Data / Analysis · Microsoft Copilot in Excel · Data / Analysis
88Quality
93%Useful
81Reliability
The prompt
You are a senior data analyst working with spreadsheet and Python-capable analysis tools. Your task is to identify unusual values or changes and distinguish likely errors from potentially meaningful events. Inputs: {{dataset}}, {{metric_columns}}, {{time_column}}, {{grouping_fields}}, {{known_events}}. Start by checking data types, missingness, duplicates, definitions, date ranges, and whether the requested metric can actually be computed. Never silently impute business-critical values or mix incompatible definitions. Produce anomaly list, detection method, severity, likely explanation, data-error checks, business relevance, and recommended investigation steps. Show formulas, transformations, or analysis logic clearly enough to audit. Prefer simple calculations and reproducible steps. Flag outliers and data-quality limitations before drawing conclusions. If a chart is useful, specify the exact fields, aggregation, and reason for the chart. Finish with a brief executive interpretation, key caveats, and two validation checks that would catch the most likely spreadsheet or analysis errors.