For Zapier, Make, n8n and Pipedream
Know the moment an automation stops, before your client does.
Add one HTTP step to a Zap, scenario or workflow. If it stops running, runs far less than usual, or starts sending blank data, StepCanary tells you in plain English.
Free for 3 automations. No card needed. Set up in about three minutes.
Automations fail quietly. These are the four ways we catch.
The Zap was switched off, a connection expired, or the trigger quietly stopped firing. You hear about it within minutes, not when a client asks where their leads went.
StepCanary learns what a normal Tuesday looks like. “Usually about 40 runs, only 3 yesterday” lands in your inbox at 8am.
A mapped field starts arriving empty after someone renames a form question. We notice the field went blank, without ever storing what was in it.
A workflow that triggers itself can burn a month of tasks in an hour. We spot the flood and tell you once, not five hundred times.
How it works
Name it after the automation and say how often it should run, or let us learn that for you.
Add an HTTP step as the last step of your automation and paste in the URL we give you. Press “Test” and watch it connect.
Every run checks in. If the check-ins stop or change, you get an email, Slack or Discord message.
Your automation “New lead → CRM” has stopped checking in.
What we expected: at least one ping every 2 hours, counted only during Mon–Fri, 09:00–18:00 (America/New_York), with 15 minutes of grace.
What happened: nothing has arrived in time. The last successful ping arrived Tue 10 Nov 2026, 10:42 EST (2 hours 15 minutes earlier).
Common causes: the workflow was turned off or paused, an earlier step is failing, a connected account needs reconnecting, or the trigger has stopped firing.
A real alert. It says which automation, what we expected, and what happened.
Built to be trusted, not to be noisy
Lead forms sleep at night. Give a check active hours and the timer simply pauses outside them.
We monitor ourselves every minute. If our side has a problem, you see “monitoring gap”, not a false “your automation is down”.
Payload monitoring looks at field names and whether each one is filled in. The contents are never stored.
Pricing
- 30 checks
- Slack, Discord and webhook alerts
- Schedules and active windows
- Volume baseline
- Payload drift alerts
- 150 checks
- Everything in Pro
- Client folders
- Weekly client health summary emails
Everyone starts on Free. Upgrade from the Plan page when you need more; cancel any time.
Agencies: put each client’s automations in a folder and get a weekly health summary per client, ready to forward as proof of maintenance.
Questions
Do you need access to my Zapier or Make account?
No. There is nothing to connect or authorise. You add one ordinary HTTP step to the automation, pointing at a private URL we give you. Remove the step and we are gone.
What do you store?
Your email address, your check settings, and when each ping arrived. If you turn on payload monitoring we keep the names and types of the fields (“email: text”), never the values. Payments are handled by Stripe.
Will it wake me up for nothing?
It is built not to. Business-hours automations can be given active hours so quiet nights and weekends never alert. Volume checks are compared by day of the week. And if we have a problem on our side, we show a “monitoring gap” instead of telling you your automation is down.
Does it work on Zapier’s free plan?
The step uses “Webhooks by Zapier”, which Zapier includes on its paid plans. Make, n8n and Pipedream include HTTP steps on every plan.
What happens if I move to a smaller plan?
Nothing is deleted. Checks over the smaller plan’s limit are paused, starting with the ones that pinged least recently, and resume if you upgrade again.
Can I cancel?
Any time, from the Plan page. You keep what you paid for until the end of the billing period.