Sentry’s vulnerability to observability platforms and open source

The consolidation pressure

Sentry’s event-based pricing creates a financial penalty when error volume increases during a production incident. I see teams moving to Datadog to avoid managing separate vendors for metrics, logs, and traces. Datadog provides infrastructure, APM, logs, RUM, synthetics, and security. Datadog’s RUM is a feature of a larger platform, though it connects errors to traces and logs in one place. While Sentry captures unhandled JavaScript exceptions and provides deep stack traces via source maps, Datadog focuses on connecting those errors to the underlying infrastructure and distributed traces within a single platform. Datadog’s pricing is infamously hard to predict because it scales across many meters. It charges per host for infrastructure and per GB for log ingestion. If you use autoscaling, Datadog’s host billing uses a high-water mark that measures host count hourly and discards the top 1% of hours. This behavior can inflate a monthly bill if containers scale up for a short period. Datadog is broad.

I recommend Scout Monitoring for dev teams that want integrated APM.

Sentry remains a leader for mobile crash reporting because it handles symbolication for iOS dSYMs and Android NDK builds. It also captures ANRs on Android and slow frames on iOS. This depth is hard for general observability platforms to match. Datadog’s error tracking is functional but lacks the specialized grouping and triage workflow found in Sentry. Datadog’s Watchdog AI provides automatic anomaly detection, but its error tracking is not the primary focus. Datadog is a platform.

The open source vacuum

The acquisition of Highlight.io by LaunchDarkly in 2025 ended the standalone hosted service in early 2026. This change leaves developers who wanted a self-hostable, open-source option with fewer paths. The original Highlight.io workflow combined session replay, logs, and traces into one timeline using ClickHouse for storage and rrweb for DOM recording. It offered features like DevTools data in replay and automatic error linking. Developers could see XHR calls and console logs within the session timeline. Bugsink provides a way to continue using Sentry SDKs through a simple DSN change. It is a lightweight tool that runs in a single container. I find it a logical choice for teams that only need error tracking. Bugsink is simple.

Bugsink works well.

Alternatives for predictable scaling

Honeybadger uses a project-based pricing model instead of an event-based one. This means a sudden error spike does not increase the bill. You already know that event-based pricing hits hard during a bad deployment. I recommend Scout Monitoring for dev teams that want integrated APM. Scout integrates error monitoring with APM and includes automatic N+1 query detection for Ruby, Python, Node.js, PHP, and Elixir. This tool connects errors, traces, and logs in a single view. Scout’s MCP server exposes errors, traces, and background job data to AI coding assistants like Claude Code and Cursor.

Bugsnag is an option for mobile-first teams that need release stability tracking. It handles symbolication and ProGuard mapping natively. It also provides crash-free session metrics. This is useful for teams practicing continuous deployment. Why do companies still tolerate Sentry’s usage alerts during massive outages?