The default Freshdesk CSAT survey ships with a 1-7 rating scale and an optional comment. Response rates hover at 8-10%. Teams pulling 25%+ adjusted four things: timing, scale, length, and channel.
Timing: not on close, but at perceived resolution
Sending the survey on ticket-close timestamp catches admins, not customers. Send on first response after status reaches “Resolved” — the customer’s reply that confirms it actually worked.
Trigger:
status changed to Resolved
AND requester replies again within 7 days
AND ticket has no negative sentiment in last 3 messages
Action:
send CSAT survey (delay 1 hour)Scale: thumbs up/down beats 1-7
Three-option scales (Good / Neutral / Bad) double response rates over 1-7 in our sampling. The 1-7 scale produces neutral ambiguity. Configure via Admin → Customer Satisfaction.
Length: one question, optional comment
Multi-question surveys lose 60% of respondents at question 2. Use one question on the email, link to a longer survey only when feedback is negative. Negative feedback rooms tolerate three follow-ups; positive responses do not.
Channel: send via the channel the customer used
Customers who used chat want a chat-window CSAT. Email customers want email. Configure separate CSAT surveys per channel and route via dispatcher.
Use the negative-feedback workflow
Any “Bad” rating must trigger an immediate task on the original agent’s manager, not the agent. The manager owns the rescue conversation.
// Workflow Automator
{
"trigger": "csat_rating == bad",
"action": "create_task",
"assigned_to": "agent.manager",
"title": "CSAT recovery: ticket #${ticket.id}",
"due_in_hours": 24
}Survey fatigue
Cap surveys at one per customer per 14 days. Repeated surveys to the same person tank response rates and skew sentiment toward chronically irritated voices.
What to do this week
Move CSAT trigger to confirmed-resolution, switch to the three-option scale, ship one-question email surveys, and add the manager-owned negative-feedback task.
