How to think visually with a mind map

Put the problem on the page

Visual thinking starts when a problem stops living in your head. A problem in your head has no edges and no size; on a page it has a root, a set of branches and a shape you can look at from a distance. Writing the root label alone often settles how the problem is framed.

In Canvasola the mind map is a single tree of nodes, and every node carries a label, a color, children and a position. That is the structure visual thinking needs: one thing per node, no formatting to fiddle with, and a position that carries meaning.

Start with the question, not the answer. Write the root as the thing you are trying to understand, and resist filling in branches until the root is precise. The introduction to mind maps sets out the basic idea, and a template gives you a first shape to rename.

• One problem per map, stated as a question in the root
• One idea per node, no formatting to fiddle with
• Position carries meaning, so place things deliberately
• The map is one object you can zoom, fit and export

Separate facts, guesses and questions

The fastest way to make a problem legible is to sort what you have into three kinds: what you know, what you are assuming, and what you still need to ask. Put each kind on its own first-level branch. Once separated, the assumptions stop looking like facts.

This is where a map earns its place over a list. In a list, a confident guess sits next to a fact and reads the same. On a map the branches separate them, and the branch you keep returning to during review is usually the one with the most guesses in it.

Name each branch in a way that tells you which pile it is. "Known from the data", "Guesses" and "Questions for the client" beat "Facts", "Thoughts" and "Other", because they are testable names. Rename a node with Enter when a pile turns out to be mixed.

1. Name the three piles. Create a branch for what you know, a branch for what you assume, and a branch for what you need to ask.
2. Sort what you already have. Move every idea you have into one of the three branches. Leave a branch empty if you truly have nothing there.
3. Watch which pile grows. The branch that keeps filling with new ideas is usually the one your thinking is avoiding. Read it first.

Let position show sequence

Position in a mind map is free, and that is what makes it useful for sequence. Nodes have no fixed order, so you can arrange a branch to read as a timeline, or to place the finished part of a process on one side and the open part on the other.

Be deliberate about it. A tree layout puts the root at the left and grows right, which suits a process with a beginning; a radial layout spreads branches around the center, which suits a problem with no beginning. L switches between them, and A clears manual positions and re-runs the layout.

Position also carries priority. Put the branch you are working on nearest the root and the branch you are avoiding further out, and the map starts to sort your attention for you. The layout comparison covers which mode fits which kind of problem.

• Left to right for a process with a beginning
• Radial for a problem with no natural starting point
• L switches layout, A clears manual positions
• Put the branch you are avoiding furthest from the root

Use color to show confidence

The 8-color Coggle-style palette is enough to carry one scale: how sure you are. Put a known fact in one color, a well-supported inference in a second and a hunch in a third, and the map becomes a confidence map rather than a decorated one.

Press C to cycle the color of the selected node, and the first child inherits the parent's color while siblings cycle through the palette. Inheritance is a problem when color means confidence, because children arrive with the parent's confidence. Fix it on the nodes where it matters rather than fighting the default.

Write the scale in the root note so it survives a break. A legend of three words, one per color, is enough. The color guide covers the palette, the recent-colors row and how to keep color from taking over the map.

• One color per level of certainty, three is usually enough
• C cycles the color of the selected node
• Children inherit the parent color, so fix them deliberately
• Keep the three-word legend in the root note

Note: Color is a code you set up, not a property the map has. If nobody could read the map correctly without seeing it in color, the labels are not doing their job.

Read the shape of the map back

Once the problem is on the page, stop adding to it and look at it. Fit the whole map to the screen with the fit-content control, which frames all visible elements plus the mind map, and step back far enough to see the shape. A lopsided map is telling you something about the problem.

Three shapes are worth naming. A map with one huge branch and several tiny ones is a problem you are avoiding. A map with a long chain of single children is a problem that is really a sequence, and a list would serve you better. One node with fifteen children means an unnarrowed question.

Use the minimap as well, toggled with M, when the map is bigger than the screen. Click it to jump the camera and compare the density of two regions. The folding guide explains how to reduce a crowded map to a shape you can read.

• One huge branch and tiny ones means avoidance
• A long chain of single children means you need a list
• Fifteen children on one node means an unnarrowed question
• Fit the content to view, then step back and look

Turn the map into one next decision

A map is not finished when the branches are full. It is finished when it produces a decision. Read the map back, name the thing it says the real problem is, and write that as a single next action in the root note: the call to make, the thing to check, the person to ask.

Decisions that come out of a map are usually about where to look next rather than about what to do. A map with a wide gap in one branch is telling you to fill that gap, and the action is "find the source for the second mechanism" rather than "finish the project".

Once the decision is made, prune. Delete the branches that turned out to be wrong or irrelevant, because a map that carries dead branches will mislead the next reading. Undo history keeps 20 versions, so a branch you regret deleting can be recovered.

• One next action, written in the root note
• A gap in a branch means go and look, not go and finish
• Delete the branches that turned out to be wrong
• Undo history keeps 20 versions, so pruning is reversible

A checklist you can reuse

• The root states the problem as a question, not an answer
• Facts, assumptions and open questions sit on separate branches
• Branch names are testable, not vague categories
• Position is used deliberately, and A re-runs the layout after big changes
• Color means one thing, and the legend is in the root note
• The map has been fit to the screen and read back for shape
• One next decision is written in the root note

Common mistakes to avoid

• Starting from the answer. Branches that defend a conclusion you already have hide the evidence against it. Write the question first and let the branches argue.
• One pile called Other. A catch-all branch collects the things you have not thought about yet. Name each pile precisely or it becomes a dumping ground.
• Color without a code. Coloring a node because it feels important means nothing to a reader. Decide what a color means before you use it.
• Never stepping back. You cannot see a lopsided map from inside it. Fit the content to the view and read the shape before deciding what to do next.
• Ending with five actions. A map that hands back five next actions has not decided anything. Pick the one the map says matters and write it down.

Frequently asked questions

Is drawing a picture not just writing a list?

A map is not a picture, it is a structure. The branches separate what you know from what you assume, and the position of a branch shows where it sits in the argument. A list can hold the same words, but it cannot show you that one branch has swallowed the rest of your thinking.

How long should a thinking map take to build?

Long enough to get the problem out of your head, usually fifteen minutes for a single question. The map is a thinking tool rather than a deliverable, so build it quickly, read the shape back, and let it change. A polished map that took two hours has been over-designed.

Can I use a canvas instead for this?

Yes, and the canvas is the right tool when the answer is a diagram rather than a structure. The mind map is better when the work is about relationships and priority, because every node stays in one tree. The guide to combining drawing and mind mapping covers when to use each.

What if the map just shows how confused I am?

That is a result, not a failure. A map with one oversized branch usually names the part you have been avoiding, and a map with a long chain of single children tells you the problem is a sequence, so a list will be easier to work with. Both findings come from reading the shape rather than from adding more branches.

This guide is part of the mind mapping guides collection. For the wider product tour, read the complete Canvasola guide, and keep the keyboard shortcut cheat sheet open while you work.