Skip to content
Tutorials Center
My learning

Topics

Lesson 6 of 37 ยท Concepts and Setup

Core Concepts: Organizations, Folders, Dashboards and Panels

Learn Grafana's building blocks: organizations, users, teams, roles, folders, dashboards, panels, queries and data frames, and how they relate to each other.

Grafana has a handful of core concepts. Once you know how they nest, the interface stops feeling like a maze. This lesson defines each term in the order you meet them, from the outside (who owns what) to the inside (how a query result is shaped).

The hierarchy

flowchart TB
  S["Grafana server"] --> O["Organization"]
  O --> U["Users and teams"]
  O --> DS["Data sources"]
  O --> F["Folders"]
  F --> D["Dashboards"]
  D --> P["Panels"]
  P --> Q["Queries"]
  Q -. "use" .-> DS

An organization owns data sources, folders and dashboards. Panels hold queries that use data sources.

Organizations

An organization is an isolated workspace inside one Grafana server. Dashboards, data sources, API keys and folders belong to exactly one organization. A user can belong to several organizations with a different role in each. Most installations use a single organization (the default Main Org.); multiple ones suit hosting for separate customers.

Users, teams and roles

Users sign in with an account. Teams group users so you can grant access once. Each user has an organization role:

Role Can do
Viewer See dashboards and use Explore (in some setups read-only)
Editor Create and edit dashboards and folders
Admin Manage data sources, users, teams and org settings
No basic role Only what fine-grained permissions grant

Separate from this, a Grafana Admin (server admin) manages organizations and server-wide settings. Fine-grained access control (RBAC, Enterprise/Cloud) adds custom roles.

Folders and permissions

Folders organize dashboards and are the main place to assign permissions: grant a team Edit on a folder and every dashboard inside inherits it. Folders can be nested in current versions. Alert rules are also stored in folders.

Dashboards

A dashboard has a title, tags, a UID (a stable identifier used in URLs and provisioning), a time range, optional variables, and a list of panels. Dashboards are versioned: every save is a new version you can compare and restore.

Panels

A panel is one rectangle on a dashboard. It has a title, a visualization type, one or more queries, transformations that reshape the result, and options for display (units, thresholds, colors). Panels can be shared across dashboards as library panels.

Data sources and queries

A data source is a configured connection (URL, credentials, defaults). A query is a request written in that source's language: PromQL for Prometheus, SQL for MySQL, LogQL for Loki. Each query has a refId like A, B, C used to refer to it in transformations and expressions.

Data frames

Whatever the source, Grafana converts results into data frames: tables with named, typed columns (fields). A time series is a frame with a time field and a number field, and the labels become part of the field metadata. Because every source returns frames, any panel can draw any result, and transformations work on all of them. The stand-in below models a frame so you can see what panels receive.

JS Stand-in: a data frame

Press Run (or Ctrl + Enter) to see the console output here.

No output.

Stand-in: role checks

The following sketch encodes the three basic roles and asks whether an action is allowed. It is a teaching model, not Grafana's real permission engine (which is far more detailed).

JS Stand-in: simple role check

Press Run (or Ctrl + Enter) to see the console output here.

No output.

UIDs, IDs and URLs

Each dashboard has a numeric ID (internal, avoid depending on it) and a UID (a short string you can set yourself). Dashboard URLs look like /d/<uid>/<slug>. Use UIDs in provisioning, links and API calls because they stay stable when you export and import.

Annotations mark events (a deploy, an outage) on graphs. Links navigate between dashboards, often passing the current time range and variables. Both are dashboard-level features you will use in the dashboards chapter of the Grafana course.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. What owns data sources, folders and dashboards?

    These resources belong to one organization; users may be members of several.
  2. 2. Which role can manage data sources?

    Org Admins manage data sources, users and settings; Editors work with dashboards.
  3. 3. What is a refId?

    RefIds name queries so transformations and expressions can refer to them.
  4. 4. Why use a dashboard UID rather than its numeric ID in automation?

    UIDs travel with the dashboard JSON; numeric IDs are instance-specific.

Summary

  • Organization > folders > dashboards > panels > queries.
  • Users have roles; teams simplify permissions; folders carry permissions.
  • Data sources are connections; queries use their native languages.
  • Query results become data frames, so panels and transformations are source-independent.
  • Use UIDs for stable references.

Next: Configuring Grafana with grafana.ini