Why it matters
Most import failures and wrong reports come from a handful of problems in the data itself: a date written the American way, a price with a pound sign, a total of "N/A", a stray space after a postcode, or text that went through the wrong encoding and came out as "Café". They're easy to miss in a spreadsheet and slow to find one column at a time.
This profiler reads the whole file in your browser and reports on every column at once: what type it holds, how much is missing, the spread of values and anything that doesn't fit. It also checks the file as a whole for duplicate and ragged rows and encoding damage. The statistics match pandas.
How to use it
- Open the CSV or paste it in.
- Check the delimiter and header settings.
- Read the column table and the file checks.
- Pick a column to see its statistics and distribution.
FAQ
Is my data uploaded?
No. The file is read and profiled in your browser.
How are outliers found?
Values more than 1.5 times the interquartile range below the first quartile or above the third, the rule used in box plots.
What is mojibake?
Garbled text such as “Café”, made when UTF-8 text is read as another encoding and saved again. The original characters can usually be recovered.
Why are some dates flagged as US order?
A date like 10/13/2026 only works as month/day, so if the rest of the column reads as day/month, one of them is wrong.
