Sentry

What it is

Sentry is an application monitoring platform focused on error and performance tracking. Every uncaught exception, every slow database query, every failed API call: Sentry captures it, groups it, and links it to the exact line of code that caused it. It has SDKs for every major language and framework, with Next.js support that is unusually good.

Who it's for

• Any production app where you care about what's breaking
• Next.js and React teams (the source map integration is first-class)
• Teams that want to tie errors to releases and deployments
• Anyone who's been debugging "it works on my machine" with console.logs
• NOT for: simple apps where server logs are sufficient — overkill if you're not shipping to real users
• NOT for: teams with on-prem monitoring requirements (self-hosting Sentry is possible but operationally complex)

The good

Error grouping is the standout feature. Similar errors are automatically deduplicated and grouped. You don't drown in alert noise. You see "this function has thrown 412 times in the last 24h" not 412 individual alerts.

Source maps work. This sounds basic but isn't. Sentry's Next.js SDK handles source map uploading at build time so your production errors show the original TypeScript, not minified junk. This alone is worth the setup cost.

Performance monitoring earns its keep. Transaction tracing, N+1 query detection, slow endpoint flagging. Not afterthought features, but things that have saved real debugging hours.

Release tracking closes the feedback loop. Tag errors to releases, see if a deploy made error rates spike, set up alerts for error rate thresholds. The deployment pipeline integration is well thought out.

The bad

The free tier runs out fast. 5,000 errors/month sounds like a lot until you have a JS error that fires on every page load. You'll burn through it quickly if something breaks on a popular page. The $26/mo Developer plan covers 50k errors, still finite.

The UI has gotten complex. Sentry has grown into a full observability platform. The dashboard has a lot of surface area and settings. It's not overwhelming, but it's not as focused as it used to be.

Performance pricing adds up. Error tracking and performance monitoring are separately metered on higher plans. Budget both if you want the full picture.

SDK size is non-trivial. The Sentry Next.js SDK adds weight to your bundle and a small startup overhead. Use sampleRate to control it.

Onboarding experience

SDK install and Sentry.init() take under 10 minutes. The Next.js setup wizard handles source map configuration automatically, a step that previously required manual webpack config. First real error appears in the dashboard within minutes of deploying. The onboarding flow is code-first and doesn't gate you behind a demo call. The free tier activates immediately with a working project.

Support reality

Free tier gets documentation and community forums with limited official response. Paid plans (Team and above) get email support with a 24 to 48 hour SLA for non-critical issues. Business and Enterprise tiers get SLA-backed response. The Sentry GitHub and community forums are active. For SDK configuration questions, community answers are typically faster than tickets. For billing or account issues, official support is necessary.

Migration and lock-in

Moderate lock-in. Error data doesn't export in a portable format — moving to a competitor means starting fresh with no historical error history. The SDK instrumentation is partially compatible with OpenTelemetry at the traces layer, but Sentry's alert rules, grouping logic, and dashboards are proprietary configuration. Budget a few weeks to rebuild your alert setup on a new platform if you ever switch. Grafana Cloud and Datadog are the realistic alternatives at scale.

Pricing reality

Free: 5k errors/month, 1 user, 1 project. Fine for side projects with low traffic. Developer plan at $26/mo covers 50k errors, replays, and more projects. This is where most solo devs land. At real scale, the per-event pricing becomes significant; evaluate Grafana Cloud or self-hosted Sentry at that point (it's OSS and self-hostable, though the ops complexity is real).

Vendor signals

• Founded in 2012, OSS core, stable company with clear recurring revenue
• Self-hostable via open-source Sentry (documented, though operationally complex)
• Public changelog and accurate status page with real incident history
• Has expanded scope (performance, replays, crons, profiling) without breaking the core product
• No acquisition risk signals; no concerning pivot signs

Verdict

Sentry is the default answer for error tracking and you'd need a good reason to pick something else. The DX, the source map integration, the grouping: all genuinely better than competitors. The pricing beyond the free tier is the only real friction point.

If you're on the free tier and burning through errors fast: fix the error causing noise, or pay $26/mo, as it is worth the cost. If you're at scale and comparing to Grafana Cloud or Datadog: Sentry's developer experience stays ahead, but cost modeling at 500k+ events/month deserves a careful look.

Would I recommend a colleague start here? Yes. For any production app with real users, not knowing what's broken costs more than $26/month.