In this demo deployment, each hook reads as a sentence: when this happens in this project, do that.
Hooks
When something happens in Countly,
tell the rest of your stack
A hook is one trigger and the actions that follow it. Data arrives, a cohort changes, an alert fires or the clock strikes, and Countly calls your webhook, emails a team or runs your own code. No middleware to host, no polling.
Part of engagement
The automations
teams set up first
Each is one trigger and one or two actions, built in a form, with no code unless you want it.
“When someone signs up, add them to the CRM.”
A tracked event triggers an HTTP request with the whole payload as JSON.
“When a crash appears for the first time, post it to Slack.”
The new crash activity triggers a request to a Slack webhook.
“When a customer slips into the at-risk cohort, email their account manager.”
Entering a cohort triggers an email, to anyone, with or without a Countly account.
“When an alert fires, open a ticket.”
An alert triggers a request to your ticketing tool’s API.
“When a variable changes, tell the release channel.”
Changes to Adaptive Variables are activities a hook can follow.
“Let our backend start it.”
A hook gets its own URL. Call it, and its parameters become the payload.
Triggers and actions
One trigger,
as many actions as you need
The actions run in order, and one failing does not stop the next.
Tracked data
The moment matching data arrives from your apps.
Internal activity
Something Countly itself did.
Schedule and API
On the clock, or when something calls it.
Actions
What happens next.
Built in a form
Set up in a minute,
tested before it runs
Pick the trigger, add the actions, run it once for real, then turn it on.
Safe by design
Automation that cannot
be turned against you
A hook sends your data out, so the limits are part of the design.
- 01HTTP
No reaching inside your network
Destination addresses are checked against internal and private ranges when a hook is saved and every time it runs.
- 02Code
A sandbox for your code
Custom JavaScript runs isolated, for up to three seconds and 128 MB, with no network, files or access to the rest of Countly.
- 03HTTP
Requests that cannot hang
A destination that does not answer within fifteen seconds is recorded as a failure.
- 04Monitoring
Failures where you can see them
Each hook keeps its ten most recent failures with their errors.
- 05Server
A rate limit if you want one
Cap how often hooks run in a time window. Runs beyond it are dropped and recorded.
Works with
Hooks, with the
rest of Countly
Every feature reads the same live data, so an answer in Hooks doesn’t stop there. These are the handoffs built in.
All featuresData ownership
Your data goes
only where you send it
Hooks run on your Countly server. Nothing is forwarded until you point a hook at a destination.
Data privacyNo middleware
No integration service to host or trust in between. The server that holds your data makes the call.
Send only what is needed
Reshape the payload before it leaves, so the receiving system gets three fields, not thirty.
Self-hosted, or dedicated
Run Countly in your own cloud, your data center or an air-gapped network. Or let us run a single-tenant instance in the Google Cloud region you choose.
*Countly Ltd. is a TISAX® Participant. TISAX® is a registered trademark of and governed by ENX Association. Learn more at enx.com/tisax