Metrics (Trace Connected) in Beta

[Dhrumil-Sentry]

We’re introducing Metrics that let you send counters, gauges, and distributions from your code to track things like email.sent, checkout.failed, or queue.depth — and pivot directly into the related traces, logs, and errors when something looks off.

TraceConnectedMetrics.mp4

If you'd like access for JavaScript or Python please comment with your org slug below or drop us a note at feedback-metrics@sentry.io

Why we built this

Metrics in most tools stop at charts — they show that something changed, but not why. When a checkout.failed counter spikes, you’re forced to switch tools and manually line up logs or traces to find the cause.

With Sentry, every metric event carries a trace_id.

That means you can:

• click into a sample and see the exact trace that produced the metric,
• open the related spans, logs, and errors,
• understand why the metric moved — slow database query, failed API call, bad deploy, etc.

So instead of “metric alert → guesswork,” debugging becomes a single flow: Metric spike → open a sample → trace waterfall → root cause.

How the UX works

Aggregates tab shows trends and totals across any attributes (e.g. sum(metrics_value) grouped by email_type).

Samples tab shows individual metric events with direct links to their trace — so you can jump straight into the logs, spans, and errors that occurred at the same moment. Allowing you to drill into any spikes or dips in your metrics and find relevant errors,traces or logs to look at.

How to get started

If you'd like access please comment with your org slug below or drop us a note at feedback-metrics@sentry.io

Python

You need to be on a min version of 2.43.0 of our Python SDKs

import sentry_sdk
from sentry_sdk import metrics

sentry_sdk.init(
    dsn="___PUBLIC_DSN___",
    _experiments={
        "enable_metrics": True,
    },
)
metrics.count("checkout.failed", 1, attributes={"route": "/checkout", "tenant": "acme"})
metrics.gauge("queue.depth", 42, attributes={"queue": "emails"})
metrics.distribution("cart.amount_usd", 187.5, attributes={"currency": "USD", "tenant": "acme"})

More details: getsentry/sentry-python#5042

JavaScript

You need to be on a min version of 10.20.0 for our JavaScript SDKs

Sentry.init({
  dsn: '<your-dsn>',
  _experiments: { enableMetrics: true },
});

// Counter — increment a metric (defaults to +1)
Sentry.metrics.count('checkout.failed', 1, {
  attributes: {
    route: '/checkout',
    tenant: 'acme',
    provider: 'stripe',
  },
});

// Gauge — set a current value
Sentry.metrics.gauge('queue.depth', 42, {
  unit: 'items',
  attributes: {
    queue: 'emails',
    region: 'us-east-1',
  },
});

// Distribution — record numeric values to compute p95, avg, sum, etc.
Sentry.metrics.distribution('cart.amount_usd', 187.5, {
  unit: 'usd',
  attributes: {
    currency: 'USD',
    tenant: 'acme',
  },
});

Each metric event includes:

• name, kind, value, attributes
• trace_id, span_id → used to link with other telemetry
• optional unit, byte_size

More details: getsentry/sentry-javascript#18055

When (and why) to use Metrics

Metrics are best for application and code based health signals — numeric indicators of whether your application logic is behaving as expected.

You should reach for metrics when you want to quantify what your code is doing

Examples:

• email.sent / email.failed
• job.processed / job.retried
• checkout.started / checkout.failed
• payment.declined / refund.issued
• queue.depth / pool.in_use
• search.zero_results

Feedback

We’re working on shipping Alerts, Saved Queries and dashboard widget building support very soon

We’d love to hear your feedback about the following

• Are the SDK APIs easy enough to send your desired metrics ?
• Any feedback about the Metrics Explorer UX - What could make this a better experience for you ?
• Any default metrics that you’d like Sentry to send automatically