Features
Everything You Need to Explore Structured Data
Discover the analytical tools that make EuroData Score a practical workspace for organizing, examining, and visualizing datasets.
Structured Dataset Processing
Upload a CSV or JSON file and EuroData Score parses it on the server, detects the delimiter, normalizes header names and classifies every column as numeric, date, boolean, categorical or text.
- Problem solved
- Spreadsheets mix numbers, dates and labels in ways that make analysis error-prone. Explicit column types tell you exactly how each field will be treated.
- Practical example
- A sales export with 1,200 orders becomes a typed table in seconds, ready for statistics, frequencies and trends.
- Accepted data
- CSV or JSON, up to 5 MB, 50,000 rows and 100 columns. Excel workbooks must be saved as CSV first.
Records
1,122
Columns
10
Valid cells
99.6%
- order_idText
- order_dateDate
- regionCategorical
- categoryCategorical
- channelCategorical
- unitsNumeric
- unit_priceNumeric
- discount_pctNumeric
- revenueNumeric
- customer_ratingNumeric
Data Quality and Cleaning
A data-quality panel counts missing values, exact duplicate rows and values that do not match their column type, and lists the columns that need attention with examples of the problem values.
- Problem solved
- Duplicates inflate counts and sums; invalid entries silently distort averages. Seeing them before you analyze avoids misleading conclusions.
- Practical example
- Spot “n/a” entries in a price column and create a de-duplicated copy of the dataset with one click, keeping the original intact.
- Accepted data
- Works on every uploaded dataset and every column type.
Missing values
47
Duplicate rows
4
Invalid values
2
Columns needing attention
- region — 1 missing value(s)
- category — 1 missing value(s)
- unit_price — 2 value(s) do not match the number type (e.g. “n/a”)
- customer_rating — 45 missing value(s)
Descriptive Statistics
For every numeric column: count, sum, mean, median, mode, minimum, maximum, range, sample and population variance and standard deviation, quartiles, IQR, percentiles from P5 to P95 and skewness.
- Problem solved
- Calculating summary statistics by hand or with spreadsheet formulas is slow and easy to get wrong. Methods here are documented and tested.
- Practical example
- Compare the mean and median order value to see whether a few large orders are pulling the average up.
- Accepted data
- Numeric columns. Missing and invalid cells are excluded and reported separately.
- Count
- 1,122
- Mean
- 473.56
- Median
- 184.88
- Std. deviation
- 775.96
- Minimum
- 2.7
- Q1 (25th pct.)
- 57.59
- Q3 (75th pct.)
- 502.47
- Maximum
- 5,719.99
- Skewness
- 3.17
0 invalid and 0 missing value(s) excluded · sample standard deviation (n − 1)
Frequency Analysis
Count how often each value occurs, with percentages and cumulative percentages, sortable by count or value. Numeric columns also get an equal-width histogram with adjustable bins.
- Problem solved
- Understanding how values are distributed is the first step in most analyses, but pivot tables take time to set up for every column.
- Practical example
- See which product categories make up most orders, then switch to a histogram of unit prices.
- Accepted data
- Any column. Histograms use numeric columns; a date-range filter is available when the dataset has a date column.
- Office Supplies353 · 31.5%
- Accessories236 · 21.1%
- Electronics213 · 19.0%
- Software173 · 15.4%
- Furniture146 · 13.0%
5 distinct values · 1 missing excluded from percentages
1,120 values in 12 equal-width bins
Trend Visualization
Aggregate a numeric metric by day, ISO week, month, quarter or year using sum, mean, median, count, minimum or maximum. See period-over-period changes and compare two date ranges.
- Problem solved
- Time-based questions — “how did this change month to month?” — normally require manual grouping and formulas.
- Practical example
- Plot monthly revenue totals, find the highest and lowest months, and compare Q1 with Q4.
- Accepted data
- A date column (YYYY-MM-DD or YYYY/MM/DD) and a numeric metric.
Highest · 2025-11
69,530
Lowest · 2025-08
18,435
Pattern Exploration
A Pearson correlation matrix across numeric columns, unusual-value detection using Tukey's IQR fences or z-scores, group comparisons of a metric across categories, and cross-tabulations of two categorical columns.
- Problem solved
- Relationships and anomalies are hard to see in raw tables. These methods surface them — with their limitations stated alongside the results.
- Practical example
- Check whether units and revenue move together, then compare average revenue by sales channel.
- Accepted data
- Numeric columns for correlations and unusual values; categorical columns for groups and cross-tabs.
| Column | units | unit_price | discount_pct | revenue | customer_rating |
|---|---|---|---|---|---|
| units | 1.00 | -0.20 | 0.02 | 0.11 | -0.01 |
| unit_price | -0.20 | 1.00 | -0.05 | 0.76 | -0.05 |
| discount_pct | 0.02 | -0.05 | 1.00 | -0.04 | -0.11 |
| revenue | 0.11 | 0.76 | -0.04 | 1.00 | -0.05 |
| customer_rating | -0.01 | -0.05 | -0.11 | -0.05 | 1.00 |
127 unusual values in revenue outside Tukey's fences (-609.74 to 1,169.8).
Pearson r, pairwise-complete observations. Correlation does not imply causation.
Exportable Reports
Save any analysis as a report containing the dataset, date, columns and settings used, a short methodology, statistics, charts, observations and limitations. Print it or save it as PDF from your browser.
- Problem solved
- Results are only useful if they can be shared and revisited. Reports freeze the numbers so they stay consistent.
- Practical example
- Create a one-page summary of regional order volumes and export the underlying table as CSV.
- Accepted data
- Tables export as CSV (UTF-8). Reports print or save as PDF through your browser's print dialog.
Analysis report
Units by region — summary
Dataset: Sample Retail Orders 2025 (synthetic)
Methodology: Descriptive statistics and frequency counts on non-missing values.
Statistics: units — mean 4.85, median 3, max 24
Distribution: North 22.4% · East 21.8% · West 21.4%
Limitations: Results describe this dataset only and do not predict future values.
Organized Workspace
Keep datasets, saved analyses and reports in one place. The History page lets you reopen an analysis with the same settings, and every item can be renamed or deleted.
- Problem solved
- Analyses scattered across files and tabs are hard to reproduce. A single workspace keeps your work traceable.
- Practical example
- Reopen last week's frequency analysis on the same dataset and update the report.
- Accepted data
- Up to 50 datasets per account.
- Overview
- My Datasets
- Statistical Analysis
- Frequency Analysis
- Trends
- Insights
- Reports
- History
My Datasets
1,122 records · 10 columns
Use cases
Designed for Everyday Data Analysis
EuroData Score focuses on the analytical tasks that come up most often when working with tables of records.
Reviewing spreadsheets
Export a sheet as CSV and check its quality, types and totals before sharing it.
Analyzing numerical sets
Summarize measurements, prices or scores with documented statistical methods.
Exploring frequencies
See which categories, responses or values occur most and least often.
Organizing records
Find duplicates, missing fields and inconsistent values across thousands of rows.
Comparing metrics
Compare a metric across groups or between two date ranges.
Preparing reports
Turn results into a structured report with methodology and limitations for colleagues or clients.
See the tools on real sample data
EuroData Score — Full Access is a one-time payment of €42. No subscription, no recurring charges, and a 60-day money-back guarantee.