Visualization
Azure Monitor offers several ways to visualize telemetry, from quick ad-hoc charts to shared, parameterized reports. Pick by audience and lifespan.
| Tool | Best for | Data | Sharing |
|---|---|---|---|
| Metrics Explorer | Quick, ad-hoc metric charts | Metrics | Pin to a dashboard |
| Workbooks | Rich, parameterized, interactive reports | Metrics + logs (KQL) | Shared, RBAC, exportable as ARM |
| Azure dashboards | At-a-glance operational boards | Pinned tiles | Shared via RBAC |
| Azure Managed Grafana | Grafana experience over Azure (and other) sources | Metrics, logs, Managed Prometheus | Grafana sharing model |
Metrics Explorer
The fastest way to look at a platform or custom metric. Select a resource, a metric (e.g. Percentage CPU), an aggregation (avg, max, p95), and optionally split by a dimension to break one chart into per-instance series. Pin the result to an Azure dashboard, or use Save to workbook for something more permanent. Metrics Explorer is where you validate a signal before building a metric alert on it.
Workbooks
Workbooks are the most capable visualization surface — interactive documents that combine text, parameters, metric charts, and KQL log queries into a single report.
- Parameters — time range, subscription, resource, or free-text pickers that flow into every query on the page.
- Mixed data — a metrics chart and a KQL grid side by side, driven by the same parameters.
- Templates — Azure ships curated workbooks (VM performance, failures, cost) you can clone and adapt.
- As code — export a workbook's definition to JSON and deploy it with Terraform so dashboards are versioned alongside the resources they monitor.
resource "azurerm_application_insights_workbook" "vm_perf" {
name = "11111111-1111-1111-1111-111111111111" # any GUID
resource_group_name = azurerm_resource_group.monitoring.name
location = azurerm_resource_group.monitoring.location
display_name = "VM Performance"
# Author the workbook in the portal, then export "Gallery Template" JSON:
data_json = file("${path.module}/workbooks/vm-perf.json")
}
Tip
Workbooks are the natural home for a KQL query you have already written for investigation or alerting. Promote a good infrastructure KQL example into a workbook tile so the whole team gets it without touching the query editor.
Azure dashboards
Azure portal dashboards are tile boards you assemble by pinning charts (from Metrics Explorer, Log Analytics, or Application Insights). They are good for a shared operational "single pane" but are less interactive than workbooks — no parameters, limited layout. Share via RBAC on the dashboard resource.
Azure Managed Grafana
If your team already knows and prefers Grafana, Azure Managed Grafana is a fully managed Grafana service with native Azure Monitor and Managed Prometheus data sources and Entra ID authentication. It lets you keep a single Grafana experience across the self-hosted and Azure-native stacks — the same dashboards can query Prometheus and Azure Monitor together.
- Native Azure Monitor data source (metrics + logs) and Azure Monitor Managed Service for Prometheus (PromQL).
- Managed upgrades, high availability, and Entra ID / RBAC integration.
- Import existing Grafana dashboards, including the community dashboards used with the self-hosted stack.
Note
This bridges the two stacks: teams standardizing on Grafana can run self-hosted Grafana for on-prem/Kubernetes and Azure Managed Grafana for Azure — or point a single Managed Grafana instance at both Prometheus and Azure Monitor.
Related
- Azure Monitor overview — the stack and its components.
- Data Collection — the telemetry these tools visualize.
- Azure Alerts — alerting on the same signals.
- Grafana — the self-hosted visualization layer.
- KQL — the query language behind workbook log tiles.