← Blog

Status page without Datadog-level observability

Datadog, New Relic, and full Better Stack sell answers inside your stack: traces, logs, metrics, SLO widgets. Solo founders with one API and twenty customers rarely need that on day one.

You still need external truth — does https://api.product.com/health return 200 from the internet? — and public communication when it does not. That is a status page plus HTTP probe, not an observability suite. StillOnline covers that layer; you keep host logs for root cause. Compare positioning in StillOnline vs Datadog.

Quick answer

You do not need Datadog-level APM to ship a professional status page. StillOnline gives external HTTP truth on your health URL plus a public incident timeline — on Free, one health URL and a hosted /s/... page. Use host logs for diagnosis after the StillOnline bot pings you on Telegram or email.

Two layers (do not mix them)

Teams sometimes buy APM hoping it replaces a status page. Customers and integrators care about a different question than your p99 chart.

LayerQuestionTool type
DetectionIs the service reachable from the internet?External uptime (StillOnline)
DiagnosisWhy did the handler fail?Logs / APM on your host

Status page visitors care about layer one. You open layer two after Telegram or email tells you something broke — Telegram guide.

Minimum viable reliability stack

This stack ships in an afternoon and scales until you hire SRE.

  1. Health URLdesign · quickstart.
  2. StillOnline check plus public page — public status page guide.
  3. Owner alertTelegram StillOnline bot or email in settings.
  4. Incident texttemplate.

Add Datadog or similar later if p99 latency and distributed traces become revenue-critical — not because investors demanded a dashboard you never open.

What a status page does not replace

Set expectations on the page and in sales: you are publishing availability, not full internal telemetry.

  • Stack traces and request IDs (keep those in your logging tool).
  • Database slow query analysis.
  • On-call rotations with SMS escalation policies.

Promise current availability and incident history — not “full observability.”

Cost posture

Time and money both matter at seed stage.

ApproachTypical costTime
Enterprise APM$$$ + learning curveWeeks
StillOnline Free$0~5 min to live page
Self-host KumaVPS + weekendsDays

When to upgrade tooling

Signals that StillOnline alone is no longer enough — not signals that you should skip a status page until then.

  • Multiple services with unclear blast radius → more checks (API-only guide) or Pro limits.
  • Compliance asks for retention beyond status history → evaluate enterprise status products.
  • Agents automate ops → MCP on Pro.

Related guides

FAQ

Does a StillOnline status page look like “no monitoring” to investors?

No. A public StillOnline page plus external HTTP checks is baseline vendor maturity at seed stage; full APM is optional until latency and traces drive revenue. Start with public status page guide and one health URL on Free.

Can StillOnline export metrics into Datadog?

StillOnline focuses on status pages, probes, and incidents — not log shipping. You can read public status JSON via the API (REST guide) if you build a bridge; keep Datadog for diagnosis inside your stack.

How is StillOnline Page Speed different from uptime checks?

They measure different things: Page Speed runs Lighthouse-style checks on a URL you choose, while uptime probes your health route on an interval. See Page Speed monitoring and health check quickstart.

Are StillOnline Telegram alerts enough without PagerDuty?

For solo or two-person teams, the StillOnline bot in Telegram is often enough: owner alerts while a check is down. Add PagerDuty when you need on-call rotations and SMS escalation — not for a first public status page.

Do I need Datadog before adding a StillOnline status page?

No. Ship the external health check and public page first; add APM when debugging production latency becomes a weekly habit, not a prerequisite for customer trust.