I am configuring real-time adherence alerts for our operation of 800 agents. Currently, if an agent is out of adherence, the alert triggers almost immediately at exactly 09:00:15 JST when their shift started at 09:00:00 JST.
This is creating too much noise for our supervisors. Is there a method to configure the threshold so that the notification only fires if the agent is out of adherence for more than 300 seconds?
From a workforce management perspective, this is a very common issue when rolling out RTA (Real-Time Adherence).
You don’t configure this in the alerts menu directly. You need to go into your WFM Management Unit settings. Under the ‘Adherence’ tab, you can define the ‘Severe’ versus ‘Warning’ thresholds in minutes. If you set the threshold to 5 minutes, the system will allow the agent a 5-minute grace period before it flags them as severely out of adherence and triggers the supervisor notification.
When configuring these alerts, I must remind you of the data privacy implications.
If these alerts are being routed outside of the Genesys Cloud platform (e.g., via a webhook to a third-party messaging system like Slack or Teams), please ensure the notification payload does not log the agent’s full name or employee ID in a non-compliant environment. In accordance with GDPR data minimization principles, the alert should only contain the internal GC User ID.
I actually tried to write a Python script to listen to the adherence topic to build a custom dashboard, but it crashed.
# My code
event = get_adherence_event()
print(event.body['user'])
It threw a KeyError. I’m sorry for the noob question, but does the event payload use a different key for the agent ID when the threshold is crossed?