Issue detection in Data Manager helps identify issues in your data that can lead to poor data quality. At a high-level, Data Manager surfaces two (2) types of issues:
- Outliers
- Gaps
Outliers
Outlier detection in Data Manager helps identify data points that deviate significantly from expected patterns. This process ensures data integrity and accuracy across energy and water meter reading values.
Outlier Types
Data Manager currently supports the following types of outliers:
-
Usage Outliers
- Usage value may be incorrectly low
- Usage value may be incorrectly high
-
Cost Outliers
- Cost value may be incorrectly low
- Cost value may be incorrectly high
-
Cost vs Usage Ratio Outliers
- Cost per usage may be incorrectly low
- Cost per usage may be incorrectly high
- Usage per cost may be incorrectly low
- Usage per cost may be incorrectly high
-
Duplicate Outliers
- Reading may be incorrectly duplicated in another meter
- Reading may be incorrectly duplicated within its own meter
-
Date Outliers
- Period for reading is unusually long
- End date for reading is in the future
Gaps
Data Manager currently supports the following gap issues:
- Gaps (or True Gaps)
- End Gaps
Gaps
Gaps, or “True Gaps”, are when there is missing data between two or more meter readings. The “Action Needed” column in the Gaps table under Check will suggest the best option to resolve the issue.
End Gaps
End Gaps occur when a meter’s last reading is over 90 days before today’s date. Depending on the source of data for the meter, data may need to be entered by hand, uploaded through Bill Upload, or an account-level or provider-level issue in Connect needs to be resolved.
Both of these types of Gaps can be resolved through the Core application. Please reference the following guide for more details on this workflow: Data Integrity Monitoring at the Site-Level.