Vercel Adds Custom metrics to Observability for App-Level Monitoring.

Vercel is expanding its built-in monitoring beyond infrastructure signals. With Custom metrics, developers can now emit application-specific measurements from Vercel Functions, then analyze that data alongside the platform’s existing observability information.

The feature, detailed in an official changelog entry published August 20, 2026, gives teams a native way to track values such as database latency, transaction activity and events generated by custom business logic. Previously, those measurements often required a separate analytics or monitoring service.

Custom metrics arrive as Vercel also broadens its production diagnostics, following the addition of always-on tracing for production and preview traffic. Together, the releases push Vercel Observability toward a more complete view of how applications behave, not just how the underlying deployment performs.

Developers record a metric using the metric() function from the @vercel/functions package. The function accepts a metric name, a numeric value and optional attributes that can later be used to filter or group the results.

Vercel’s example measures the duration of a database query:

import { metric } from "@vercel/functions";

metric("database.query_ms", 120, {
  table: "my-table",
  db: "my-database"
});

In this case, the application records a value of 120 for database.query_ms, while identifying the relevant table and database. A team could use the same approach to measure checkout processing time, jobs completed, AI inference duration or another numerical signal relevant to its application.

Vercel automatically attaches contextual attributes, including the source deployment and the region where the function ran. That reduces the need for developers to add routine infrastructure metadata themselves. It also makes it possible to investigate whether an application-level change is isolated to a particular deployment or location.

Once collected, the measurements are available through three parts of Vercel Observability:

• Query builder: Metrics can be filtered, grouped and aggregated interactively.
• Notebooks: Teams can turn queries into dashboards and share the resulting analysis.
• CLI: Developers can retrieve metrics from a terminal using the vc metrics command.

The combination matters for smaller teams. A developer can investigate an incident from the command line, while a product or operations team can follow the same underlying signal through a shared notebook.

Default platform metrics are useful for identifying broad technical problems, but they do not necessarily explain what an application was doing when the issue occurred. A latency spike, for example, becomes more actionable when it can be compared with a custom measurement for database query time or another workload-specific operation.

Business events can also sit closer to operational data. A SaaS company could record the volume of completed imports, while a marketplace might measure the number of transactions processed by a function. Because attributes can be attached to each measurement, teams can segment results without creating a separate metric name for every variation.

There are cost and access considerations. Vercel says Custom metrics are available to Pro and Enterprise customers using Observability Plus. The measurements are billed as observability events, so teams generating high-volume metrics will need to account for that event usage when deciding what to instrument.

The announcement does not specify separate pricing, retention periods or event limits for Custom metrics. Teams evaluating the feature should therefore consult their current Observability Plus terms before instrumenting frequently executed functions.

What is Vercel Custom metrics?

Custom metrics is a Vercel Observability feature that lets developers record application-specific numerical measurements from Vercel Functions. Those measurements can include attributes for filtering and grouping, and Vercel automatically adds context such as the deployment and function region.

Which Vercel plans support Custom metrics?

Custom metrics are available on Pro and Enterprise plans with Observability Plus. Vercel says the emitted measurements are billed as observability events.

Where can teams analyze Custom metrics?

Custom metrics can be analyzed in Vercel Observability’s query builder and used in shareable Notebooks. Developers can also query them from the command line with vc metrics.