What this service solves
This plan gives the owners of live automations a clear way to notice and investigate failures in the flows covered by the engagement. We agree what healthy operation looks like, which signals to watch, and where issues go when they exceed the available repair time or need approval from a system owner.
What you get
Monthly health summary
A review of monitored flows, issues investigated, agreed fixes completed, and unresolved problems for the owner to follow up.
Exception and escalation path
A defined route for alerts and findings, including who receives the details when a fix exceeds the plan or requires an internal decision.
What’s included
Monitor agreed flows
Watch the agreed flows using the available run signals and alert channels for the system.
Investigate alerts and perform bounded fixes
Investigate alerts against run history and perform agreed fixes within the plan's monthly allowance.
Report health and unresolved issues
Summarize observed health and unresolved issues so the owner can decide which follow-up needs its own work.
How implementation works
Onboard the covered flows
Confirm the flow inventory, normal execution pattern, and escalation contact. Agree the service windows and alert sources.
Monitor and triage
Review alerts in the agreed window, use available logs to assess impact, and record fixes or escalations.
Review the month
Review the month with the owner, including recurring incidents and any issue that needs a broader repair or change.
Technical considerations
- Detecting a missed run
- A missing-run check needs an expected schedule and an available record of execution or output. For an event-driven flow with variable traffic, agree which observable condition can support a useful alert.
Systems, inputs & prerequisites
A baseline flow inventory, authorized read access, run history, and an escalation contact.
- Provide the flow inventory, authorized read access, and recent run history. Identify the expected cadence and output of each flow.
- Name an escalation contact and agree the conditions and authorization for any change to a live workflow.
Starting scope & pricing
Base scope
One system, up to five workflows, and up to two hours of agreed fixes per month.
Illustrative workflow
Read signals
Check the agreed signal for the expected daily run.
Triage alert
Review the run history and report whether an input or a workflow step failed.
Review month
Summarize recurrence and open items for the report owner.
Validation & handover
- Confirm the covered inventory, the signals available for each flow, the service window, and the escalation contact.
- Use available history or an agreed test example to check that an alert can be tied to a flow and that the recipient knows which details to escalate.
- Review the monthly summary with the owner. Distinguish completed agreed fixes, issues still open, and work that needs a separate decision.
Limitations & possible extras
Limitations
Service windows and escalation terms are agreed for the engagement. Critical out-of-hours coverage, additional repairs, and platform fees require separate scope.
Service windows and escalation terms are set in the engagement. Critical out-of-hours coverage requires separate scope.
Possible extras
New builds, extra repair hours, expanded monitoring, or platform fees.
Questions about this service
What if an incident needs more repair time than the plan includes?
The plan includes up to two hours of agreed fixes per month. Record the evidence and escalate a larger issue so the owner can decide on separate repair scope or a revised plan.
When will alerts be investigated?
Service windows and escalation terms are set for the engagement. Identify flows that affect critical operations during onboarding so the owner can decide whether separate out-of-hours coverage is needed.
Can monitoring tell whether the destination received every expected record?
That depends on the agreed signals and access. Run history can show execution status. Confirming destination delivery requires an available output or reconciliation signal and a defined expectation.
Can we use monitoring time for a new automation?
The monitoring plan covers the agreed flows and bounded fixes to their operation. A new build requires separate scope and its own acceptance checks.