AUDIPY is a versatile tool for analysing, processing and visualising structured data sets: search, filter and evaluate large volumes of data efficiently, without complicated workflows or additional tools. Below is the path through AUDIPY in seven steps, followed by the features in detail.
1Importread in data2Exploreget an overview3Prepareget data ready4Ask AMYask in plain language5Analysespot patterns6Auditrecord results7Automaterepeat workflows
Import
Basics & Import
Supported File Formats
- Parquet: Highly optimised column format (standard format)
- Excel: XLSX, XLSB, XLS (all versions)
- CSV / TXT: Automatic character set and delimiter detection
- XML: Structured data imports
- JSON: Support for nested structures
- DBF: DBase files
- ORC: Apache ORC format
- Power BI: Direct extraction from PBIX files
Database Connectors
- Microsoft SQL Server: With Windows authentication
- MySQL & MariaDB: Full compatibility
- PostgreSQL: Native support
- SQLite: Embedded databases
- Microsoft Access: MDB/ACCDB via ODBC
- ODBC: Connection to any data source
- Encryption: SSL/TLS for secure connections
Import Wizards
- Data Wizard: CSV/TXT with preview and delimiter detection
- Headers and footers: skip them, choose the character set (UTF-8, ISO, etc.)
- Excel Wizard: Smart Merge and automatic header detection
- Worksheets: merge multiple sheets
- Report Wizard: Extraction from PDF files
- Database Wizard: with SQL Query Builder
- Connection test: error diagnostics integrated
Special Imports
- GoBD import: compliant XML import (Index.xml with references)
- SAP DART files: Automatic structure import
- XRechnung: Extraction and visualisation
- Power BI integration: Access to DAX & M queries
- Batch import: Read multiple files simultaneously
- Folder import: Recursive processing of entire directories
Explorer & Project Management
- File explorer: hierarchical with tree view
- Project management: projects and cases with quick switching
- Data lineage: Tracking of derivations
- Folder structures: create and manage your own folders
- Tab system: multiple files open simultaneously
- Drag & drop: Import files directly
Explore
Exploration & Interaction
Column Formatting
- Date formats: automatic detection and conversion
- Timestamps: Unix timestamps and Excel serial numbers
- Number formats: Integers, decimals, EU/US separators
- Display: Thousands separators, percentage and currency
- Other types: Formatting for text, Boolean and hyperlinks
- Precision: Decimal places and display precision configurable
Column and Row Actions
- Manage columns: duplicate, rename, delete or hide
- Drag & drop: Reordering with automatic saving
- Column widths: adjust individually and save permanently
- Sorting: by one or more columns
- Copy cell contents: individually or in blocks
- Editable columns: direct editing in the table
- Edit rows: add or delete in edit mode
Advanced Filtering
- Simple filters: Equal, Not equal, Contains, Starts with, Ends with
- Range filters: Greater than, Less than, Between
- Date filters: Year, month, week, weekday or time periods
- Fuzzy filter: Similar values with configurable accuracy
- Multi-value filter: Select multiple values (In / Not In)
- AND/OR logic: Combine multiple filter conditions
- Filter bookmarks: Save, name and reuse
Column Statistics
- Key figures: Sum, mean, median, minimum and maximum
- Dispersion: Standard deviation, variance and quantiles (25%, 50%, 75%)
- Unique values: Frequency and cardinality analysis
- Null values: Detection and percentage display
- Type-adaptive statistics: Automatic detection of text, number or date
- Inline display: statistical results directly in the table
External Actions & Integration
- Google Search: Start directly with the cell content
- Google Translate: Translate selected content
- Google Maps: Interactive address search and map view
- XRechnung: Visualisation and display of structured invoice data
- Cross-File Lookup: Link to external Parquet files
- External applications: Launch from context menus
- Python: run your own scripts or analyses
User Interface & Interaction
- Multilingual interface: German, English, Dutch
- Design themes: Light, Dark, Custom
- Zebra pattern: Alternating row colours for better readability
- Font size: adjustable, DPI scaling for high-resolution displays
- Context menus: Quick access to all table functions
- Responsive: Layouts for different screen sizes
Prepare
Processing & Transformation
Stream Processing
- Partial sections: process very large data volumes in sections
- Memory consumption: significantly reduced thanks to streaming
- Stream extraction: Real-time filtering during import
- Stream grouping: Aggregation without memory overflow
- Stream join: Joining large files
- Stream append: Merging many files
- Stream pivot: Creating large cross-tables in the stream
- Based on DuckDB: Highest performance and efficiency
Data Joins
- Inner Join: Only matching records
- Left Join: All records from the left table
- Right Join: All records from the right table
- Full/Outer Join: All records from both tables
- Anti Join: Only non-matching records
- Fuzzy Join: Similarity-based matching
- Multiple keys: Join across multiple key columns
- Append / Concatenate: Add tables row by row
Data Cleansing
- Duplicates: find and remove automatically
- Delete empty rows: completely or partially
- Missing values: fill in (Forward/Backward Fill, mean, median)
- Outliers: Detection using statistical methods
- Data type validation: automatic correction
Column Transformation
- Merge: Combine columns with delimiters
- Split: By delimiter, length or pattern
- Extract: Numbers only, text only or by RegEx
- Replace: Substring or exact match
- Text operations: Trim, convert to upper/lower case
- Date operations: Extract year, month, day or time components
Aggregation & Pivot
- Grouping: by one or more columns
- Standard aggregations: Sum, average, count, min, max
- Advanced aggregations: Median, variance, standard deviation
- Pivot tables: Flexible cross-tables with multiple values
- Drill-down: Interactive detail view in pivot results
- Custom calculations: user-defined calculations and groupings
Calculated Columns
- Formula editor: with real-time preview
- Operators: +, -, *, /, %, power
- Statistical functions: Mean, median, standard deviation
- Date/time differences: Calculate time intervals
- Lag/Lead: Reference to the previous or next row
- If-then functions: Conditions nested as deeply as needed
- Cumulative sums: with grouping
- Python expressions: full support
New Columns
- Text columns: editable for manual input
- Number columns: editable for calculations
- Index column: Sequential numbering
- Check column: Manual markings (✔, ✖ or empty)
- Constant column: Fixed values for all rows
Comparison & Difference Analysis
- Table comparison: Compare two data sets directly
- Row-by-row analysis: Added, removed, changed entries
- Column-by-column differences: Make field changes visible
- Diff report: Clear summary of changes
- Excel export: Export differences directly
Ask AMY
AI Functions (AMY)
AI Chat Interface
- Ask my Data (AMY): Ask questions in natural language
- Answer formats: Text, table or interactive charts
- Code generation: automatic Python code for analyses
- Code help: Error correction and improvement suggestions
- Context preservation: Follow-up questions and multi-step dialogues
- Gatekeeper function: If a task is imprecise, AMY asks you to specify it
- Favourite prompts: Save prompts and run them via quick command
Supported AI Models
- AUDIPY AI: Mistral AI with zero data retention, no data rows sent to the AI
- OpenAI: various models (optionally integrable)
- Deepseek: various models
- GPT4All: Local models without internet connection
- Custom APIs: Connection to your own endpoints
- With AUDIPY AI: Only metadata (e.g. column names) is processed
- Data protection: no sensitive data content leaves the system
Magic Prompts & Playbooks
- Magic Prompt Generator: Context-dependent analysis suggestions
- Context: Industry, data content and system context are taken into account
- Automatic data quality check: Missing values, duplicates
- Outliers: Anomaly detection via prompt
- Distributions: Analysis to identify data patterns
- Playbooks: Creation for secure analysis with AI agents
AI-Assisted Column Creation
- AI Column Generator: Create new columns automatically
- No programming: complex transformations via AI
- Text patterns: Extraction via AI analysis
- Context values: automated calculation of context-related values
AI Visualisation & Insights
- Automatic chart generation: Selects the best-fitting chart
- Insight generation: Important findings are highlighted
- Trend detection: Identification of developments in the data set
- Anomaly explanation: AI provides reasons for unusual values
- Visual storyboards: Automatically generated analyses as a report
Documentation & Sharing
- AMY bookmarks: Save and re-run AI queries
- Documentation: all AI analyses performed are recorded automatically
- Python code: view, edit and export
- Save charts: interactive as HTML or as PNG
- Reproducible workflows: Analyses exactly repeatable
- Column description: created automatically by AMY
Analyse
Analysis & Visualisation
Descriptive Statistics
- Unique values (all columns): Identify primary keys
- Unique values (column): Frequency analysis
- Missing values: Check data completeness
- Value distribution: Percentage and absolute frequencies
- Quartiles: 25th, 50th, 75th percentiles
- Summary: All important key figures at a glance
Outlier Analysis (Anomalies)
- Isolation Forest: AI-based outlier detection
- Z-Score and Modified Z-Score: Classic statistical methods
- Rolling Z-Score: Optimised for time series
- Automatic mode: Selects the best method based on data type
- Category-based analysis: Group-specific evaluation
- Interactive visualisation: Display with rolling median
Distribution Analysis
- Histogram: various bin methods (Sturges, Rice, Scott, Freedman-Diaconis)
- Kernel Density Estimation (KDE): Smooth density distribution
- Box plot and violin plot: Compare quartiles and distribution
- Normal distribution: Comparison with the theoretical distribution
- Category-based overlays: Multiple groups in one view
Time Series Analysis
- Seasonal Decomposition: Breakdown into trend, seasonality and residuals
- Trends: Recognise patterns over time
- Rolling averages: Moving averages
- Trend charts: Line and area charts
- Seasonal comparison: Views across multiple periods
Benford Analysis (Digit Analysis)
- Forensic method: Detection of potential manipulation
- Digit tests: first, second and combined digit
- Last digit test: Validate trailing digits
- Category-based Benford analysis: Grouped tests
- Deviations: visual comparison with the Benford distribution
Gap Analysis
- Completeness: check sequential number ranges
- Invoice numbers: check for gaps and duplicates
- Grouped number ranges: by prefix or category
- Reset fields: for periodic numbers (e.g. annually)
- Timeline: Visualisation of gaps
Advanced Visualisations
- Network diagrams: Relationships between entities
- Sankey / Money Flow: Representation of cash flows
- Parallel coordinates: Multidimensional data analysis
- Scatter plots: Correlations between variables
- Bar and line charts: Visualise shares and categories
Interactive Charts
- Plotly: interactive visualisations
- Zoom and selection: Zoom, pan and select directly in the chart
- Tooltips: Detailed information on hover
- Export: as interactive HTML or static image (PNG)
Audit
Audit & Compliance
Monetary Unit Sampling (MUS)
- Sample planning: Poisson, Alpha-Beta and hypergeometric
- Cumulative value selection: Systematic selection
- Evaluation: Evaluate samples using various methods
- Tainting & extrapolation: Projection to the population
- Reproducible: thanks to seed values
- PDF report: for documentation
Bookmark System
- Analysis bookmarks: Save complete views
- Reuse: Filter, pivot and chart bookmarks
- Notes: Descriptions for each entry
- Colour markings: for categorisation
- Quick access: via the dashboard
Data Integrity & Security
- Hash chain technology: Guarantee immutability
- SHA-256, MD5 and xxHash: Cryptographic checksums
- Signature verification: Ensure data provenance
- History report: Complete change history
- Metadata management: Store extended data information
- API keys: encrypted for secure integration
Export & Reporting
- File export: Parquet, CSV and Excel with formatting options
- PDF reports: with integrated visualisations
- HTML export: Interactive reports as web pages
- Power BI export: Create PBIX files directly
- Column selection: targeted export
- "What you see is what you get": Filters are carried over
Sampling Methods
- Random and systematic: samples drawn randomly or systematically
- Stratified: Samples by category
- Reproducible: Results through seed values
- Sample size: automatic calculation
Compliance Support
- GoBD export: hand over data and audit steps traceably
- Secure: local processing, no telemetry
- Tamper detection: chained SHA-256 hash values
- Audit trail: complete for full traceability
- Privacy by Design: Data protection integrated from the start
Automate
Automation & Agent Playbooks
Audit Steps
- Recording: capture work steps at the push of a button and save them as an audit step
- Audit Steps: create audit steps once, reuse them as often as you like
- Audit Step Editor: build your own audit routines with input dialogs
- My Audit Manager: bundle and manage audit workflows
- Log: every action documented automatically
Agent Playbooks (in all packages)
- Agent Playbook: Step-by-step instructions that an AI agent can read and execute independently in AUDIPY
- Visual creation: Build audit steps in the editor as usual, click, drag, done
- Export as Agent: One click automatically creates the complete instructions for the AI
- Automatic execution: The AI agent reads the playbook and carries out all steps independently
- Traceable: Written in plain language, anyone can read what the AI will do
Configuration Options
- Language: Choose German, English or Dutch
- Level of detail: Compact, Detailed or Expert
- Column mapping: Flexible (recommended) or Strict
- Shareable: Pass on and share playbooks as files
- Flexible: Automatically adapts to your data
Error Handling & Reports
- Ask: The AI asks when problems occur, you decide
- Stop: The AI halts and displays the error
- Continue: The AI skips the issue and documents it automatically
- Automatic report: PDF report with results after the analysis
- Save results: Automatic saving of all analysis data
- AUDIPY required: AUDIPY must be available for execution but can run in the background. The playbook is executed by an AI agent of your choice via MCP
Benefits at a Glance
- Time savings: Routine analyses in minutes instead of hours
- Consistency: Always the same result, no deviations
- No IT knowledge required: Visual editor, no programming needed
- Multilingual: German, English and Dutch
- Error-proof: Intelligent error handling built in
Python & Developer Mode
- Python scripts: automate individual audits
- Metadata management: Read, delete, replace
- Debug logging: for detailed error analysis
- Feature toggles: Activate experimental functions
Foundation
Performance & Scalability
Big Data Handling
- Streaming operations: Process large data volumes without fully loading them
- DuckDB integration: In-memory analytics for millions of rows
- Chunked processing: Processing in flexible blocks
- Intelligent memory management: Automatic memory cleanup
- Lazy loading: Load only the data currently needed
- Column selection: Read only relevant columns
Asynchronous Processing
- Background processing: The UI remains responsive during processing
- Multi-threading: Parallel execution of tasks
- Resource management: Automatic release of unused memory
Visualisation of Large Data Volumes
- Figure resampling: Automatic reduction of data points for high performance
- Zoom-based aggregation: Detailed view when zooming in
- Interactive downsampling: Smooth navigation without loss of detail
- Efficient rendering: Optimised display even with millions of points
- GPU-accelerated visualisation: Maximum performance at high resolution
Optional add-on module
AuditPlus
Add-on module for your AUDIPY licence. More about AuditPlus →
DataFlows
- DATEV: ready-made preparation up to the standard journal
- SAP DART: ready-made flow for structured data preparation
- Reusable: set up a flow once and build on it in the next analysis
AuditSteps
- Balance sheet structure: Map accounts, create balance sheet, P&L and statistics
- ForecastAudit: Examine time series, forecasts and deviations
- Word search: Search posting texts in a targeted way, fuzzy too
Data Models
- Templates: Data model templates for DATEV, SAP DART and XRechnung
- Your own models: Link tables yourself, relationships remain visible and editable
- Check locally: Verify links before using them in analyses
- With AMY: Have relationships suggested (with AI package)
Dashboards
- Templates: e.g. annual financial statements, posting review and journal entry testing
- Editable: Switch data fields, add key figures, adjust the display
- Interactive filtering: one click filters all connected charts
- Drilldown: from the year down to the month and to the underlying data
- Library: Store the dashboards of an audit in an organised way
- With AMY: Create a dashboard from your question (with AI package)
No matching feature found. Feel free to ask us directly.
See the features live.
On request, we show you AUDIPY online in English, free of charge, focusing on the features that matter for your work.