Data and files
Counterparty data quality audit — upload a CSV and see what is wrong
Before you connect a counterparty database to an e-invoicing system, migrate it, or run a bulk mailing from it, it is worth knowing what state it is in. This tool detects columns automatically, verifies the checksums of every identifier, looks for duplicates by number and by name, and flags columns with inconsistent formats. The file is read locally and never uploaded.
Free · no signup · runs in your browser
Drag a CSV file here, or click to choose one
The file is read in your browser and never uploaded. From Excel: File, Save As, CSV.
How it works
- 01
Upload a CSV file
Comma or semicolon separator, UTF-8 or Windows-1250 encoding - we handle both. From Excel: Save As, CSV.
- 02
Columns are detected automatically
By header and by content: tax IDs, company registers, IBAN, EU VAT, EAN, email, phone, postcode, date, amount.
- 03
Get a report and a list of rows to fix
Column fill rates, failed checksums, duplicates, inconsistent formats. The list of problem rows downloads as a CSV.
Assumptions and limits
- The file is read entirely in your browser. It is not uploaded to our server, not stored, and we have no access to it - you can disconnect from the internet after the page loads and the tool still works.
- We support CSV, not XLSX. That is deliberate: parsing Excel files would require a large library that would slow the site down for everyone, including people not using the audit. Exporting to CSV takes seconds in Excel.
- Column detection is heuristic: it is based on the header and on what the cells actually contain. If a column is detected wrongly, you can change its type by hand.
- Duplicates by name are found after normalisation: we ignore case, legal forms, punctuation and Polish diacritics. Two spellings of the same company name will be flagged as one entity.
- The result is guidance, not a verdict. A record failing a checksum almost always needs fixing, but a name duplicate may be two genuinely different branches.
- Large files are processed in chunks so the browser does not freeze. Tens of thousands of rows still take seconds.
Frequently asked questions
Does my customer data file go anywhere? +
No. The entire analysis runs in your browser - the file is not sent to any server, does not reach any log and is not stored by us. You can verify it: load the page, disconnect from the internet and drop the file in. It still works.
I have an XLSX file, not CSV +
In Excel choose File, Save As and pick CSV. We deliberately did not build in an XLSX parser, because it is a large library that would slow the site down for every visitor.
What do I do with the result? +
Download the list of problem rows as CSV, fix them in the source system and run the audit again. Start with failed checksums and gaps in key fields, then duplicates - those need a human decision.
Why does this matter before an integration or migration? +
Because target systems reject bad records, and when importing tens of thousands of rows the rejections surface one at a time. Cleaning up first turns weeks of fixes into a single run.
Can you do this for us on the full production database? +
Yes, cleaning and deduplicating data is a standard part of preparing for a migration or integration. A technical audit starts from PLN 3,000, and you get a ballpark instantly in our quote calculator. A human replies within 24 hours.
Related tools
All tools →Number validators
Nine identifiers in one place: checksum, detected type and the reason for any error.
JPK and KSeF viewer
Upload a JPK_V7 file or a KSeF FA(3) invoice and get a readable summary plus detected anomalies.
Integration cost
Estimate the effort and cost range of connecting an ERP, webshop, warehouse and accounting.
Turn the numbers into a project
Get a ballpark instantly, a reply from a human within 24 hours and a firm quote after a short call. No obligation.
Estimate your project in 2 min