HRIS Import Validator
Every import error in a file at once, checked against your system’s rules, with the fixes that have one right answer already made.
Reads .xlsx and .csv inside your browser. Nothing is uploaded, least of all to the system you are about to load it into.
The file you are about to import
Drop a spreadsheet here
The spreadsheet you plan to load into your HR system, payroll or any other system with an import. Rules are suggested from its headers, and you can change every one.
An import stops at the first error
Most HR systems reject a file one problem at a time, and often quote a row number that is off by one because they count from the first row of data. So somebody fixes row 212, uploads again, and finds out about row 340. This checks every row against every rule at once and gives row numbers the way your spreadsheet shows them, with the header as row 1.
A spreadsheet has usually been at the file
Somewhere between the system it came from and the import, the file was opened in a spreadsheet, and that quietly does three things. Employee numbers lose their leading zeros. Dates turn into numbers like 45123. Long numbers become 1.23457E+11, which has already thrown the last digits away. The first two have one right answer and are put back. The third cannot be, and the tool says so, because the only fix is a fresh export opened as text.
A picklist is exact
“Full time” is not “Full-Time” to an import, and neither is “Full-Time” with a space on the end that nobody can see. When a value matches exactly one allowed value once case, spacing and punctuation are ignored, it is changed to the spelling the system uses. Anything else is left for a person, with the closest allowed value offered as a hint, never applied as a guess.
A manager has to exist first
A manager ID that is neither in the file nor already in the system fails the row, or loads it with no manager at all. Paste in the IDs already in the system and they count. Some systems check rows in order, so a manager listed below their own report fails too; the ready file moves each manager up to just above their first report and leaves everything else where it was. And a clean row whose manager’s row has a problem is held back with it, since it cannot load until its manager does.
Only fixes with one right answer
Every correction is listed in the change log with what it was, what it is now and why: stray spaces and invisible characters, an allowed value’s exact spelling, a date rewritten in the format the system asks for, zeros restored on a number of fixed length, and names like “José” decoded back to “José”. An exact copy of a row is left out. Two rows that share an ID but disagree are both held back, because deciding which is right needs a person.
What this does not do
It checks the rules you give it. It does not know your system’s own rules, so set them from the system’s import template and save them to use next time. A file that passes here can still be rejected for a rule it was never told about. For checking a file’s general health before you have any rules at all, the HR Data Quality Checker does that job.