Sentry is the most popular error tracking tool for developers. The free tier is generous, the SDK is excellent, and session replay is genuinely useful. But when your app hits a spike (a bad deploy that throws 100,000 errors in an hour) the pricing model starts to matter.
Pricing Tiers
| Developer (Free) | Team ($26/mo) | Business ($80/mo) | |
|---|---|---|---|
| Errors/month | 5,000 | 50,000 | 100,000 |
| Performance transactions | 10,000 | 100,000 | 100,000 |
| Session replays | 50 | 500 | 500 |
| Users | 1 | Unlimited | Unlimited |
| Data retention | 30 days | 90 days | 90 days |
| SSO / SAML | No | No | Yes |
| Custom dashboards | No | Yes | Yes |
| Overage pricing | Events dropped | ~$0.000315/error | ~$0.000290/error |
Real Cost Examples
Solo dev, side project
Errors: ~2,000/month
Plan: Developer (free)
Cost: $0/monthTeam of 5, production SaaS
Errors: ~80,000/month (50K base + 30K overage)
Plan: Team ($26/month)
Overage: 30,000 × $0.000315 = $9.45
Cost: ~$36/monthTeam of 10, high-traffic app
Errors: ~300,000/month
Plan: Business ($80/month)
Overage: 200,000 × $0.000290 = $58
Cost: ~$138/monthBad day scenario (production incident)
Normal: 50,000 errors/month
Bad deploy: +500,000 errors in one day
Plan: Team ($26/month)
Overage: 500,000 × $0.000315 = $157.50
Cost that month: ~$184
Tip: set a spending cap in Sentry settings to prevent thisWhat the Free Tier Actually Includes
Sentry's free tier is genuinely useful, not a crippled trial:
- 5,000 errors/month (dropped, not billed, when exceeded)
- 10,000 performance transactions/month
- 50 session replays/month
- All SDKs (Next.js, React, Node, Python, etc.)
- Error grouping, stack traces, breadcrumbs
- Release tracking
- 1 user only, that's the main limitation
For a solo developer with a production app, this is enough. The moment you need a second team member with access, you need Team ($26/month).
Sentry vs Competitors: Price Comparison
| Tool | What it does | Starting price | Free tier |
|---|---|---|---|
| Sentry | Error tracking + perf | $26/mo | 5K errors, 1 user |
| Datadog Error Tracking | APM + errors | $31/host/mo | 14-day trial |
| Bugsnag | Error tracking | $59/mo | 7,500 events |
| Rollbar | Error tracking | $13/mo | 5,000 events |
| New Relic Errors | Full observability | $49/user/mo | 100GB data |
Sentry is the best value for dedicated error tracking. Datadog and New Relic are full observability platforms that happen to include error tracking, much more expensive if errors are your primary need.
When Sentry Is Enough (and When It's Not)
Sentry is enough when:
- Error tracking is your primary monitoring need
- You want session replay to debug user-facing issues
- Basic performance monitoring (transaction tracing) is sufficient
- Budget is under $100/month for monitoring
Sentry is NOT enough when:
- You need real-time API health monitoring (uptime, latency P95, error rates per endpoint)
- You need infrastructure metrics (CPU, memory, disk)
- You need instant alerts when an endpoint degrades (Sentry alerts on errors, not latency)
- You need WhatsApp/SMS alerts that bypass DND at 3 AM
The Best Combo for Next.js Teams
Sentry (error tracking + session replay) + a dedicated uptime / API monitoring tool (for example UptimeRobot or Better Stack) for external checks and alerts.
- Sentry catches errors with stack traces and tells you which deploy caused them
- The monitoring tool checks API health from outside (status codes, response time) and alerts the on-call person when an endpoint degrades or goes down
- Together: you know WHAT broke (Sentry) and WHEN it started affecting users (the monitor)