Spreadsheet Cleaner

Fix dates written many ways

05/01/2024, 5 janvier 2024, Jan 5, 2024: all in one format. When a date is ambiguous, you choose.

  • No account to create
  • Never saved on our servers
  • Every change logged, nothing erased
The report is free. You only pay to download.
A small bike shop's back office: on the laptop, the report of a messy client list, 99 things to fix.
Illustrative photo, fictional data.

What Spreadsheet Cleaner fixes

The problems you find in a customer list, a contact export, a product list or a sales log.

Rows and columns

  • Exact duplicates
    Marie Tremblay (twice)Marie Tremblay
  • Likely duplicatesYour call
    Marie-Ève Côté · marie@gmail.com Marie Eve Cote · MARIE@gmail.comMarie-Ève Côté · marie@gmail.com
  • Empty rows
    Empty rowRemoved
  • Headers to sort out
    Name, City (repeated row 240)Removed
  • Merged cells
    Region merged over 3 rowsEast · East · East

Contact details

  • Phone numbers written many ways
    (514)555.0123514 555-0123
  • Postal codes
    h2x1y4H2X 1Y4
  • Emails to fix
    Luc@GMIAL.conluc@gmail.com
  • Split first and last nameYour call
    Tremblay, Marie-ÈveMarie-Ève | Tremblay

Values

  • Dates written many ways
    Jan 5, 20242024-01-05
  • Amounts and numbers stored as text
    $1,234.561234.56
  • Long numbers in scientific notation
    1.23457E+151234567890123456
  • Extra spaces
      Laval Laval
  • Names in all caps or all lower case
    SAINT-JÉRÔMESaint-Jérôme
  • Broken accents
    QuébecQuébec

What you get

The real file produced from the fictional sample, as it opens in Excel: three tabs, nothing lost in silence.

storefront-clean.xlsx
NameEmailCityZIP codeSignup date
Emily Carteremily.carter@example.comChicago606142024-03-02
Jason Millerjason.miller@example.comDenver802022024-03-14
Megan Brooksmegan.brooks@example.comBoston021082024-03-22
Daniel Reyesdaniel.reyes@example.comAustin787012024-04-01
  • ZIP code

    The leading zero restored

    A 4 digit ZIP code in a column of 5 digit codes becomes 02108 again, not 2108.

  • Removed rows

    Removed rows, kept

    The exact duplicate is not erased: it is copied as is onto its own tab, with the reason.

  • An Excel file (.xlsx) with these three tabs, or a plain CSV for your accounting software.

The cleaned file printed on a sheet, the postal code column highlighted in yellow, a highlighter resting on it.
Illustrative photo, fictional data.

See it work

A real recording of the site with a fictional file: dropping it, the check, and the preview of the fixes.

Real recording, no sound. Fictional file and data.

Free to see. Paid to download.

You see the full report and the first clean rows without paying anything. If you like the result, the exact price, currency and tax are shown before any card is asked for.

  • The report and the preview are free
  • One payment per file, no subscription
  • No account, no email asked
  • Refunded on request within 30 days
  • Nothing to fix? Nothing to pay.
Drop a file

Your file stays private

Processed in memory

The file is read in memory while it is processed, then forgotten. It is never written to a disk or kept on our servers.

No contents in our logs

Our logs record the file size and the number of rows, never a name, an email or an amount.

Nothing for Stripe

Stripe receives the price and a fingerprint of the file to link the payment to the right file. Never its contents.

The rest stays on the server

Before payment, your browser only receives the report and the first rows. The blurred rows are empty shapes.

Deeper check

To catch duplicates and typos that simple rules miss, a small part of the file is sent to Anthropic, a US based AI provider, which processes it in the United States. That part is a few examples from the report and, only if you turn on likely duplicates, the columns that identify a person, a company, a product or a city (name, company, email, phone, postal code, city) for rows that look alike. Anthropic does not train its models on this data and deletes it within 30 days, with the exceptions listed in our policy.

Read the privacy policy

Questions

Is my file changed?

No. Your original file stays as it is on your computer. You download a new, clean copy.

Which files work?

Excel (.xlsx) and CSV files, up to 4 MB and 20,000 rows. For an old .xls file, open it in Excel and save it as .xlsx. If the workbook has several sheets, the first sheet with data is cleaned.

What if a fix is wrong?

Every change is logged on the Changes tab, with the value before. Removed rows are kept on their own tab. Fixes that can remove information start turned off.

What about product codes with leading zeros?

They stay intact. A code like 00123 stays 00123, and a long number stays as text so it never shows in scientific notation.

Why is 03/04/2024 not fixed?

It can mean April 3 or March 4. If the rest of the column does not settle it, we ask you to choose instead of guessing.

Do you use artificial intelligence?

Rules written in advance do the work first. Then an AI service (Anthropic) checks for likely duplicates and typos the rules miss, and explains the problems in plain words. Every suggestion is checked by our rules before it is applied, and it is marked on the Changes tab.

How do I get a refund?

Write to bonjour@samrostudios.com within 30 days with your receipt. No need to explain.