How to set up CSAT surveys on resolution#

Audience: admin — survey policy is an org-wide decision, and the mail transport it depends on is an Email-settings decision.

Prerequisites#

• Admin access to Settings → Surveys.
• An outbound mail transport configured under Settings → Email. With transport none, nothing below sends — surveys are never queued and the CSAT sections on scorecards stay honestly empty.

One sentence of doctrine#

"One survey per resolved ticket, at most one per client per interval, never inside quiet hours, and every link works exactly once."

1. Turn surveys on#

In Settings → Surveys, check Send a satisfaction survey when a ticket resolves and save. From the next resolution on, the worker queues a survey for the ticket's client.

If mail is not configured, the page says so in plain text and no surveys are queued. Turn the transport on first; the toggle does not pretend.

2. Shape the policy#

• Minimum interval per client (default 168 hours): a client who already received a survey inside the window is skipped. The skip is recorded (min_interval) so you can see the policy working rather than guessing.
• Response link validity (default 14 days): how long the emailed link works. Links are single-use regardless. A second open gets an honest "invalid, used, or expired" page, never a second ballot.
• Quiet hours (optional HH:MM window in the org timezone): surveys arriving inside the window are deferred to its end, then send. Leave both bounds empty to send as soon as a survey is queued.

3. Where responses land#

• Scorecards — each tech's card gains a CSAT block (responses, average score, low scores at ≤ 2) once the org has real responses. Before that the section stays "not yet available" instead of showing a fake zero.
• Team scorecards → Client satisfaction — the per-client rollup table for owners and admins.
• Automations — the CSAT response recorded trigger with the CSAT score below N condition fires on low scores. Use it to notify a manager or add an internal note while the ticket is still fresh.

4. The response experience#

Clients open the tokenized link from the resolution email and sign in to the portal — the sign-in is the attribution, so responses are never anonymous. They pick 1–5 with an optional comment and see a thank-you state. The link in the email and the in-app route are the same single-use token.

Troubleshooting#

• No surveys arriving — check the transport (Email settings), the send-on-resolve toggle, and whether the client is inside the minimum interval.
• A link does not work — it was already used or its validity has run out; there is no override, by design.