The seven folder system

First, a quick correction: I realized many days too late that I said in my last email that we crossed over another quarterly boundary: July 30. Except...we still haven't hit July 30, and it's not a quarterly boundary anyway, nor is July 31. I meant June 30, of course.

Oh well. At least you know I'm not using AI to write this stuff, eh?

Now, a brief update on my new Q3 2025 Consistency Challenge project: migrating my family photo workflow over to Immich and iCloud Photo Downloader. This week, I kept things simple:

1. Make sure Immich, which I installed to experiment with months ago, is updated to the latest version and still working. (It is now.)
2. Make sure iCloud Photo Downloader, also installed months ago, is updated to the latest version. (It also is now.)
3. Make sure iCloud Photo Downloader is still logged in to my iCloud account, or refresh the login. (It needed to be refreshed--consequences of this behavior still TBD.)
4. Check on my annual Mylio subscription renewal date, in case I decide to revert to the free version of Mylio. (I have until the end of August, so...not very long.)

Next week, I'll try to figure out exactly how to automate synchronization that is smart enough to catch deletions only when we actually want that to happen. I suspect that particular behavior is going to be the hardest part to accomplish.

The Seven Folder System

Today, I want to pass on a data organization method that I somehow never heard about until this week. You can apply it to any digital organization task where you have control over a folder structure hierarchy, including something like label definitions in Gmail. It doesn't fit well everywhere, but it's flexible enough to serve as a good rule (or guideline, anyway) in many situations.

I first came across this method in this blog post from SERgroup, but it came across as bland corporate-speak. So, I hunted down this article from TeamNext that I feel does a better job and presenting the idea in an accessible and intuitive way:

1. A maximum of 7 folders per level should be created.
2. Each of 7 main folders therefore contains a maximum of 7 subfolders, and each subfolder contains a maximum of 7 "sub-subfolders."
3. No more than 3 levels are intended in the 7-folder system, resulting in a maximum of 343 folders (7 x 7 x 7).

Adhering to this rule forces you to think carefully about how to categorize information, particularly avoiding the tendency to create a new folder (or sub-sub-sub-subfolder) for every slightly new topic.

Of course, there are cases where it doesn't make sense to do this. A business project folder might logically have a list of 50 files or folders inside, each of which corresponds to a single client or customer. Breaking this apart into groups such that only 7 folders exist per level would add needless friction rather than simplicity.

This is a similar approach to the PARA method, which specifies that the top level have only four folders (Projects, Areas, Resources, and Archives). You can mix them together if you want: use PARA at the top level, then enforce the Seven Folder System under each of the main four folders in PARA--or at least in the first three, since Archives tend to grow without the same limitations and have to be touched far less often.

The big data archive cleanup project I've been working on through the first half of 2025 could benefit from this approach, even if I don't adhere to the 7-folder limits everywhere. As I've quickly categorized everything into high-level groups, I so far ended up with 17 folders; however, half a dozen of those have to do with various forms of media (photos, videos, music, non-music audio, etc.), so those could easily go into a single "Media" top-level folder if I wanted to. I could squash a few other folders together logically as well, probably getting down to the 7-folder limit at least at the top level.

I'm not planning to do that for sure, but I like the simplicity of the rule: no more than 7 folders in any parent folder, no more than 3 levels deep. It's easy to remember and encourages conscious naming and organization without boxing you in too tightly to one particular set of categories.

For anyone faced with an data organization task that could potentially follow that guideline, how does it sound to you? Refreshingly easy? Too strict? Not helpful enough to answer the hard questions about where you actually need to put stuff? Let me know! Maybe we can improve upon it and make a better guideline.

Until next week, happy data-taming!