Ask customers how the conversation went — in the chat or by email after — and get scores you can actually audit.
A CSAT survey tool asks customers to rate a support conversation just after the conversation ends. inrelay sends CSAT surveys on the channels a team enables — inline in the chat widget for live chat, by hosted form for email and tickets — and collects scores, comments and delivery outcomes in one place.
What it prevents is flying blind. Reply-time stats tell you a message went out; they say nothing about whether the answer helped. A fast queue can still be a wrong queue, and without a survey the first sign is a customer who simply never comes back.
CSAT is configured once, in desktop settings, and then runs itself at the end of conversations on the channels you choose.
CSAT formula. CSAT (customer satisfaction score) equals the number of satisfied responses divided by the total number of survey responses, multiplied by 100. A response counts as satisfied when the rating falls in the top tiers of the scale — a 4 or 5 on a five-point scale. A CSAT of 80 means eight of ten respondents rated the conversation positively.
Two habits keep the number honest. Read it per channel and per period rather than as one lifetime score, because a rough week on live chat hides easily inside a good quarter of email. And treat small samples gently — ten responses move the score ten points at a time, so the trend matters more than any single week’s figure.
The failure mode a CSAT survey tool removes is confident ignorance. Small teams optimise what they can see, and without surveys the only visible numbers are volume and speed — so that is what improves, even when answers are going out fast and wrong. A recurring low score on one channel, or one kind of question, is the earliest signal you will get that a policy, a template or a missing help article is quietly costing you customers.
The comments matter as much as the score. A 2-out-of-5 with “took three messages to get a returns link” attached is not a grade — it is the next knowledge base article, already specified.
Honesty, mechanically enforced. Because each survey snapshots its settings, you can change the question or the scale without corrupting history — responses are never re-read against a question they were not asked. And the delivery workers record truthful outcomes: a survey email is marked queued, sent, skipped or failed, and provider acceptance is never dressed up as final delivery without evidence. When the score is good, you can trust it, because the misses are counted too.
Turn surveys on and pick the channels that should ask.
Choose the scale type and order, write your question, and decide whether comments are optional or required.
Write the confirmation copy and enable the email follow-up for conversations that end outside the widget.
Scores and comments land alongside the conversations they rate. The help docs cover every setting.
CSAT is one piece of the inrelay feature family — these are its closest neighbours.
Low scores point at missing articles. Write them where you reply.
See the knowledge base → TRIAGEGet each conversation to the right teammate before the survey ever has to say it went to the wrong one.
See triage → MOBILECatch a low score on your phone and make it right from the same queue.
See mobile →Free to start with your own inbox — every feature, no card, never per seat.
Give your customers faster answers and your team their evenings back.
Desktop and mobile. Free for solo.