Zovo Decision Tree

Decision trees with expected values.

Zovo Decision Tree turns a decision, its options, the chance events after them and the payoffs at the end into a decision tree drawing with the expected value of each node. Write the tree as an indented outline or a JSON list. Probabilities can be decimals, fractions such as 1/3 or percents. Expected values are computed with fractions, not floating point, so 1/3 + 1/3 + 1/3 is 1 and nothing drifts. The probabilities of each chance node must add up to 1; when they do not, the tool names the node and shows the sum. Each decision picks the option with the highest expected value (or the lowest, when the payoffs are costs), ties are listed, and branch costs are applied by a stated rule. A second calculation, from the probability and value of each path, must give the same root value. You get a vector PDF (Letter or A4), an SVG and a CSV with the fraction and the rounded value of each node, the chosen options and the risk profile of the chosen strategy.

What it does

Try asking

“Decision tree: launch costs 200, strong demand 60% pays 1000, weak demand 40% pays -300, or do not launch for 0”
“Check that the probabilities in my decision tree add up to 1”
“Draw my decision tree as an A4 PDF with values in EUR”

Limitations

Add it

ChatGPT: once Zovo Decision Tree is approved, add it from the ChatGPT plugin directory.
Claude: go to Customize > Connectors > Add custom connector and enter https://decision-tree.zovo.one/mcp. No sign-in is needed. Once Zovo Decision Tree is listed in the Claude connector directory, you can add it from there too.

Other MCP clients can connect to the same endpoint: https://decision-tree.zovo.one/mcp