Connected Sheets now features anomaly detection for BigQuery data, enabling users to easily find outliers in time-series data. This feature allows identification of irregularities without the need for manual model training or complex SQL queries. The system uses BigQuery ML and TimesFM for "zero-shot" analysis, offering immediate AI insights. Anomaly detection complements the existing forecasting feature for a comprehensive predictive toolkit. Users can configure the analysis through a user-friendly panel without writing any SQL code. Results are clearly formatted with columns indicating anomalies and boundaries for easy interpretation. Users can customize the analysis with time periods, probability thresholds, and data filtering options. Like other Connected Sheets objects, anomaly detection can automatically refresh to ensure insights remain up-to-date. There are no specific admin controls needed for this feature. Access requires BigQuery project permissions with enabled billing. The feature is available to all Google Workspace customers and personal Google account users.