Your data
Getting your data in, and being able to take it out again
A template to download, a dry run that tells you row by row what is wrong before the database is touched, and an upload that is all or nothing. Upload the same file twice and not one line is duplicated.
Nobody is afraid of the new tool. They are afraid of the move.
The software is not the worry. The worry is the Saturday spent typing in eighty people by hand, and the Monday when you find twelve are missing and three are in twice. So the import in Fichaloto is built around one rule that comes before all the others: it never writes anything without showing you first what it is going to write.
How an upload goes, start to finish
Two stages, and the first one touches nothing:
- You download the template — headers and one example row, flagged so it is skipped if you upload it as it came. The export works too: the format exported is exactly the format imported, so a file downloaded from Fichaloto can be corrected and sent straight back up.
- You run the dry run — nothing is written. Out comes a report with one line per issue: row, column, value, what is wrong and what to do about it. Errors stop the upload; warnings do not.
- The report downloads as CSV — so you can fix the file with the report open beside it, instead of hunting for row 1,743 on screen.
- You apply it — all or nothing, in one transaction. A single error and not one row is written. There is no such thing as a half-finished upload here, which is the kind that makes Monday impossible to reconstruct.
- The report is kept — you can reopen it later to see what went in and when.
Uploading the same file twice duplicates nothing
Every entity declares which of its columns are the key. If the key already exists, that row is updated; if it does not, it is created. Projects by code, and by name where there is no code. Time entries by their external reference, and where they arrive without one a fingerprint is derived from the person, the date, the project and the note, so an exported row recognises itself when it comes back up.
And one detail that saves more time than it sounds like: a blank cell leaves that field alone. Upload a file with three columns filled in and the rest empty and you update those three, leaving everything else as it was. To empty a field deliberately you write a dash, which is an explicit decision rather than an oversight.
What you can move
Sixty-five entities export and fifty-three import. It is not just the headcount: clients, projects, phases, tasks, milestones, assignments, time entries, costs, absences, calendars and public holidays, working patterns and shifts, clock-in records, work centres and kiosks, sales opportunities, tickets, invoices and their lines, numbering series, payments received, roles and permissions, and the ten account catalogues.
Twelve of them are download-only, for a reason: attachments, ticket history, clock marks, invoice lines, payments received. Those are the record. Writing over a record from a spreadsheet is the quickest way for it to stop being one.
The same entities are in the API, under the same names. The API.
The details that stop a European CSV falling over
Half the trouble in an import is not the data, it is the file. These are handled and there is nothing for you to set:
- The delimiter is detected — semicolon or comma, whichever wins on the first line. Excel on a European locale saves with semicolons and you do not have to tell anyone.
- The
sep=;line Excel writes is discarded, and so is the byte order mark. - Files that are not UTF-8 are converted from Windows-1252 — accents and non-English characters in an old file come through intact.
- Exports follow the language for the delimiter — semicolon in Spanish, comma in English, and the same for the decimal separator.
- 8 MB and 20,000 rows per file — a year of time entries for a mid-sized headcount fits comfortably; a large migration is done in batches.
- Or you paste the rows — no file at all, straight out of the spreadsheet.
Frequently asked questions
Can I upload an .xlsx directly?
No: it is CSV. Excel and Google Sheets both save as CSV from a menu, and in return the format is readable, opens in anything and does not depend on which version of Office anyone is running. We would rather be clear about that now than later.
What happens if an import fails halfway?
It cannot: it runs in a transaction. Either the whole file goes in or none of it does, and either way you are left holding the report of what needed fixing.
Can the import delete whatever is not in the file?
There are two modes, update and replace. Replace is blocked on entities with history hanging off them, and if something left over cannot be deleted for integrity reasons, that is an error and nothing is applied.
Can I take all my data with me if I leave?
Yes, and you do not have to ask: all sixty-five entities download as CSV from inside the application, with whatever filters you have applied, in the language you are working in. A product you cannot leave is a product you should not join.
Who can import?
It goes by permission, and there are two of them: one to download and one to upload. Somebody can export reports every month without being able to write a single row.
Getting started
Fourteen full days, no card and nothing to install. If it has not saved you work by then, there is nothing to cancel. And if you would rather see it first, book a demo and we will set it up with two or three of your real projects, which is where you find out whether it fits.