Tailoring Cloud Provider Observability: A Guide to Customizing Dashboards in Grafana Cloud
Whether you're monitoring AWS, Azure, or Google Cloud, Cloud Provider Observability in Grafana Cloud gives you prebuilt dashboards and drill-downs right out of the box. But every team has unique workflows, and you may already have trusted internal dashboards or want a view that better fits your debugging process. Now you can customize these cloud service views without leaving the app. This guide explains three key customization methods: connecting an existing dashboard, creating one with AI and wiring it in, and editing instance-level drill-down panels. You'll also learn where to manage these settings and how they stay consistent across Grafana surfaces like the entity graph and Database Observability.
What customization options does Cloud Provider Observability offer for AWS, Azure, and Google Cloud?
Cloud Provider Observability provides three powerful customization capabilities. First, you can set a default dashboard—either the built-in view or one of your own—so that when users open a cloud service from the Services tab, entity graph, or other entry points, they see exactly what you want. Second, you can add custom dashboards as quick links, giving your team easy access to other trusted views. Third, you can customize the instance drill-down panels that appear when you click into a single instance. This means the same drill-down view is used everywhere: in Cloud Provider Observability, Database Observability, and the entity graph. Additionally, you can create dashboards using AI generation and add them just like any other custom dashboard. All these options live on a single Configure page for each cloud service.
How can I set an existing dashboard as the default view for a cloud service?
If you already have a dashboard that perfectly fits a cloud service—say, your internal RDS or Lambda view—you can attach it and optionally make it the default. Go to the Services tab and click Configure for the service you want to edit. In the section labeled “Customize your quick links and add new ones to your custom dashboards,” use the “Select a dashboard” dropdown to pick your dashboard. It appears as a quick link. To make it the default view, mark it as default in the same section. This ensures that users who open that service from anywhere in Grafana (services tab, entity graph, etc.) land on your chosen dashboard. The built-in view remains available as another quick link, so your team can still access it when needed. Changes are saved per service and reused across all relevant Grafana surfaces.
How do I create and wire in an AI-generated dashboard?
Grafana Cloud includes AI capabilities to help you build dashboards tailored to your needs. To create one for a specific cloud service, start by generating a dashboard that uses the correct variables and methodology (for example, using the $CloudProvider variable for filtering). Once your AI-generated dashboard is ready, you add it like any other custom dashboard. Navigate to the service’s Configure page, find the quick links section, and select your new AI dashboard from the dropdown. You can also set it as the default view for that service if you want it to be the primary front door. This integrates the AI dashboard into the same workflows and debugging paths as manual dashboards. The ai approach saves time and helps you create comprehensive views without starting from scratch. Remember to test that the dashboard’s queries and variables work correctly with the cloud provider data before setting it as default.
How can I edit the instance drill-down views in Cloud Provider Observability?
The instance-level drill-down views appear when you click into a single resource (like one EC2 instance or one RDS database) from any Grafana surface: Cloud Provider Observability, Database Observability, or the entity graph. You can tailor these views on the service’s Configure page. Look for the section labeled “Customize the panels…” (or similar). Here you can add, remove, or modify the panels and their queries that render when drilling into an instance. For example, you might replace a default CPU panel with a custom query that shows memory usage over a specific time window. These changes are global—meaning the same panels appear everywhere the instance drill-down is used. This ensures consistency for your team. You can also reorder panels and adjust their sizes. Save your changes, and the next time someone clicks into an instance of that service type, they’ll see your customized layout.
Where do I manage all these customizations?
All customization for a given cloud service (e.g., Amazon RDS, GCP Cloud SQL, Azure Virtual Machines) is managed from that service’s Configure page. To access it, go to the Services tab in Cloud Provider Observability and click Configure for the service you want to edit. The page is divided into clear sections. At the top, you’ll see the Preconfigured dashboard—the built-in out-of-the-box view. Below that, you can manage Custom dashboards that you add as quick links, with the ability to mark one as default. Next, there’s a section for Instance drill-down customization where you adjust panels and queries. Finally, you’ll find Explore-style links for metrics and Grafana Metrics Drilldown. Everything you add or change here is saved per service and automatically applied everywhere that service appears in Grafana—whether it’s the services tab, entity graph, or Database Observability. No need to configure the same settings in multiple places.
How are customizations applied consistently across different parts of Grafana?
One of the key benefits of this customization feature is consistency. When you set a default dashboard on a service’s Configure page, that dashboard becomes the default view for that service everywhere—whether you open it from the Services tab, the entity graph, or any other entry point. Similarly, the instance drill-down panels you configure under “Customize the panels…” appear identically in Cloud Provider Observability, Database Observability, the entity graph, and anywhere else that shows a drill-down for that instance type. This means your team gets a unified experience: they never have to wonder why the dashboard looks different in one view versus another. The same applies to quick links—they show up consistently. This single-source-of-configuration approach reduces confusion and saves time, especially in complex environments where multiple teams or surfaces reference the same cloud services. Remember that changes take effect immediately after saving.