See How AI Arms Service Organizations with More Case Insights to Empower Support Engineers
God is in the detail. A lot of useful information can be extracted from the comment streams. For instance, if your turnaround time (TAT) is consistently going up, customer frustration might be on the rise as well. Escalation Predictor uses machine learning to closely monitor and identify such deviations, and raise an alert so that corrective action can be taken.
The search-powered application is also adept at evaluating queries and gauging the underlying customer emotion. Towards this end, natural language understanding (NLU) tags in to glean sentiment packaged with the case description or comments. This helps to prioritize cases with a high emotional quotient and those low on emotions but high dissatisfaction.
When a case is not assigned to the right agent, unnecessary transfers follow. This wastes a lot of time and further aggravates a disgruntled customer. Escalation Prediction picks out words and phrases and rummages through historical cases to find matches. Then, intent extraction kicks in and assigns tags the case to the right category. This makes it easier to ensure the best agent is on it.
Median case resolution time is the median time to successfully resolve all the cases created by a customer without any escalation. This time window becomes the customer’s base expectation for a query to be answered and closed. Escalation Predictor keeps a tab on this metric and if a case is taking longer than usual, it might be en route escalation.