# Overview
RootCause turns structured business data into an explainable causal model that you can use to test decisions before acting. It connects to your data, discovers cause-and-effect relationships, combines them with human judgment, and produces a runnable Digital Twin. You can then simulate interventions, find optimal strategies, explain past outcomes, and forecast what happens next — all backed by causal evidence, not just historical patterns.
### How It Works
RootCause follows a seven-step workflow. Each step builds on the last, producing a specific artifact that feeds the next stage:
1. [**Connect Data**](/start-here/workflow.md#1-connect-data) — Import data from files, databases, or APIs
2. [**Build Ontology**](/start-here/workflow.md#2-build-ontology) — Map your data into a shared semantic layer
3. [**Build 360 Table**](/start-here/workflow.md#3-build-360-table) — Prepare a single, analysis-ready dataset
4. [**Build Causal Graph**](/start-here/workflow.md#4-build-causal-graph) — Discover what drives what
5. [**Build Digital Twin**](/start-here/workflow.md#5-build-digital-twin) — Create a runnable causal model
6. [**Run Simulations**](/start-here/workflow.md#6-run-simulations) — Test decisions before acting
7. [**Produce Reports**](/start-here/workflow.md#7-produce-reports) — Package findings into living documents
For the full walkthrough, see the [Seven-Step Workflow](/start-here/workflow.md). If you want to jump straight in with sample data, try the [Quick Start Tutorial](/start-here/quick-start-tutorial.md).
***
### Key Concepts
Understanding these terms will help you navigate the platform:
**Dataset**
Raw data imported into RootCause from files or connectors. This is your original data before any transformations.
**Data View**
A transformed and prepared version of one or more datasets, ready for analysis. Data Views let you join, filter, and reshape data without modifying the originals. In the seven-step workflow, this is the "360 Table."
**Ontology Concept**
A unified representation of a column that may appear across multiple datasets. For example, "Customer ID" might appear in your sales data, support tickets, and CRM export — ontology concepts link these together so RootCause knows they refer to the same thing.
**Digital Twin**
A causal model of your business or system that can be used for simulation and optimization. Unlike traditional predictive models that just forecast, a Digital Twin understands *why* things happen, so it can predict the effects of actions you've never taken before.
**Simulation**
A what-if analysis run against your Digital Twin. Simulations let you test interventions ("what if we raise prices?"), find optimal strategies ("what's the best marketing mix?"), or explain outcomes ("why did sales drop last quarter?").
***
### Getting Help
**RootCause Assistant** (BETA)
Use the chat interface in the bottom-right corner to ask questions about your data, generate simulations, or get help navigating the platform. See [RootCause Assistant](/more-details/rootcause-assistant.md).
**Support**
Contact your organization's administrator or reach out to the RootCause team.
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://docs.rootcause.ai/start-here/readme.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.