Why does this setting prevent the predictive routing engine from picking up the assigned skills?
We are currently migrating our Zendesk support queue to Genesys Cloud. The goal is to replicate Zendesk’s tag-based assignment logic using Genesys skills. In our Architect flow, we successfully parse the Zendesk ticket tags and map them to specific Genesys skills using a Set Skills block. The flow logs show the skills are correctly assigned to the interaction before it enters the queue.
However, when we enable Predictive Routing for the skill group, the interactions sit in the queue with a status of ‘Waiting’ and never route to an agent. If we switch the skill group to ‘Longest Idle Agent’, the routing works perfectly.
Predictive Routing Error: Insufficient Historical Data
The Genesys Cloud Analytics dashboard shows a warning about insufficient historical data for the model to train, despite us having pushed 500 past interactions via the API for training purposes. The API endpoint /api/v2/analytics/interactions/summary returns the data, but the predictive model seems to ignore it.
Is there a specific delay or configuration step required to sync this historical data with the predictive routing model? In Zendesk, we just updated the automation rules and it worked immediately. We are using the standard Genesys Cloud API v2 for data ingestion. Any guidance on bridging this gap would be appreciated.