Operations

Dataset Analysis Reporter

Profile data quality, identify patterns, and deliver prioritized recommendations

Overview Capabilites Agent Workflow Example prompt

Overview

The Dataset Analysis Reporter profiles data quality, identifies patterns, and delivers prioritized recommendations—enabling data teams to quickly assess new datasets and make informed decisions about data readiness, cleaning needs, and analytical potential. When working with new data sources, analysts spend significant time on exploratory analysis: checking data types, finding missing values, identifying outliers, understanding distributions, and assessing overall quality. This agent automates this critical first step by performing comprehensive dataset profiling, generating statistical summaries, visualizing distributions, flagging quality issues, and recommending specific actions to prepare data for analysis. Whether onboarding new data sources, auditing existing datasets, or preparing data for modeling, it accelerates the path from raw data to actionable insights.

Capabilities

Profile dataset structure including columns, data types, and statistical summaries
Identify data quality issues including missing values, outliers, and inconsistencies
Analyze distributions and relationships between variables with visualizations
Flag potential problems and recommend specific data cleaning actions
Generate comprehensive reports prioritizing issues by impact and effort to resolve

Agent Workflow

Input: User uploads dataset (CSV, Excel, database table) for analysis
Data Profiling: Agent examines structure, types, distributions, and basic statistics
Quality Assessment: Identifies missing values, outliers, duplicates, and inconsistencies
Pattern Analysis: Explores relationships, correlations, and notable patterns in data
Recommendation Generation: Prioritizes issues and suggests specific remediation actions
Output: Delivers comprehensive analysis report with quality score and action plan

Example prompt

"Analyze the attached customer transaction dataset (50,000 rows, 15 columns) that we're planning to use for churn prediction modeling. Provide a comprehensive data quality and readiness report including: 1) Dataset overview - number of records, columns, date range covered, and overall completeness score, 2) Column-by-column profile - data type, missing value percentage, unique value count, and distribution summary for each field, 3) Data quality issues - identify missing values, outliers (using IQR method), duplicate records, inconsistent formatting, and any columns with suspicious patterns, 4) Relationships and correlations - flag any strong correlations between variables and potential multicollinearity issues, 5) Readiness assessment - is this dataset ready for modeling, or what cleaning is required? Prioritize the top 5 data quality issues by impact on model performance and provide specific recommendations for addressing each (e.g., 'Impute missing values in Age column using median' or 'Remove 47 duplicate records based on Customer_ID'). Include visualizations for the 3 most important variables. Format as a 2-page executive report suitable for presenting to our data science team lead."