Save Agents Time and Improve Accuracy and Completion with Einstein Case Classification
How: Add the fields that you want to predict for your agents. Then build the predictive model. Einstein Case Classification analyzes the fields in your closed-case data to determine recommendations for these fields in new cases.
After you build the model, review the recommendations. For each field, you can choose to turn on Select Best Recommendation. This setting controls whether your agents must select the correct recommended value for the field or whether the best recommendation (if available) is selected for them.
Activate Einstein Case Classification to display recommendations to your agents.
In the Service Console, your agents see an alert that Einstein recommendations are available. To view the recommendations, agents click the alert.
- A minimum of 1000 cases but 10,000 or more is ideal for best performance.
- A minimum of 100 closed cases that used each field and value that you want to predict for your agents.
- Fewer than 100 values per field. If you have more than 100 possible values per field, model accuracy is reduced.