FlowCSV
How it worksClean a CSV

CSV cleanup for operations teams

Turn ugly operational data into system-ready data.

Clean, standardize, map and export messy CSV files in minutes — without spreadsheets, formulas or technical expertise.

Clean a CSVTry demo data

No account needed · files are processed in your browser and never uploaded

customers_export_legacy.csv

Needs attention
BeforeAfter

Customer Name

··JOHN·SMITH·John Smith

Email Address

John.Smith@Email.COM·john.smith@email.com

Phone Number

(519)·555-1234519-555-1234

Province

ontarioON

Province

OntarioON

Signup Date

07/03/20232023-07-03

Real values from the built-in demo file. Nothing is changed until you approve it.

The problem

Exports arrive as five spellings of the same thing.

A legacy system exports customers. The province column contains ON, Ontario and ontario. Phone numbers use brackets in some rows and dots in others. Names shout in capitals. Half the emails carry a trailing space, so the import fails on row 12 and you spend an afternoon in a spreadsheet finding out why.

Same value, four spellings

  • ON
  • Ontario
  • ontario
  • Ont.

Same number, four formats

  • (519)·555-1234
  • 519.555.1234
  • 5195551234
  • +1·519·555·1234

Same date, four layouts

  • 07/03/2023
  • 2023-07-03
  • Jul·3·2023
  • 3·July·2023

Same person, twice

  • John·Smith
  • john·smith·
  • JOHN·SMITH
  • John··Smith

A note on dates: numeric dates such as 07/03/2023 are read month-first, so that is 3 July 2023. Anything FlowCSV cannot read confidently is left exactly as it was.

How it works

Five steps, no formulas.

  1. 01

    Upload

    Drop a CSV in. It is read on your own device.

  2. 02

    Analyze

    See a plain-language report of what is actually wrong.

  3. 03

    Clean

    Approve the fixes you want. Switch off anything you don't.

  4. 04

    Map

    Rename, reorder and drop columns for the receiving system.

  5. 05

    Export

    Download a clean file, and save the setup as a profile.

What FlowCSV can detect

Read-only analysis, before anything changes.

Duplicate records

Rows that repeat once whitespace and casing are ignored.

Empty fields

Blank values in columns where the rest of the file has data.

Formatting inconsistencies

The same value written five different ways.

Invalid emails

Missing domains, stray spaces, uppercase addresses.

Phone problems

Brackets, dots, spaces, country codes, too few digits.

Date inconsistencies

Day-first, month-first, dotted and ISO dates mixed together.

Province and state values

ON, Ontario, ontario, Ont. treated as one value.

Column naming problems

Trailing spaces, mixed casing, unusable header names.

Whitespace

Leading, trailing and doubled spaces inside values.

Wrong column order

Columns renamed, reordered or dropped for the receiving system.

Demo you can run right now

A customer export, deliberately broken.

FlowCSV ships with a 33-row legacy customer file containing duplicates, four date formats, inconsistent provinces, shouting names, padded values, malformed emails and blank fields. Load it without uploading anything of your own and watch it come out the other side.

Run the demo

What the analysis reports

  • Duplicate rowsremoved once approved
  • Inconsistent province valuesstandardized to two-letter codes
  • Mixed date formatsconverted to YYYY-MM-DD
  • Malformed phone numbersreformatted, or flagged and left alone
  • Email formatting issueslowercased and trimmed
  • Whitespace correctionsleading, trailing and doubled spaces

Exact counts depend on your file — FlowCSV reports what it actually finds.

Templates

Column layouts for the usual destinations.

Each template sets the export column names, their order and the rules that start switched on. They are FlowCSV layouts modelled on common import formats — not official integrations — and you can save your own as a reusable profile.

QuickBooks Customer Import

Contact-style output with snake_case field names, standardized provinces and de-duplicated rows.

CRM Contact Import

First/last name, email and phone for a typical CRM contact import.

Shopify Product Import

Handle, title, SKU and price columns for a product catalogue upload.

Accounting Import

Transaction-style output with ISO dates and a single amount column.

Inventory Import

SKU, description, quantity and location for a stock count upload.

Payroll Import

Employee identifiers, hours and pay rate with ISO dates.

Custom Format

No target schema. Column names are converted to snake_case and you choose the rules yourself.

Got an ugly CSV? Clean it.

No account, no upload, no formulas. Your file is read in this browser and comes back ready for the system that rejected it.

Start cleaning