Methods and systems for automatically detecting docketing data anomalies are provided. The method includes storing in a docketing system docketing information for a plurality of matters, each of the plurality of matters including a plurality of activities and a plurality of documents. Retrieving a first document from the plurality of documents associated with a first matter of the plurality of matters. Determining a document type of the first document. Extracting one or more features from the first document and the plurality of activities associated with the first matter. Training a machine learning model, based on the extracted features and the document type of the first document, to determine one or more expected docketing activities for a new document determined to match the document type.