PageLantern for SaaS
A SaaS product fails in layers: the API returns the wrong body at a healthy 200, the login page breaks only in a browser, a billing worker silently stops, and customers find out before you do. PageLantern watches each layer from the outside and turns failures into incidents your team can route.
The API layer
For a SaaS product, "the API is up" is a contract about content: the login endpoint issues a token, the list endpoint returns the right shape, the health endpoint says every dependency is green. PageLantern API checks send real requests with your method, headers, and body, then assert on the response — status code, JSONPath expressions against the JSON, and a latency budget — so a 200 with a broken payload fails the check the way it fails your customers.
On paid plans, multi-step API checks chain requests with variable extraction — sign in, take the token, call an authenticated endpoint — and validate responses against a JSON Schema. You can import the endpoints straight from an OpenAPI spec or a Postman collection instead of re-describing your API by hand.
Every layer
The API contract is one layer. A SaaS product also has a front end that breaks only in a browser and a background fleet whose failures produce no HTTP error anywhere.
Hosted Playwright checks walk login, signup, and dashboard flows in a real browser and assert the flow reached its expected state — with screenshots, traces, and console output captured as evidence for triage. 250 minutes a month on Starter, 1,000 on Team.
Every queue worker, billing run, backup, and nightly job gets a unique URL to ping on success. A missed ping past its grace window opens an incident — catching the silent failures uptime checks cannot see. Included on every plan.
Uptime checks on the marketing site and app shell, SSL certificate expiry warnings, and domain expiry checks — the boring failures that are somehow always the embarrassing ones.
Failures open incidents automatically, and paid plans add timeline entries and maintenance windows; Team adds responder assignment and postmortem fields so the follow-up has a home.
Customer trust
During an incident, the cheapest support ticket is the one your status page answered first. PageLantern hosts a public status page driven by your actual monitors — you choose which monitors it reflects, publish incident updates and maintenance windows to it, and customers can subscribe by email for updates. One hosted page is included free; Starter has 3 and Team has 10, which is enough to separate an app status page from a per-product or per-API one.
Alert routing
Alerts deliver to email, Slack, Microsoft Teams, Discord, webhooks, and PagerDuty, plus a metered SMS allowance on paid plans (50 messages a month on Starter, 200 on Team). Notification rules control when a signal becomes a page: thresholds on failure counts, latency, success rate, SSL and domain days remaining, quiet hours, and reminder intervals — so the pager fires on customer impact, not on every blip.
Workspaces have seats and roles: 1 seat on Free, 3 on Starter, and 10 on Team, with role-based access control on paid plans so an engineer, an on-call responder, and a viewer see the right things.
At a glance
| What can fail | The check that watches it | Plan |
|---|---|---|
| API returns wrong data at a healthy 200 | API check with JSONPath assertions | Free |
| Authenticated, multi-request API flows | Multi-step API checks with variable extraction | Starter ($9/mo) |
| Login or signup breaks only in a browser | Hosted browser check with screenshots and traces | Starter ($9/mo) |
| A worker, cron job, or billing run silently stops | Heartbeat monitor with a grace window | Free |
| Certificate or domain quietly expires | SSL and domain expiry monitoring | Free |
| Customers need answers during an incident | Hosted public status page with email subscribers | Free (1 page) |
| On-call needs paging, not email | PagerDuty alert channel | Free (1 channel) |
Plan shapes are the published plans — see the pricing page for the full matrix.
Plans at a glance
| Limit | Free | Starter | Team |
|---|---|---|---|
| Price | $0 | $9/mo or $7/mo annual | $24/mo or $19/mo annual |
| Monitors | 15 | 50 | 500 |
| Check speed | 5 minutes | 1 minute | 30 seconds for 25 critical monitors |
| Users | 1 | 3 | 10 |
| History | 30 days | 90 days | 1 year |
| Status pages | 1 hosted page | 3 pages | 10 pages |
| SMS alerts | Not included | 50 SMS per month | 200 SMS per month |
| Alert channels | 1 channel | 5 channels | Unlimited |
Figures are the published plans — the full matrix is on the pricing page.
Next step
FAQ
No. PageLantern is external synthetic monitoring: it checks your product from the outside — APIs, pages, browser journeys, heartbeats — and turns failures into incidents, alerts, and a status page. It does not instrument your code, collect traces or host metrics, or do log analytics. Many teams run it alongside an APM; small teams often find outside-in checks are the layer they actually needed.
Yes. API checks send your method, headers, and body, so a static token or API key works on any plan. On paid plans, multi-step checks handle real authentication flows: one step signs in, extracts the token from the response, and later steps use it against protected endpoints — with JSONPath and JSON Schema assertions at each step.
Yes — PagerDuty is a built-in channel alongside email, Slack, Microsoft Teams, Discord, webhooks, and SMS. Free includes 1 alert channel, Starter includes 5, and Team is unlimited, with 50 and 200 SMS a month on the paid plans. Notification rules decide which signals page and which just notify.
With heartbeat monitors: each job gets a unique URL to request on success, and you set the expected interval plus a grace window. When a ping is late or stops, PageLantern opens an incident and alerts your channels — no agent, no log scraping, one outbound request from the job. Heartbeats are included on every plan.
Yes. Status pages are public, reflect the monitors you choose, and support email subscribers, incident updates, and maintenance windows. One hosted page is free; Starter includes 3 and Team includes 10.
1 seat on Free, 3 on Starter, and 10 on Team. Paid plans include role-based access control so responders, engineers, and read-only stakeholders get appropriate access, and Team adds responder assignment on incidents plus a full audit log.