# Meerie

- [Meerie Docs](https://docs.meerietechnology.com/index.md): Make better decisions under uncertainty: frame a decision, put honest ranges on it, and see which option wins before you commit.
- [The six steps at a glance](https://docs.meerietechnology.com/journey/overview.md): The journey every decision follows, from framing to revising.
- [How to create a decision](https://docs.meerietechnology.com/frame/create-decision.md): Step 1: set the goal, the options, the assumptions and the background of a decision.
- [How to build an influence diagram](https://docs.meerietechnology.com/structure/influence-diagram.md): Step 2: map how inputs combine into the outcome.
- [Types of nodes in the influence diagram](https://docs.meerietechnology.com/structure/node-types.md): What each box on the diagram means, how to spot it, and when to use it.
- [How to set estimates and notes](https://docs.meerietechnology.com/estimate/set-estimates.md): Step 3: give every uncertain number a range, and record where it came from.
- [Why ranges, not single numbers](https://docs.meerietechnology.com/estimate/why-ranges.md): The future is uncertain. A single number hides the risk, and plans built on averages are usually wrong.
- [Choosing how to describe an estimate](https://docs.meerietechnology.com/estimate/describe-estimates.md): The kinds of answer you can give, and why the kind decides which shapes are allowed.
- [Bounded shapes](https://docs.meerietechnology.com/estimate/bounded-shapes.md): Triangle, rounded peak (PERT) and even spread: distributions for hard limits.
- [Skewed and percentage shapes](https://docs.meerietechnology.com/estimate/skewed-shapes.md): Lognormal, normal, split normal, logit-normal and Brown–Johnson: distributions for a likely range.
- [Correlation](https://docs.meerietechnology.com/estimate/correlation.md): Estimates that move together, and what happens if you ignore it.
- [Monte Carlo simulation](https://docs.meerietechnology.com/simulate/monte-carlo.md): Step 4: how thousands of possible futures are played out, and why runs differ slightly.
- [How to set up win criteria](https://docs.meerietechnology.com/analyze/win-criteria.md): Decide what an option must achieve to count as winning, before you see the results.
- [How to read the results](https://docs.meerietechnology.com/analyze/read-results.md): Step 5: the Decision Analysis Report has two screens. Compare the options, then go into one option in detail.
- [Reading the numbers](https://docs.meerietechnology.com/analyze/reading-numbers.md): Expected value, shortfall, percentiles and risk appetite, in plain language.
- [Sensitivity and value of information](https://docs.meerietechnology.com/revise/sensitivity-voi.md): What drives the answer, and what learning more is worth.
- [How to revise](https://docs.meerietechnology.com/revise/revise.md): Step 6: decide now, research more, add options, and restore earlier versions.
- [How to use Dave, the AI analyst](https://docs.meerietechnology.com/tools/dave.md): Ask Dave to build, research and explain, in plain language.
- [How to use notes](https://docs.meerietechnology.com/tools/notes.md): Notes are the primary store of research. Variables link to them.
- [How to share and manage access](https://docs.meerietechnology.com/tools/sharing.md): The access code, roles, and who can see what.
- [How to connect to MCP](https://docs.meerietechnology.com/tools/mcp.md): Connect Claude or another AI assistant to your Meerie decisions.
- [Ask Claude about a decision](https://docs.meerietechnology.com/mcp/explore-decision.md): List your decisions, read a model, see the diagram and check it before running.
- [Build or change a model with Claude](https://docs.meerietechnology.com/mcp/build-model.md): Edit estimates, options and the diagram. Every accepted change becomes a new version.
- [Run and compare simulations with Claude](https://docs.meerietechnology.com/mcp/run-simulations.md): Run Monte Carlo, stress a scenario, compare runs and find break-even points.
- [Research an input with Claude](https://docs.meerietechnology.com/mcp/research.md): Find evidence for an estimate, then accept, edit or reject the proposal.
- [Write memos and documents with Claude](https://docs.meerietechnology.com/mcp/write-up.md): Turn a saved run into a decision memo or a documented brief.
- [Formulas and model code](https://docs.meerietechnology.com/advanced/formulas.md): Write calculations, use the function library, and see the code the engine runs.
- [Constraints](https://docs.meerietechnology.com/advanced/constraints.md): Rules an option must satisfy, such as budget limits and shares that total 100%.
- [Stress tests and scenarios](https://docs.meerietechnology.com/advanced/stress-tests.md): Rerun the simulation under a stated what-if and compare it with the base run.
- [Decision trees](https://docs.meerietechnology.com/advanced/decision-trees.md): Show choices, chances and results on one page.
- [Research sources](https://docs.meerietechnology.com/advanced/research-sources.md): Web search, internal data, experts and the knowledge base, and the rules research follows.
- [Calibration and the specialist view](https://docs.meerietechnology.com/advanced/calibration-specialist.md): Practise giving honest ranges, and reveal extra analytical detail.
- [Best practice](https://docs.meerietechnology.com/best-practice.md): When a quantitative decision model is worth building, and how to tell a good model from a bad one.
- [Glossary](https://docs.meerietechnology.com/reference/glossary.md): The terms used across the docs.
- [Troubleshooting](https://docs.meerietechnology.com/reference/troubleshooting.md): Common problems and what to do.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.