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

# Core Concepts

> How projects, surveys, sources, streams, and respondent outcomes are organized in Research Desk.

Before creating your first project, it helps to understand how Research Desk structures work.

## Object hierarchy

| Object                 | What it represents                                                                                                                                                                           |
| ---------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Project**            | The top-level container for a research initiative. One project can include multiple surveys (for example, different countries, audience segments, or waves).                                 |
| **Survey**             | A specific fieldwork configuration within a project. Each survey links to one survey URL with one country and language target.                                                               |
| **Source**             | How respondents reach a survey — manual supplier traffic, automated marketplace allocations, or internal test traffic.                                                                       |
| **Stream**             | A targeting configuration within a source. Streams carry qualifications and quotas that determine which respondents can enter.                                                               |
| **Respondent session** | An individual respondent's interaction with a survey, tracked by **RDUD** (Research Desk's respondent identifier), including their outcome (complete, terminate, over-quota, security flag). |

Think of it this way: a **project** is a folder, **surveys** are the studies inside that folder, **sources** are the traffic channels feeding each survey, and **streams** define who can enter through each channel.

## Project lifecycle

* Scope each project to a single research initiative with a defined end date.
* Avoid using one project for all surveys indefinitely. A typical project spans **1–2 months** of fieldwork.
* Assign a **project manager** who is responsible for monitoring fieldwork and reconciliation.

## Survey configuration

Each survey defines:

* Target country and language
* Sample size and quota structure
* Length of interview and incidence rate
* Device compatibility (desktop, mobile, tablet)
* Survey entry link to your programming platform
* Status: **Draft**, **Live**, or **Complete**

<Warning>
  Quotas must be defined when creating a survey. Avoid contradictory qualification combinations — for example, requiring a respondent to be unemployed while also specifying a job title — because those combinations can block all traffic.
</Warning>

## Respondent outcomes

When a respondent finishes (or is screened out of) a survey, Research Desk records an outcome. Common categories include:

| Category               | Examples                                                                         |
| ---------------------- | -------------------------------------------------------------------------------- |
| **Complete**           | Respondent successfully finished the survey                                      |
| **Terminate**          | Respondent screened out for standard reasons                                     |
| **Over-quota**         | Respondent matched a quota that was already full                                 |
| **Security / quality** | Respondent flagged by Research Defender (speeder, bot detection, bypass attempt) |

These outcomes drive field monitoring, reconciliation, and billing.

## Calculation types

Research Desk offers three source-level calculation types: **Completes**, **Survey Starts**, and **Hybrid**. The type you choose affects how quotas fill and whether over-fill or under-fill is more likely. See [Source Quota Calculation Types](/research-desk/guides/10-source-quota-calculation-types).

## Next steps

<CardGroup cols={2}>
  <Card title="Creating Projects" icon="folder-open" href="/research-desk/guides/01-creating-projects">
    Set up your first project.
  </Card>

  <Card title="Creating Surveys" icon="clipboard-list" href="/research-desk/guides/02-creating-surveys">
    Configure survey targeting and quotas.
  </Card>
</CardGroup>