basic edits to it — straighten, crop, color correct. Capture One, irritatingly, does not “touch” the files in any way, so that even if I correct time and date it exists only in Capture One’s database, not written to the original file. Conversely, as I understand it, Photo Mechanic will just read the jpg render of each RAW file that is embedded as a part of each RAW file; it cannot be set up to e.g., read Capture One’s jpeg renders.
The difference is philosophical. Of course if one shoots journalistically, quickly, and with a specific purpose of providing deliverables you are going to know from the embedded jpeg whether you “got it” or not. And plenty of times I know I’ve got it too and can see that in the plain-jane, rendered-by-the-camera jpeg.
But a larger portion of what I shoot I only “discover” after I’ve made the picture, and in my case, after I’ve looked at it with a proper horizon, exposure corrected, or even perspective corrected. There can be startling finds that I would not have seen in, e.g., Photo Mechanic’s browser. So, specifically: in my last week of photo of the day shooting, about half of what I selected I knew would be the picture (Photo Mechanic would be great here) but the other half I only discovered after I’d see what I had (Photo Mechanic’s renders would have led me to miss them.) Winogrand said, “I photograph to see what the world looks like photographed.” I photograph to see what the world looks like after it’s rendered digitally with my tools.
