| Insufficient filled fields for training |
If the case fields that you selected as inputs for training don’t contain
enough information for a service rep to classify the case, Einstein probably can’t
classify it. |
- Improve your case submission process to encourage customers and service reps
to include clear and consistent information in specific fields, especially the
fields used to train your model. Consider adding a list of common case issues to
select from or a minimum character count for important case fields.
|
| Duplicate cases |
Einstein assumes that your closed cases are filled out correctly. If Case 1 and
Case 2 are duplicates, any differences between their field values send Einstein
mixed messages about correct data patterns. |
|
| Incorrectly completed fields |
Cases whose fields aren’t filled out correctly make it hard for Einstein to
know what’s correct. |
- Label high-quality cases, and use them as your example case set.
|
| Overlapping field values |
When a picklist field’s values are too similar or generic, it’s hard for
service reps—and therefore for Einstein—to know which value is correct. For example,
values like “General” or “Other” can be overused, or overlapping values like
“Mobile” versus “Mobile Application” can dilute your data. |
- Identify the field’s potentially problematic values, and then exclude cases
with those values from your model.
- Simplify your case data structure to remove redundancies.
|
| Too many field values |
If a picklist field included in your model has more than 100 values, it hurts
the model’s accuracy. Here’s why:
- Einstein likely doesn’t have enough example cases to understand when each
value is appropriate.
- Because selecting from such a large list of field values requires time and expertise, service
reps are more likely to make wrong or inconsistent selections.
|
- Label high-quality cases, and use them as your example case set.
- Simplify your case data structure.
|