Note: Thank you for participating in our Metrics beta program. After careful consideration, we have ended the program and retired the Metrics beta solution on October 7th. We're actively developing a new solution that we believe will make tracking and debugging any issues in your application easier than ever.
Why did Sentry end the Metrics beta?
With Metrics, we set out to make it easier for you to track the data points that matter for your application so you can get to the root cause of issues impacting your users faster.
Based on your feedback, we’re evolving the product to deliver a more intuitive and connected debugging workflow that ensures metrics are always connected to spans and contextually linked with the rest of your Sentry data.
Our goal is also to simplify the product and purchasing experience by fully integrating Metrics with Tracing, allowing us to avoid separate charges for metrics and ensuring you can measure and debug seamlessly.
With significant changes on the horizon and no easy migration path, we want to ensure you don't invest further time in a product that will no longer be supported long-term.
What changes in the product?
The Metrics page and all adjacent functionality such as alerts or dashboard widgets using custom metrics are no longer available.
What happens to my metrics?
Any metrics sent after October 7th will not be accepted, Sentry will drop these automatically. All metrics sent before that date will still be retained for a period of up to 90 days depending on your plan. During this time, they are available in read-only mode via this link.
Will Sentry offer a metrics product in the future?
Yes, we're already working on a new solution to help you easily track the metrics that matter most, so you can quickly debug any issues in your application.
The new solution will allow you to run real-time queries and aggregations directly on your span data. Unlike traditional metrics approaches, which rely on pre-defined tags, limit your ability to drill down, and force you to worry about cardinality issues, this approach will keep raw data for flexible querying at any time. By attaching custom attributes like payloadSize or cartAmount to spans, you can perform aggregations like “Show me the p95 duration of spans where cartAmount>100” using these attributes.
You’ll be able to analyze historical data without worrying about how you set up tags or aggregation rules in advance. Since metrics are directly connected to application traces, you can more easily debug issues - viewing both historical trends and individual samples without having to switch screens or deal with fragmented data sources. This unified approach will provide deeper insights into your application’s health and performance, helping you solve problems more quickly and efficiently.
We will be publishing more updates regarding the new metric solution later in the year.
Have another question?
If you have other questions about this change and how it impacts your organization, please submit a request or reach out on Discord or GitHub.