Why does this setting break WFM schedule adherence for AI bot handoffs? We are seeing a significant discrepancy in our schedule_adherence metrics when agents are handling conversations that originated with our Genesys AI bot.
The environment is running v2024-02 in the us-east-1 region. Our WFM setup uses WFM SDK v3.1.4. The issue manifests specifically when the bot escalates to a human agent via a Transfer Queue action in the Architect flow. The transfer includes a custom data payload with the conversationId and botSummary.
When the agent accepts the transfer, the WFM system marks the time as Available rather than Talking or After Call Work until the agent manually logs the interaction type. This creates a false positive in our adherence reports, showing agents as idle while they are actually reviewing the bot summary.
We have verified that the skill assignment matches the queue skill, and the wrap-up code is correctly mapped in the WFM configuration. However, the activity_code does not automatically transition. The error log in the WFM dashboard shows Activity transition failed: source activity not compatible with target.
This is causing major issues for our shift swaps and performance reviews. Agents are being flagged for non-adherence because the system thinks they are not working. We need the activity to automatically switch to Talking upon transfer acceptance.
Has anyone configured a Data Action or a specific WFM rule to force this activity transition based on the transferType? We are considering a webhook to update the activity state, but that seems like a heavy lift for what should be a standard behavior. Any insights on the correct configuration in the WFM Settings under Activity Codes would be appreciated. We want to avoid manual overrides for every bot-assisted call.