CSAT surveys are typically triggered immediately after an interaction ends, while the experience is still fresh, with a short single question such as how satisfied the customer was with the resolution. The scores are then averaged over a period or segmented by channel, agent, or use case. Because it is a single subjective rating, CSAT is usually tracked alongside other metrics like Customer Effort Score and automated resolution rate to build a fuller picture of interaction quality.
In AI-driven customer service, comparing CSAT for AI-resolved contacts against human-resolved contacts shows whether automation is maintaining or eroding satisfaction as more volume shifts to AI. This comparison matters especially in regulated sectors like banking and healthcare in the MENA region, where trust in a new AI concierge channel needs to be earned, and where satisfaction also needs to hold up across Arabic dialects and WhatsApp interactions, not just in English.
In Eshal: Eshal collects CSAT automatically at the close of every conversation, across WhatsApp, web, and voice channels, and tracks scores separately for AI-resolved and human-escalated contacts. This lets businesses monitor whether satisfaction holds up as automated resolution rate increases, and surface any dialect, channel, or use case where AI-resolved CSAT is falling behind human-handled interactions.