Better Stack explained: uptime monitoring, on-call, status pages, logs

What Better Stack includes: the product areas

The Better Stack homepage groups the platform into separate products that share one account. The ones most IT and DevOps teams start with are:

• Uptime monitoring External checks of websites, APIs, servers and cron jobs, with screenshots of errors and traceroute and MTR output for timeouts.
• Incident management and on-call On-call schedules, escalation policies, phone and SMS alerts, and incident handling inside Slack or Microsoft Teams.
• Status pages A branded public or private page, usually at status.yourdomain.com.
• Telemetry Log management, infrastructure metrics and OpenTelemetry tracing, queried with SQL, PromQL or a drag and drop builder.
• Error tracking An exception tracker that works with Sentry SDKs.
• Real user monitoring Session replay, web vitals and product analytics.

The product list also includes an AI SRE. Better Stack says this assistant suggests hypotheses but never takes any automated actions without your approval, so it is an investigation aid rather than an automation engine.

How Better Stack uptime monitoring works

The basic check is the HTTP status code monitor. According to the monitoring quick start, it checks your URL every X seconds for a 2XX success code. If the URL fails, the monitor creates an incident and notifies the current on-call person, and if that person does not respond, the entire team is alerted.

Creating one takes a few steps: go to Monitors, enter the URL or IP address, choose how to escalate, and save. Beyond HTTP checks, the monitor types in the docs cover keyword, ping, heartbeat, Playwright, API, DNS, domain expiration, SSL certificate and TCP/UDP port monitors, plus monitoring of private networks.

A few details matter when you plan a rollout:

• Check frequency The uptime product page puts the free tier at 3-minute checks, while paid monitors go down to 30-second checks.
• Heartbeats Cover the reverse case: your cron job or serverless worker pings Better Stack, and you are alerted if it does not run.
• Playwright transaction checks Run scripted user flows in a real Chrome browser.
• Firewalls Better Stack recommends allowlisting by User-Agent rather than IP address, because its probe IPs can change without notice. The User-Agent string still contains "Better Uptime Bot", a leftover of the old name.

How on-call, alerting and escalation work

The default alert flow

Out of the box, the on-call guide describes a three-step flow: a monitor or integration triggers an incident, the current on-call person is notified first, and if they do not acknowledge it (for example after 3 minutes) it escalates to everyone on the team. The person on the schedule marked Primary gets new incidents.

For each monitor you choose the channels: phone call, SMS, e-mail, push notification, or a critical alert that ignores the mute switch and Do Not Disturb. The mobile apps let responders acknowledge or resolve from the notification itself. Phone call alerts only reach a listed set of countries, so check the list if your responders are spread across regions.

On-call schedules

Schedules are rotations: you pick colleagues, set how often shifts change (for example every day) and save. You can add one-off overrides when someone steps in, and responders get a reminder by e-mail and push notification 24 hours before their shift by default. Several schedules can be combined for backup schedules or follow-the-sun rotations.

Escalation policies

An escalation policy is a set of rules that defines who is alerted and in what order until someone acknowledges the incident. Policies can alert specific people, the current on-call team, Slack or Microsoft Teams, Zapier, webhooks or the AI SRE. Steps can wait a set number of minutes, apply time-based rules (stop or hand over to another policy at given times), route on incident metadata such as severity or ownership, post runbook instructions into the incident timeline, and repeat before falling back to another policy.

One operational detail trips people up: creating a policy does not apply it. You assign it to each monitor, heartbeat or integration, individually or in bulk. Monitors left on "Notify the primary responder" use a simple escalation that e-mails the primary on-call, escalates to the team and triggers all integrations.

Slack and Microsoft Teams

The Slack integration posts incidents into a channel with buttons to acknowledge them, and Better Stack can create a dedicated channel per incident or a thread in your incidents channel. A Slack user group such as @primary-on-call stays in sync with the schedule, and /cover lets a responder ask colleagues to take a shift. The pricing page includes a basic single-channel integration and charges extra for channel-based and thread-based workflows.

Better Stack status pages

A status page tells users about outages and scheduled maintenance so they do not need to contact support. The status page guide covers both public pages and private ones protected by a password, which suits internal stakeholders or specific clients. Setup is: pick a *.betteruptime.com subdomain or your own custom subdomain, add your logo, then choose which monitors, heartbeats and third-party services appear as components.

The status page product page says a custom subdomain needs a single CNAME record, and customers can subscribe to the whole page or only selected components. The pricing page describes monitoring and incident management as fully integrated with status pages, so the monitors that page your team also feed what customers see.

Logs, metrics and traces (Telemetry)

Telemetry is the part that came from Logtail. The logging quick start starts with a source (an application, web server or cluster), gives it a source token, and has you send a test log over HTTP, then watch it arrive in Live tail. Guides cover the Better Stack collector, Node.js, Python, Ruby on Rails, Nginx, Apache, Ubuntu, Docker and Kubernetes. Logs can be queried with SQL, and alerts can fire on the presence or absence of log patterns.

The homepage lists several cost controls aimed at log-heavy teams: marking irrelevant logs as spam so they are not billed, storing logs in your own S3 bucket, and querying logs, spans or metrics with SQL over an HTTP API or an MCP server. Metrics can come from Prometheus or OpenTelemetry collectors. If you are weighing Better Stack against larger observability stacks, the Grafana vs Datadog comparison shows the trade-offs on the other side.

Better Stack integrations

The integrations section of the Better Stack docs lists Slack, Microsoft Teams, Datadog, New Relic, Prometheus, Grafana, Zabbix, AWS CloudWatch, Amazon SNS, Google Cloud, Jira, Linear and Zapier, plus incoming and outgoing webhooks and e-mail. There is also a migration path from PagerDuty. For infrastructure as code, the pricing page lists a Terraform provider and a REST API across the incident products, and the error tracker accepts Sentry SDKs from 100+ platforms.

In practice this means Better Stack can sit downstream of an existing monitoring tool: Datadog, Grafana or Prometheus alerts arrive as incidents, and Better Stack handles the paging, escalation and status page.

How Better Stack pricing works

Pricing is modular. The figures below come from the Better Stack pricing page as published in November 2026; it offers monthly or annual billing (annual gives about two months free) and a 60-day money-back guarantee.

Two points shape the real bill. First, phone and SMS alerts are tied to the paid Responder licence: the free plan alerts by Slack and e-mail only, while the homepage says unlimited phone calls, SMS and push notifications come with every Responder licence. Second, telemetry prices vary by data region. The page shows several price sets for the same bundle, with the Nano bundle starting at $25 per month billed yearly ($30 monthly) for 40 GB each of traces, logs and metrics. Status page extras such as white-labelling, password protection, IP allowlisting and SSO are priced per page, so a customer-facing private page can cost more than the base plan.

Deployment, security and data location

Better Stack is a hosted service. The pricing page FAQ says data is stored in EU regions by default in GDPR-compliant, DIN ISO/IEC 27001-certified data centers, that Better Stack is SOC 2 Type 2 compliant, and that it is not HIPAA compliant. Data is encrypted at rest with AES-256 and in transit with TLS. For teams that need more control, the page lists hosting telemetry in your own bucket as a paid add-on, and custom data residency and custom VPC deployment at the enterprise level. Single sign-on with Google is included; Azure and Okta SSO cost extra per user, and generic SAML is an enterprise feature.

Who Better Stack fits (and who it does not)

Better Stack fits well when:

1. A small or mid-size team wants monitoring, on-call paging and a status page from one vendor instead of combining a checker, a paging tool and a status page service.
2. You already run Datadog, Grafana, Prometheus or Zabbix and want a cheaper place to handle alert routing and escalation.
3. You want to start for free and pay per responder as the on-call rotation grows, rather than per seat for everyone who views dashboards.

It fits less well when you need HIPAA compliance, which the vendor says it does not offer, or when policy requires a fully self-hosted tool, since the deployment options it lists outside its cloud are enterprise items. Teams with very large log volumes should model the per-GB pricing in their own data region before committing. To compare it with other options, browse the monitoring and observability category.