Pulsegrid watches every service, endpoint, and scheduled job you run, and tells the right engineer the moment something drifts.
On call for 2,400 engineering teams, from four-person startups to regulated fintech
HTTP, TCP, DNS, browser and certificate checks share one timeline, one alert policy and one on-call rotation.
A failing check is retried from nearby regions first. Only when several agree does Pulsegrid open an incident, so a single flaky probe never pages anyone.
regions agree. Incident opened and the primary on-call paged 11 seconds after the first failure.
Cron jobs, queue workers and backups ping Pulsegrid when they finish. Silence is an alert.
Expiry warnings 30, 14 and 3 days out, with the exact link in the chain that is failing.
Link components to monitors. Incidents post updates for you, subscribers hear first.
Every channel is included on every paid plan, with no per-SMS surcharge.
The overview every Pulsegrid team lands on: monitors, uptime and what changed. Switch the range or hover the chart.
Most pages are noise: a flapping check, twelve alerts for one root cause, a SEV-4 at 3 a.m. Pulsegrid filters them before they reach a phone.
Checks that bounce between up and down are held until they settle, then reported once.
Alerts from the same service, region and minute are grouped into a single incident with a single page.
Low-severity alerts wait for business hours. A SEV-1 still gets through, every time.
The incident, the rotation and the public status page live in one place, so nobody pastes updates between tools at 3 a.m.
p95 above 800 ms from Singapore, Tokyo and Sao Paulo. Frankfurt and N. Virginia healthy.
Push, then SMS. Escalates to Jonas Lindgren at 14:12 if unacknowledged.
Linked deploy checkout-api 4.2.1, rolled out to ap-southeast at 13:58.
Checkout API marked degraded in Asia Pacific. 312 subscribers notified.
Latency back under 250 ms in all regions within four minutes.
Five consecutive healthy checks from every region. Postmortem draft created.
Pages, acknowledgements, deploys and status updates are logged as they happen.
Overrides, holidays and follow-the-sun handoffs live in one calendar.
Status components follow their monitors, with email, RSS and webhook subscriptions.
Point-and-click in the dashboard, or keep every check in version control with the CLI and a single YAML file.
$ pulsegrid login
Signed in to Norvane (production)
$ pulsegrid checks add api-checkout \
--url https://api.example.com/v2/checkout/health \
--every 30s --regions fra,iad,sin,nrt,gru,syd \
--confirm 3 --expect-status 200
Created check api-checkout (http, 6 regions)
First results in about 30 seconds# pulsegrid.yml
policies:
platform-oncall:
rotation: platform-primary
escalate:
- after: 10m
to: platform-secondary
- after: 25m
to: eng-managers
channels: [push, sms, voice]
quiet_hours:
severity: [sev3, sev4]
from: "22:00"
to: "08:00"$ pulsegrid status-pages create "Norvane Status" \
--components api-checkout,dashboard,webhooks \
--domain status.example.com
Status page live at https://status.example.com
3 components linked to 11 monitors
Subscribers can follow by email, RSS or webhook
“We went from forty pages a week to eleven. The ones that still come through are real, and the team trusts them again.”
Invite the whole team on every plan. Upgrade, downgrade or cancel whenever you like.
For side projects and your first alerts.
No card required
For product teams with a real on-call rotation.
Billed monthly
For platform teams running many services.
Billed monthly
Custom monitor volumes, a 99.99% uptime SLA, dedicated regions and a named solutions engineer.
| Feature | Hobby | Team | Scale |
|---|
Monitoring touches production. We treat it that way, and we will send the paperwork before you ask.
Audited annually. Report available under NDA.
Choose where results and incident data are stored.
SAML, automatic provisioning and least-privilege roles.
Every change to checks, policies and rotations, exportable.
Book 30 minutes with a solutions engineer. We will review your current alerting and tell you honestly whether Pulsegrid is a fit.
Talk to salesA ping tells you something failed. Pulsegrid confirms the failure from several regions, groups related alerts into one incident, finds the person on call, escalates if they do not answer and updates your status page. The check is the easy part.
Yes. Team and Scale include SMS and voice to 120 countries with a fair-use limit of 500 messages per user per month. Almost no team reaches it, and we will call you before we ever charge for overage.
The importer reads CSV and JSON exports from the most common uptime tools and keeps names, intervals and regions. Teams usually move a few hundred checks in an afternoon, and we will run the import with you on Scale.
Checks run from 38 regions across six continents, on at least two cloud providers per region. You choose EU (Frankfurt) or US (Virginia) storage when you create a workspace, and data never leaves that region.
Check runners, alert delivery and the dashboard are separate systems in separate regions, so a dashboard outage never stops a page. Our own status is public, and we publish a postmortem for every incident.
Yes. Upgrades apply immediately and are prorated. Downgrades and cancellations take effect at the end of the billing period, and you can export every check and incident before you go.
Start with 20 free monitors. Add your team, your rotation and your status page when you are ready.
Free for 20 monitors. Invite your team and set up on-call when you are ready.
Tell us about your stack and your current alerting. A solutions engineer, not a script, will get back to you.
New features, improvements and fixes, shipped every other week.
Live health of the dashboard, check runners and alert delivery.
ChangelogThe link may be old, or the page moved. Everything else is up.