Surveys & voice of the customer

#Ask while the memory is fresh

The customer knows best how a contact went, and she knows it right afterwards. A week later she remembers the frustration, not the details. That is why Univerx asks straight after the contact, on the channel where it took place.

After a call, a short voice dialogue in the IVR takes over. After a chat the question appears in the window, after an email as a link, after an appointment by text message. Two or three questions are usually enough.

#Three metrics, three questions

No metric is right or wrong on its own. What matters is that it fits the question you want to answer. More in the article Measuring customer experience without vanity metrics.

#What the free text says

The most valuable answers are often the sentences in the comment box. Univerx sorts them by topic and recognises the sentiment, much as conversation analytics does for calls and chats.

That reveals, for instance, that many low-rated contacts revolve around the same delivery delay. Or that a new cancellation rule is applied correctly but feels cumbersome.

#Results where they make a difference

Every answer is linked to the contact it refers to. Results can therefore be analysed per queue, per topic and per step of a customer journey. Journey analytics shows where satisfaction drops.

For quality management this means conversations with low scores can be reviewed in a targeted way. Conversations with very high scores show what works and make good examples in coaching.

#Closing the loop

1. Rate. The customer gives a low score and writes a short comment.
2. Trigger. A rule creates a callback task with the score, the comment and the contact history.
3. Route. Task routing passes it to the team responsible.
4. Follow up. An agent calls back, resolves the issue and records the outcome.