> ## Documentation Index
> Fetch the complete documentation index at: https://docs.simulant.tech/llms.txt
> Use this file to discover all available pages before exploring further.

# Datasets

> Add, connect, and manage the data that grounds your simulated population.

A **dataset** is a source of real-world signal that grounds the personas in your workspace. Simulant ships with a public library so you can start immediately, and you can add your own data to sharpen the population around your market.

<Frame caption="The Datasets view — workspace sources and the public library, with records and how each dataset feeds cohorts and tests.">
  <img src="https://mintcdn.com/simulant/j7X70BC9NUZuF_VF/images/app/datasets.png?fit=max&auto=format&n=j7X70BC9NUZuF_VF&q=85&s=cc5a22b6937e27d8bfe3b6498d70c439" alt="Datasets view showing a campaign's voter file, tracking poll, and public sources like the census" width="1428" height="779" data-path="images/app/datasets.png" />
</Frame>

## The public library

Every workspace can draw on Simulant's **public library** — a curated set of official, high-quality datasets covering population, attitudes, and behavior. Browse it under **Datasets → Add dataset → Public library**, then add a dataset to your workspace to make it available to cohorts. See [How grounding works](/data/grounding) for what's included.

## Add your own data

Go to **Datasets → Add dataset** and choose a source:

<Steps>
  <Step title="Upload a file">
    Drop in a file to import — for example, a brand tracker or a survey export. Simulant reads the schema so you can confirm the fields and preview sample rows.
  </Step>

  <Step title="Connect a source">
    Point Simulant at an API endpoint and provide an authorization token to pull records from an external system.
  </Step>

  <Step title="Add from the library">
    Pick a dataset from the public library to add to your workspace.
  </Step>
</Steps>

<Steps>
  <Step title="Describe the dataset">
    Give it a name (for example, `Q2 Brand Tracker`), a description of what it contains and how it should be used, a category, and tags.
  </Step>

  <Step title="Set access">
    Choose who in the workspace can use it.
  </Step>

  <Step title="Add to workspace">
    Confirm, and Simulant imports and prepares the data.
  </Step>
</Steps>

## Schema, versions, and preview

Open any dataset to inspect its **schema** (the fields and their types), browse **sample rows**, and see its details. Datasets are **versioned** — when you re-import or update a source, Simulant keeps the version history so results stay reproducible and you can see exactly what a run was built on.

## Using datasets in cohorts

Datasets don't shape a study on their own — you select which ones a [cohort](/guides/cohorts) is built from under **Source data**. Toggling a source on or off changes the signal behind that cohort's personas.

<Note>
  Your uploaded and connected datasets stay private to your workspace. Public library datasets are shared grounding available to everyone.
</Note>

<Card title="How grounding works" icon="database" href="/data/grounding">
  Understand how datasets become a population.
</Card>