Excel
How to analyse a CSV file
A CSV has no types, no formatting and no metadata. Everything you need to know has to be derived, which is exactly what a profiling pass does.
Updated September 20, 2026
Before analysis: the boring checks
- Delimiter: comma, semicolon or tab — European exports often use semicolons
- Encoding: mojibake in names usually means UTF-8 read as ANSI
- Header row present, and unique column names
- Quoted fields containing the delimiter
Profile every column
For each column record its inferred type, filled and blank counts, unique values and, for numbers, min, max, average and total. This single table answers most first questions about a file.
Then ask the analytical questions
- What is the grain — one row per what?
- Which columns are keys, which are dimensions, which are measures?
- What period does the file cover, and is it complete?
Frequently asked questions
Is my file uploaded to a server?
No. On the free tools the file is parsed in your browser. Only a small aggregated summary is sent when you ask for AI analysis.
Which file types are supported?
CSV, TSV, TXT and XLSX or XLS.
Related reading
Analyze your data with AI
Open the GridMind workspace, bring in a CSV, Excel file or your Power BI model, and let AI build the formulas, charts and dashboards for you.