Data quality report for Excel
Upload your sheet and download a per-column report with blanks, % blanks, stray spaces, numbers stored as text and distinct values, plus a duplicate-row warning.
What it is for
Before analysing a spreadsheet it pays to know what state it arrives in. Data faults — a half-empty column, numbers that won't add up because they are stored as text, invisible spaces that split one value into two — don't show up by eye and quietly ruin any analysis built on top. This report counts them per column so that, at a glance, you know what to clean and with which tool.
It is the pre-flight check: it fixes nothing, it tells you what to fix.
What it measures, column by column
The result is an .xlsx with one row per column of your sheet and these
signals:
- Blanks and % blanks — how many cells are empty and what share of the total they are. A high percentage in a column that should be complete is the first red flag.
- Stray spaces — text cells that start or end with spaces. They are invisible
and make
"Madrid"and"Madrid "count as different values when you group, count or join. - Numbers as text — cells that look like a number but are stored as text, so they neither sum nor sort. It is the costliest fault to catch by eye.
- Distinct values — how many different values there are, handy for telling a category from an identifier.
On top of that, it checks at the end for fully duplicate rows (identical in every column) and warns you how many there are.
Which tool fixes each thing
The report is built to lead you to the fix for each problem it finds:
- Stray spaces → clean data, which trims spaces and normalises text.
- Numbers as text → convert text to number, which turns them into real numbers.
- Duplicate rows → remove duplicate rows if they are surplus, or highlight duplicates if you would rather review them first.
Once it is clean, the statistics summary or group and count work on data you can trust.
A note on leading-zero codes
A value like 00123, a tax number or a phone number starting with zero is not
flagged as "number stored as text", even though it looks like one. Converting it
to a number would eat the leading zeros and break it: those codes should stay as
text, so the report does not point at them as something to fix.
Frequently asked questions
Does this report fix my sheet?
No. It only diagnoses: it counts the problems per column and points you to the tool that fixes them. Your file is not modified; what you download is the report.
How does it detect duplicate rows?
It compares the whole row: only a row identical to another in every column counts as a duplicate. For a narrower criterion, on a single key column, use highlight duplicates.
Why doesn't a code like 00123 show up as a number stored as text?
Because it shouldn't be converted: turning it into a number would lose the leading zeros. The report deliberately leaves out values that start with zero so it never nudges you into breaking codes, phone numbers or identifiers.
Does a space in the middle of the text count as a "stray space"?
No. Only spaces at the start or end of the cell are flagged, since those are the ones that break comparisons. Spaces between words are normal and not counted.
Does it work with CSV and legacy .xls?
Yes. The report always comes out as .xlsx, whatever format you upload.