Monitoring
Know whether your automations actually worked.
Awish continuously monitors workflow health and expected business outcomes, detects silent failures and anomalies, explains what went wrong, and helps your team understand what needs attention.
A workflow that ran is not the same as work that got done. Awish checks the result each workflow was meant to produce, watches how every workflow behaves against its own normal, and when something breaks it shows the likely cause, every workflow it affects, and what to do next. It also measures the hours your automations give back.
Outcome verification and silent-failure detection on every plan · Advanced Monitoring on Business
“Success” should mean the work got done.
After every run, Awish checks the result the workflow was meant to produce — in the app where that result should be. A technical success response is not enough.
Two runs of the same workflow can both return “success”. In one, the invoice row is in the books sheet and the total matches. In the other, nothing was written. A run log shows them as identical; outcome verification is what tells them apart.
The check is part of the workflow from the start. Every template names what done means for its job, and a workflow you describe states it in your own sentence.
Catch the runs that look fine and aren’t.
A silent failure reports success while the expected result is missing, incomplete or wrong. These are the ones nobody hears about until a customer or a month-end report finds them.
A record created with a required field empty
The CRM accepted the contact, so the step succeeded. The owner field it needed for routing was blank.
An email sent to the wrong list
The send completed without an error. The expected recipients were your trial users; it went to a different segment.
Half a batch written
The run finished, but 23 of 140 rows never reached the sheet. Nothing failed loudly; the outcome is incomplete.
A trigger that fired with nothing downstream
The workflow started on schedule and ended in seconds. The report it exists to produce was never assembled.
Know when a workflow stops behaving normally.
Each workflow is measured against its own pattern — how often it runs, how long it takes, how often it fails — so a change is caught even when no single run breaks.
An unexpected drop in volume, runs that suddenly take four times longer, a failure rate creeping up: each is flagged in plain language, with when it started.
Anomaly detection, including volume, latency and trigger silence, is part of Advanced Monitoring on the Business and Enterprise plans.
See why it happened, and what to do next.
Instead of sending you into raw logs, Awish names the most likely cause, shows which workflows it affects, and recommends the next step.
“14 executions failed” becomes “HubSpot authentication expired at 09:42 — 3 workflows affected”. The recommendation follows: reconnect HubSpot and replay the affected executions.
The recommendation is yours to act on. Awish does not reconnect accounts or replay runs by itself; it tells you exactly what to press, and you decide.
See every workflow’s health, and what they depend on.
One view across the operation. When a shared connection breaks, see every workflow hanging off it — one thing to fix instead of a dozen separate alerts.
Cross-workflow health
Every workflow in one view, filterable by team, with what needs attention first.
Workflow dependency alerts
When a connection or an upstream workflow breaks, see every workflow that depends on it — one cause, not a dozen alerts.
Trigger silence detection
A workflow that normally runs forty times a day and has not run since noon is a problem, even though nothing failed.
Custom alert rules
Set the conditions that matter to your team and choose where the alert goes.
See what automation is actually giving back.
Awish measures the hours your workflows save and the outcomes they verified, by workflow and by team, so the value of automation is something you can show rather than assume.
Every plan shows hours saved. Pro breaks it down by workflow; Business adds advanced ROI across teams.
Govern automation like any other production system.
For organisations where automation touches customers, money and regulated data: the record, the change history and the streams your security team expects.
Audit logs
Who did what, in which workflow, and when — kept as a record your review can read.
Workflow change audit
Every change to a workflow, with the version before and after, so a new behaviour can be traced to the edit that caused it.
Log streaming and SIEM
Stream workflow and monitoring events into the SIEM your security team already watches.
Custom retention
Keep monitoring history for 365 days or for the period your policy requires.
Monitoring on every plan.
Normal Monitoring comes with every plan. Advanced Monitoring starts on Business, and governance on Enterprise.
| Capability | Free | Pro | Business | Enterprise |
|---|---|---|---|---|
| Level | Normal | Normal | Advanced | Advanced + Governance |
| History | 7 days | 30 days | 90 days | 365+ days or custom |
| Outcome verification | Included | Included | Included | Included |
| Silent-failure detection | Included | Included | Included | Included |
| Live status and failure alerts | Included | Included | Included | Included |
| Connection health | Included | Included | Included | Included |
| Execution timeline | Included | Included | Included | Included |
| Hours saved | Included | Included | Included | Included |
| Search, filters and manual replay | Not included | Included | Included | Included |
| Anomaly detection (volume, latency, trigger silence) | Not included | Not included | Included | Included |
| Root-cause explanation and recovery recommendation | Not included | Not included | Included | Included |
| Cross-workflow health and dependency alerts | Not included | Not included | Included | Included |
| Custom alert rules | Not included | Not included | Included | Included |
| Audit logs, change audit, SIEM and log streaming | Not included | Not included | Not included | Included |
Questions about monitoring
How does Awish know what the expected outcome is?
It comes from the workflow itself. A template names what “done” means for its job — a row in the sheet, a logged activity in the CRM, a message delivered — and a workflow you describe states it in the request. After each run, Awish checks for that result in the app where it should be.
Does Awish recover or replay failed runs by itself?
No. It explains the most likely cause, shows which workflows are affected and recommends the next step, such as reconnecting an app and replaying the affected runs. You decide, and nothing is replayed without you.
What is the difference between a failure and a silent failure?
A failure is a run that stops with an error. A silent failure is a run that reports success while the expected result is missing, incomplete or wrong. Most tools only catch the first; Awish checks for both on every plan.
Which plan includes which monitoring?
Outcome verification, silent-failure detection, live status, failure alerts, connection health, the execution timeline and hours saved are on every plan. Pro adds search, filters and manual replay. Business adds anomaly detection, root-cause explanations, recommendations, cross-workflow health and custom alerts. Enterprise adds audit logs, change audit, SIEM and custom retention.
How long is monitoring history kept?
7 days on Free, 30 days on Pro, 90 days on Business, and 365 days or a custom retention period on Enterprise.
Where do alerts go?
To the channel the workflow already reports into — WhatsApp, Slack or Telegram — and to the monitoring view in Awish. Custom alert rules on Business let you choose per rule.
Start with a template. Know it keeps working.
Outcome verification and silent-failure detection are on from the first run, on every plan.
Free plan · No credit card required