If you’ve been looking for a way to pull operational metrics out of GraphOS programmatically, you might not know that the Apollo Platform API already has you covered. There are several fields available today for querying usage and performance data directly with no Studio dashboard required.
Here’s a quick rundown of what’s available:
topOperationsReport: A point-in-time snapshot of operations run against your graph. Returns operation name, ID, signature, and request counts for a given time range, with optional filtering by client name and version. Useful for generating reports or pulling a full list of operations executed in a given window (up to 10,000).
operationInsightsTimeseriesReport: Time-bucketed operation metrics including request counts, latency percentiles, and error rates. You can group by variant, operation name, client, and more, and choose your resolution (minute, hour, or day). Returns data as GraphQL records or CSV.
subgraphInsightsTimeseriesReport: The same timeseries format, but focused on subgraph and connector fetch counts and latencies. Great for tracking service-level trends across your federation.
schemaCoordinateInsightsTimeseriesReport: Usage and error data scoped to specific object fields, input fields, or enum values. Handy for tracking field adoption or identifying problem areas in your schema.
You can explore the schema and try out sample queries in the Platform API Explorer, and our full blog post walks through concrete example queries for common use cases.
Have you been using the Platform API for metrics or reporting? We’d love to hear how, drop your use case or any questions below!