My RPG index card file

In October of 2019, I started putting ideas I had about TTRPGs on index cards. I’d just read about Niklas Luhmann’s Zettelkasten, which was a knowledge-management method that he’d used to write a ton of books and articles, by capturing ideas on index cards, each with a unique ID that allowed easy cross-references and lookup. (Zettelkasten had an odd spike in popularity last summer, with a lot of productivity grind content farms publishing articles about it.)

Even though I don’t have any intention of writing books, I did like the idea of being able to capture a certain line of thought, branch new ideas off of it, set it aside for a while, and then come back later and pick up where I’d left off. Sort of like an external memory for someone who loves the idea of “always be prepping”, but doesn’t have the same data retention and recall abilities of a certain bearded podcast host.

Why paper cards?

Among some people (myself included), there’s a tendency to want to find or build a system for this sort of thing in software. After all, storing and retrieving information is what computers do. And there are plenty of software solutions out there. But… I liked the idea of just writing things on index cards.

  • I remember things more when I write them than when I type them.
  • I won’t be tempted to just copy & paste interesting things without thinking about and interpretting them.
  • I like being able to arrange cards on my desk as I follow links from one to another.
  • Flipping through sorted cards engages the same sort of spatial memory that helps me remember where to find specific passages in rule books.
  • Seeing other cards while I’m looking for a particular one means there’s a chance to make serendipitous connections between previously unrelated thoughts.

Now, there are also downsides. There’s no easy way to make backups, so there’s just the one copy. There’s no full-text search, no instant access to a particular card. I wouldn’t try using my card file at the table during a game, because things would grind to a halt while the players watch me shuffle through a thick stack of index cards. But for capturing ideas for later expansion and review, it’s terrific.

How does it work?

When I have an idea that I want to capture, I pull out a blank index card. In the upper right corner, I write the date and time. In the upper left corner, I write an ID number. These numbers follow a particular pattern to make it easy to find a card based on its ID, but they’re otherwise meaningless.

IDs start with a number. You add a “child” card by adding a letter after that number. Then you can go another level deeper by adding another number, then another letter, and so on. When you add another child under a card, give it the next number or letter in the sequence. For example:

  • The first card I wrote had an ID of 1, and it had some thoughts about the advantages and disadvantages of random character generation.
  • The next card was 1a, about “semi-random” character generation which was immediately continued on 1a1.
  • Much later, I wrote 1a2 about DMs rolling ability scores in OD&D.
  • I started another branch off of 1 about the card-based method I have for generating ability scores. That was 1b.
  • The same day that I wrote 1, I also started trees for
    • 2 (descending AC, moving off into some alternative combat mechanics)
    • 3 (RPG systems with interesting mechanics worth borrowing from)
    • 4 (inspiration from other media - books, movies, music, etc.)

Those are not hard categories, and the parent-child relationship between ID numbers doesn’t necessarily mean there’s a relationship between the topics of cards with similar prefixes. For example, one of the sub-trees underneath 3 pretty quickly turned into ideas for a new RPG system rather than thoughts about existing ones.

The ID system just means you can always insert a card between any two other cards, by incrementing or appending a number or letter, and you can sort the cards by ID so you can find them later. If you run out of letters, or the nesting gets too deep, just start a new root. I’m up to 11 now.

Linking cards

To link one card to another, I just write the target card’s ID and draw a box around it. Usually I’ll add a back-link to the target card at the same time, but not always. I try to make sure each new card that I add has at least one link from an existing card. Sometimes I discover an unexpected connection later on, and update cards to add new links.

Besides box links, I also have “continued” links that are indicated with a down arrow, and mean that the thought on this card is continued on the linked card. Likewise, I also have “context” links that are indicated with an up arrow, and mean that the linked card provides some additional context for this one.

These links are the real structure of the card file. Instead of a rigid hierarchy, there are clusters of tightly linked cards with more tenuous connections between clusters. The categories evolve over time as more cards and links are added.


When a particular card has a large number of children, I add a new “index” card that just contains a list of links to the most important or interesting ones. This index card gets an ID that ends in either a 0 or λ, and gets sorted before other cards. For example, under 3 I have , and under 3a1a I have 3a1a0. At the very top, I have a 0 “master index” that’s continued on 0a and 0b.

These indexes aren’t rigorously updated, and are only somewhat useful except as a reminder of roughly where I might find certain types of cards.

Is it worth it?

Maybe? The biggest benefit that I’ve found so far is that I can develop a line of thinking over a day or two, writing several cards on one topic, and then leave it alone for several months before picking back up from where I left off.

Most of the cards I’ve written have been about mechanics, but a few are more about setting or encounters. I think developing those might have a bigger impact on games that I run.


This type of thought organization has been on my “to do” list for years. I purchased pullout index card boxes, dividers, and many cards (of varying colors). I keep starting to write thoughts on the cards and soon after find myself writing most of my thoughts in my notebooks while prep’ing adventures. It ends up being good “stream of conscious” stuff that I tell myself I will write on the cards. @jim I thought I was the only lunatic that wanted to use this method instead of entering info on spreadsheets. About how many cards do you have?

For me, I think it is similar to when someone decides to build a shed and uses a hammer and loose nails even though they have a nail gun. The “sweat equity” is a good feeling to me. I understand it is not the most efficient way to build it, but efficiency is not the goal. I feel more invested in the project even though I know that it doesn’t really make total sense.

I think this post has inspired me to grab a stack of cards and have at them. I’ll let you know how I make out. Thanks.

Feelin’ Good about Card Catalogs

1 Like

Somewhere between 100 and 150, I think. What makes this particular system sustainable for me is that there’s almost no up-front planning about categories or structure for different types of cards. Just the stuff on card 5 (pictured above), and that’s enough to make sure everything is ordered enough that I can find it again, given a couple of minutes to look for it.

The branching ID scheme and fixed card size means that when I do start writing in a stream of consciousness, it’s broken up into units that I can later insert new ideas into, or branch off into another stream at any point. And the linking notation means I can connect it all together in whatever way makes sense to me at the time.

Yeah, there’s a bit of that involved. But there’s not that much work to it either. Just pull out a blank card and start writing. It might take a minute or two to find the “right” ID number so that it’s close to related cards, but that’s not at all critical. Compared with how much tinkering and fine tuning I know I’d do with a software version (not to mention going back and revising existing notes instead of writing new ones), I might come out ahead here.

Do it.

1 Like

Holy WOW! Very impressive!

1 Like

@jim and so it begins…


Love it! Looks like you’re off to a good start.