STRAX Reach · Alerts and push
Escalate client issues to PagerDuty
Some client messages mean something is broken. Reach can raise a PagerDuty incident so the on-call engineer hears about it straight away.
- Events API incidents
- Per-service routing
- Logged on the record
From a client report to the on-call rota
Each escalation target is a Reach contact whose PagerDuty identity is a service routing key. Queue a message to that contact and Reach triggers an incident on the service, with the subject and text as its summary.
The deduplication key comes from the Reach message, so a retried send never pages twice. Severity and source are settings per environment, which keeps staging tests off the live rota.
Client conversations are often the first sign that something is failing. Escalating from Reach puts the client's own words in front of the engineer on call, with a reference back to the conversation for context.
Incidents raised this way carry the originating message text, the client reference and a link path your engineers recognise. After the fix, the agent who escalated can close the loop with affected clients from the same conversation.
Where it helps
Payment failures reported by clients
When several clients report failed card payments, escalate to the payments on-call engineer.
Suspected fraud waves
Page the fraud team when agents notice a cluster of scam reports arriving across several channels within one afternoon.
Channel outages
Raise an incident when a messaging provider stops accepting messages.
Premium client escalations
When a key merchant reports a settlement delay, page the relationship manager on call instead of waiting for office hours.
What you need
-
A PagerDuty Events API integration
On the default escalation service, with its routing key.
-
Escalation contacts
One Reach contact per service you want to page.
-
Severity and source
Set per environment in the Reach settings.
-
Reach Core
The records escalations are logged in.
Good to know
This channel only triggers incidents. Acknowledgements and resolutions stay in PagerDuty.
Questions
Who decides when to escalate?
Your workflows, your agents or your systems, by queuing a message to the escalation contact.
Can we page different teams?
Yes, create one contact per service with that service's routing key.
What is still missing?
Acknowledge and resolve events back into the conversation, and an Opsgenie variant.
Can escalations be automatic?
Yes, a workflow can queue an escalation when a rule matches, for example several payment-failure reports within a few minutes.
Does PagerDuty see client personal data?
Only what you put in the escalation message. Keep the summary to the problem and a reference, and let the engineer open the full conversation in Reach.
Read more
-
The whole Reach family →
Reach Core and every channel package, with what each one needs.
Add PagerDuty to Reach
Tell us which channels your clients use and we will plan the rollout with your team.
Premium Requires STRAX Reach - Core, which is licensed.
About this listing
| Version | 1.0.0 |
| Author | STRAX |
| Category | Communication |
| Published | 2026-09-24 |
What's inside
| Credential | 1 |
| Global variable | 2 |
| Workflow | 1 |
What happens on install
After the import, and only with your consent, the installer runs:
- Reach - Register Channel : Registers the PagerDuty channel with Reach Core, disabled until its credentials are entered
Free listing
Install it from Feature Marketplace inside STRAX
It builds on STRAX Reach - Core, a premium package. Ask us for a quote to license it first.
Open Feature Marketplace
In your STRAX installation, open Feature Marketplace from the navigation.
Find "STRAX Reach - PagerDuty"
Search or browse to this listing and review the contents disclosure; every package shows what it contains before anything runs.
Install
Choose Install, or "Customise with AI" to adapt the package to your environment first. No licence is needed for free listings.
PagerDuty is a trademark of its owner, named here only to identify the service this package connects to.
See it on your own systems.
A demo takes an hour, on your data landscape rather than ours.