It's increasingly common for organisations to ask their customers how satisfactory their latest interaction was, often by asking them to participate in an automated survey at the end of a phone call, or by inviting them to complete an online feedback form after a physical interaction such as an airline flight or a virtual one such as using the company's web site.
But by that time, it's too late to head off any negative outcomes - the interaction is complete, so all you can do is try to stop problems occurring again or provide the customer with some sort of compensation to make up for it.
The feature has been in beta test for five months ago, and in that time has analysed more than 1.82 million customer interactions.
Satisfaction Prediction uses hundreds of signals - including text description, number of replies and total wait time - to calculate the likelihood of a positive satisfaction rating. This score can be used to help prioritise workflows, drive business rules, or trigger downstream integrations based on data-driven analysis, the company said.
Pinterest is one of the beta users. "We've been using Satisfaction Prediction to detect conversations with customers that are most at risk of a poor customer experience," said reactive support lead Maggie Armato.
"Previously, we had a manual process where a dedicated team member would look through our tickets and proactively flag experiences identified as potentially negative. Now, we use the prediction score to accurately and automatically identify these types of tickets so our agents can focus on higher value areas."
Zendesk senior vice president of product development Adrian McDermott said "By having an early warning system that identifies high-risk interactions [across multiple channels], companies can course-correct negative experiences before they ever even happen."