Zettelkasten Forum

Paper-Based Zettelkasten Processes for Problem Solving

Here are several ideas on a paper-based, Zettelkasten-inspired process for tackling problems.
(I've tried to keep the post self-contained, my apologies if some things seem a bit too obvious to readers of this forum.)

We suggest the following materials:

  • An index card box in A5 landscape format.
    A paper size of A6 seems too small for diagrams, A4 is certainly OK and for diagrams (or for math with bulky terms) arguably better than A5, but it is about twice as clunky.
    Landscape format fits better with the sheet layout we describe below.

  • Matching paper in A5.
    Paper sheets are thinner than index cards, so the box capacity is much larger.

  • Matching tabs for the box.
    These are often sold together with the box. They are needed for setting up sections and are useful as place markers.

  • Pencil, eraser etc.
    Writing material is a matter of taste, but with diagrams and a constant need for modifications, pencil and eraser are a natural choice.

These simple elements can support our attempts to solve a problem.

  • We can use the sheets for all types of notes - in words and phrases, formulas, diagrams, tables, lists or mind maps.
  • If we divide the sheets by a vertical line, we get a layout with space for side remarks, questions etc. This is an important building block for doing "more metacognition": We can start to think about weird points in the left column using the right column.
  • We can number the sheets (for example in the upper right corner) in a decimal notation, like 12, 12.1, 12.1.1, 13.
  • With this notation, we can reference between sheets: where did an idea come from, where is an idea examined in more detail, where do we find related ideas?
  • We can order the sheets in the index card box according to their number.
    Having the highest number at the front is convenient.

  • Finding sheets works via the numbers. (See the index idea below for an additional access to the material.)

  • Using the index card box tabs, we can set up several sections - one for the main part with notes on problems we try to solve, one for spontaneous ideas (perhaps with sheet numbers I1, I2) or a section for open topics we want to examine later.
  • The index card box may contain an organically growing section on thinking tools for problem solving: general questions, collections of creativity techniques, tricks for special problem domains (e.g. in science).
    This tool collection may contain references to other sections of the index card box.

  • When we take sheets from the box, a tab can help us to put them back later.

How to do the thinking about the problem itself?

  • One approach is "thinking on paper" - a close interplay between thinking and writing:
    We can write down questions, even half-formed ones. We can note that we are confused, and find out in writing why. We can collect ideas, or produce ideas with creativity tools, collect even hazy ones, and make them more precise, or we can play around with diagrams and terms on the paper.

  • Arguably one of the most important processes is reflection:
    We can ask on a new sheet what happened on the previous one. Where are the main obstacles? What can we do now? What variations are possible? What problem solving tools (perhaps from the tool collection) may be useful?

  • The notes are a "work in progress".
    They should not be written in the spirit of a polished solution - such a document can be written later, when the search for a solution was successful.
    A polished version can become part of the index card box.

  • If necessary, we can install an index to find material for a given topic.

How to stay on track?

  • It's easy to get lost while looking for a solution. Here's a practice that may help.
    The basic idea is to use "control tower sheets" - a name inspired by David N Perkins' lecture "40 Years of Teaching Thinking: Revolution, Evolution, and What Next?", available on YouTube.

  • We start with a new control tower sheet and name it "CT 123", if 123 is the next free sheet number.

  • On sheet CT 123, we write down the topic (like "Problem XY") and then a reference like "Questions (123.1)".
  • We start a new sheet 123.1 and collect questions. We choose the most promising question and examine it on a sheet 123.2, and reference 123.2 in the control tower sheet.
  • Each new sheet is referenced in the control tower sheet, with graphical links to its predecessors, forming a kind of mind map.
  • If a topic deserves a control tower of its own, we can start a new control tower sheet CT 124 and so on.
  • With this practice, we produce little "dossiers" of 10-20 sheets for each topic of a control tower sheet.
  • It's convenient to treat the sheets of a dossier as a group and arrange them on the desk while working on them.

Here are several remarks.

  • The process outlined above can be modified in countless ways.
    Feel free to adapt the process to find your own "personal problem solving method".

  • At the core of the method is a small number of principles:
    Externalize: Find a lasting representation of your thoughts. Write them down.
    Organize these documents.
    Use reflection and metacognition: Find out where the obstacles are.
    Use tools.
    These principles could be used in isolation, but I find them much more powerful when they form an "integrated problem solving environment" - in analogy to the "integrated development environments" in computer programming, like Eclipse or Visual Studio.

  • In my mind, a paper-based method has two major advantages over a computer-based one:
    It causes much much less distractions.
    It is much easier to combine text, diagrams and formulas while concentrating on content and not on technology.

  • However, if a computer seems to be the best medium, something like wiki software may be a good choice.

  • Searching for a solution to a problem often leads to tree-like structures - we start with a collection of seminal ideas, we pursue them to a certain point, from there, we can follow several different paths etc.
    The control tower sheets mirror these structures.


  • Thanks for sharing your current approach here, Thomas!

    I like the term "meta-cognition"; separating these transient notes into the subject matter to the left and reflections/comments to the right makes total sense to work on problems.

    Could you add more details about what happens when a problem seems to be solved (if that ever really is the case)? You stressed the point that the notes are not polished, but work-in-progress. Do polished results emerge as new notes inside this box, or does a polished result live outside it?

    Author at Zettelkasten.de • https://christiantietze.de/

  • This is excellent! I am struggling with how to let the zk spur more insights based on “random connections” and weak links in addition to the obvious sources based on the tags and this gives me a great starting point. I think i have to actually do it by hand to develop a more visceral understanding and muscle memory but this gives me a great starting point, thanks!

  • Interesting! Glad to see another approach to a paper Zettelkasten. Mine is half paper and half digital. If you have not seen it, you can see it here: https://forum.zettelkasten.de/discussion/355/the-analog-digital-zettelkasten

    It has a different aim than yours, but still, we are essentially aiming at the same: contribution.

  • Thanks for the responses! Just been away from the keyboard; -)


    To the present day, there are no polished results of mine inside the zettelkasten, or, alas, outside it. Still, I spend a lot of time with my ZK in an effort to strengthen it as a "PCA = personal cognitive assistant". The focus here has shifted from ZK organisation itself to the question of thinking tools that work for me.


    For what it's worth: I like to view a central "module" of my zettelkasten as an "idea generation machine" that can be fed with existing ideas (from inside or outside the zettelkasten) and then churns out new ideas.

    Here are some methods for idea generation that I find helpful.

    Stimuli, random or systematic:

    • How can I transfer a concept from another domain to my topic?
    • How can I force a general concept to fit into my topic?
    • Fairly general concepts like "segmentation" or "state transitions" seem to work better for me. (In part, the use of general principles is inspired by the TRIZ method.)

    Modification methods:

    • This is in the spirit of the SCAMPER tool - how can I Substitute, Combine, Adapt, Maximize/Minimize, Put to other uses, Eliminate or Rearrange elements in my topic?
    • In particular, the construction of opposites and complete reversals sometimes leads to new ideas.

    Perspective methods:

    • What questions might an economist or a biologist ask, with their frameworks of markets and prices or evolution and ecology.

    Morphological analysis:

    • What are basic categories of your topic? (E.g. size, substance, basic mechanism,…) What else could be instances for this category?

    Thinking in graphs:

    • How do central features of your topic change with time? What are the crucial points? Are there changes from one state to another? What causes them?

    Of course I have to ask how relevant these ideas are and what else I may have done in the time producing them. To prevent the idea generation from being too unfocused, I often start with the question "What is missing here?" and then use methods like the above to find hints to answers.


    I found the images very instructive, thank you so much!

    The next part is a bit tentative: At this time, I'm very much focused on the question how a zettelkasten could help to overcome "obstacles" in finding a solution to a problem. Does this obstacle concept (unelaborated as it is) make sense in your line of work? Are there practices that have shown to be helpful?

  • @thomasteepe thanks for the detailed methods . They are very actionable. I now will make a zettel or two just out of your suggestions so as not to lose the value. Need to figure out how to cite the source ;). Thanks, this gives me a lot to work with. Please keep more thoughts and suggestions coming.

  • Here's an update on my paper zettelkasten methods since the above posts from January and November 2018. Some bullet points deal with thinking tools rather than immediate zettelkasten practices - I hope their inclusion is OK.

    • The goal of "making progress through a reliable process" has played a central role.
      One approach is to start with a diagram / text summary of what I know about a certain aspect of a topic.
      Then I look for points of unclarity, confusion, conflict, missing parts, obstacles etc., and I ask for possible next steps, improvements, alternative models etc. From these thoughts I work out a new diagram / text unit.
      This process leads to a series or sometimes a tree of diagram / text units which I see as normally containing relevant new insights.
      I have a collection of thinking tools like the ones mentioned prominently on my desk.

    • Diagrams and formulas play a much bigger role than in the past.

    • I find something like "proto-math" surprisingly useful. I try to express something in a math form, like
      W = p * T,
      W is the output (in the form of new insights, ideas, concepts) from a zettelkasten work session,
      p is the "power" of the practices and tools I use (at this stage, I do not yet need a precise understanding what that means) and
      T is the duration of the zettelkasten session.
      This is, of course, painfully simplistic and wrong - but in trying to make it less wrong and more precise, I sometimes hit on new ideas - for example the question, how I should allocate the total time of zettelkasten work between "epistemic" work on the topic itself and "heuristic" work on improving tools and practices, or about the co-evolution of a person and their "cognitive assistance systems".

    • I use self-made tabs from index cards and blank address stickers. (I find prefab register cards with tabs fairly expensive.)
      These tabs contain the number and the main subjects covered in a dossier, which works well as an index.
      The index card itself sometimes has a short list of key words for the dossier.

    • The control tower sheets mentioned above play a much more important role in guiding my work on one dossier and in encouraging me to work on yet another approach to a topic.

    • I use a small document holder on my desk for sheets that are work in progress, for the current sheet for spontaneous non-topic ideas I want to record, or for additional tool sheets.
  • Several people on this forum use paper mind maps in their idea processing - see this example from @Phil.
    Over the last few weeks I've done some experiments with arrays of small paper-based mind maps combined with thinking tools.
    It's in an early stage, and feedback is very welcome.

    Here comes a short description and some obersevations.
    (As in my earlier ideas along these lines, these methods of "thinking on paper" are primarily intended for creating thoughts, not necessarily for storing them or for communicating them to others.)

    Here's the description.

    • I use a blank A4 sheet in landscape format.
    • I divide the sheet in 4 x 4 cells of equal size.
    • An important technical remark: I use a mechanical pencil (0.5mm) and tiny handwriting, so I fit about 10 lines in one cell. Alternatively, I could use 3 x 3 or another number of cells, or use A3 paper to get a convenient "cell capacity" in terms of words per cell.
    • For the layout of single cells, I can choose between small mind maps or ordinary text or diagrams.
      Mind maps are my default choice, but diagrams appear time and again.

    • I can now fill the cells row by row and colum by column.

    • To start a new cell, I often use small "thinking tools". A present favourite is "F/P", which stands for "focus and progress" - first I look for an aspect of my topic I want to examine closer, and then I try to make progress on this aspect - usually by asking questions, or by collecting ideas, or by applying creativity tools.
    • At my desk, I have a "tool collection" in the form of about two dozens of sticky notes. They contain basic thinking operators like F/P, collections of idea generation stimuli (like the TRIZ principles, or verbs of modification), basic diagram types (like timelines, pie diagrams or decision trees), or basic "lenses" with which I can look at my topic ("look at points of transitions, look at points with special properties, look at quantiles..."). With many of these tools, I try to transfer a useful and sometimes even sophisticated general concept to my topic.

    Here are some observations I've made during my experiments.

    • First, the crucial point: The capacity of one cell is very limited. In my experience, this leads without extra efforts to a much higher frequency of refocusing, of asking questions and of doing useful forms of metacognition, in comparison with my ordinary mind mapping habits. For me, this is the main strength of the method.
    • The threshold for writing down first thoughts is very low. It feels like I do not have to fill an entire sheet, but just one cell. And if I mess up with one cell, I can restart better in the next.
    • Finding a convenient cell capacity may require some experiments. However, having a fixed cell size rather than a variable one seems to work better for me at the moment.
    • If I have an idea overflow for one cell, I simply start the next on the same topic or on the particular branch I want to expand. This is of course a compromise, but in my present delight with the method I see it as outweighed by its advantages.
    • I find it convenient to use flexible default tools like the above F/P that lead to insights in a broad range of situations.
    • The method seems to work especially well with idea generation using stimuli.
    • It's easy to assemble a personalized "thinking on paper" method - by combining sheet organisation (in a ZK) and sheet layout and cell layout and tool collections with their architecture and their elements.
Sign In or Register to comment.