GenerateRandomSearch

Decision Tree Generator

People are reliably bad at estimating how many ways a sequence of decisions can play out. Three decisions with three options each is twenty-seven outcomes, not nine, and the gap between what that feels like and what it is is where planning goes wrong. This builds the complete tree — every path, all the same length — so the shape of the problem is visible before you start pruning it.

What this generator does

Enumerates rather than advises. The decision matrix on this site scores options you can already list; this generates the list when the options come from combining choices. The probability tree next door attaches likelihoods to each path — this one is about structure alone, and it produces a complete tree where every path is the same length and every decision has the same number of branches.

How to use this tool

  1. Type the question at the top.
  2. Give the branch labels, comma-separated.
  3. Set how many levels deep and how many branches per decision.
  4. Build it, then copy the paths and prune the ones that do not apply.

Understanding the controls

The question at the top
The first decision. Everything else hangs off it.
Branch labels
Comma-separated. The same labels are used at every level, shuffled so the tree does not look mechanical. Give fewer than the branch count and the rest are numbered, with a note.
Levels and branches
Together these fix the number of outcomes at branches to the power of levels. The combination is capped at 256, because past that a tree stops being something you can read.

Worked examples

2 branches, 3 levels
8 paths.
3 branches, 3 levels
27 paths — the number people underestimate.
4 branches, 4 levels
256 paths, which is the cap.
Two labels for three branches
The third is numbered, and the page says so.

Common use cases

  • Mapping the outcomes of a sequence of yes/no decisions
  • Planning branching scenarios for training, a game or a workshop
  • Showing a group how quickly options multiply
  • Building a skeleton before writing a real decision procedure
  • Enumerating test paths through a branching flow

How this generator works

The tree is built recursively to the requested depth with every internal node given the full branching factor, which is what makes the leaf count exactly branches^levels and makes the whole structure checkable. The finished tree is verified before display: every path from the root has the same length, no node has a partial set of children, and the leaf count matches the arithmetic.

Randomness and fairness

The only randomness is the order the branch labels appear in at each node, which stops the tree looking like a mechanical repetition. The structure — depth, branching and the number of leaves — is fixed by your settings.

For how randomness is produced across the whole site, see how Generate Random works.

Assumptions this tool makes

  • Every decision has the same number of options, which is what makes the tree complete and countable.

Limitations and good to know

  • It builds complete trees. Real decisions prune early, and this deliberately does not — pruning is your job once you can see the shape.
  • The same branch labels are used at every level; genuinely different questions at each level need editing afterwards.
  • There are no probabilities or payoffs here. For those, use the probability tree.
  • The tree is not kept between visits; copy the paths before pruning them somewhere else.

Common mistakes

Building a deep tree and expecting it to be readable
Four levels of four is 256 paths. Build two or three levels, prune, then extend the branch that matters.
Expecting the tool to recommend a path
It enumerates. Which path is best needs weights or probabilities, which the decision matrix and probability tree provide.

Practical tips

  • Build the tree, then delete the impossible paths — what is left is usually far smaller and much clearer.
  • Copy the paths into a checklist if you are using them as test cases.

Privacy and your data

The question and labels you type stay in your browser. Nothing is uploaded, nothing is stored between visits, and none of it is written into the page address. Analytics records that the tool ran and how many outcomes it produced, never any text.

Frequently asked questions

How is this different from the probability tree?
This one enumerates structure — every path, no likelihoods. The probability tree attaches a chance to each branch and works out how likely each outcome is. Use this to see the shape, that one to see the odds.
Why is it capped at 256 outcomes?
Because a tree with more paths than that is not something anybody reads. If you genuinely need more, the useful move is to prune the early branches first and build out from what survives.
Can each level have different options?
Not inside the tool — the same labels are used throughout, which is what keeps the tree complete and countable. Copy the paths and edit them once you have the structure.