Metrics

The Metrics page (sidebar > Metrics) gives you a deeper analytical view of your API usage. It shows your overall outcome distribution and breaks down traffic by API key, client IP, or target domain.

For real-time summary cards and timeline charts (Concurrency, Requests, Bytes, Response Time), use the Dashboard Overview page.

Filters

Two controls sit above the page: an API key selector and an interval selector.

Key Filter

Use the API Key dropdown to scope the page to a specific key. Only keys within your access scope appear: personal keys, org keys you administer, or keys shared through teams.

Interval Selector

Use the interval buttons to control the time window:

Button Window
1M Last 1 minute
5M Last 5 minutes
30M Last 30 minutes
1H Last 1 hour
6H Last 6 hours
1D Last 24 hours

Both the donut and the table follow whatever you pick.

Outcome Distribution

A donut chart at the top shows the success/error breakdown for the selected interval. Hover any slice to see the exact request count and percentage. Use it to quickly spot whether a sudden drop in success rate lines up with a specific outcome category.

Outcome Types

Every API request is classified into one of these outcomes:

Outcome Meaning
success Target returned HTTP 200 with no errors
application_error Target returned HTTP 200 but the response contained an error
application_fail Target returned a non-200 status code
client_error Your request had invalid parameters or format
rate_limit Request was rejected due to rate limiting
service_error Unexpected error in the request pipeline
service_fail Network failure (timeout, connection refused, DNS error)

Multi-View Usage Table

Below the donut, a usage table breaks down your traffic with three view tabs:

Tab Groups data by
API Key Each of your API keys
Client IP Source IP addresses making requests
Domain Target domains in your requests

Each tab supports four scope chips on the right that change the columns shown:

Scope What it shows
Bandwidth Request count, bytes in, bytes out
Response Time Request count, min, average, and max latency
Concurrency Request count plus concurrent request counts (API Key view only)
Outcomes Request count plus a per-outcome breakdown

Concurrency only makes sense per API key, so the chip is hidden when you switch to the Client IP or Domain view.

Last updated: May 6, 2026