Choose Files is the first stage of an upload. You add the files you exported from your core system, and for each one you say what kind of file it is. Vintage reads each file’s columns in your browser so it can check them, but nothing is uploaded at this stage.
The three file groups
Section titled “The three file groups”The screen shows three cards. Each is a file group, one kind of file within the upload:
| Card | Use it for | One row is |
|---|---|---|
| Loan Snapshots | Your portfolio at a point in time (monthly, quarterly, or yearly) | One loan, as it stood at that moment |
| Origination Files | The original terms and balances of each loan when it was made | One loan, at origination |
| Transaction Files | Individual loan events: charge-offs, recoveries, payoffs, prepayments | One event on one loan |
By choosing a card, you declare the file’s type. Vintage never guesses it from the file’s contents. The type you choose decides which Standard fields (the fields Vintage understands and models with) are offered for its columns at Column Mapping (a transaction file is offered the charge-off and transaction fields, for example, not the snapshot-only Payoff Flag and Prepayment Flag).
You can use any combination of the cards in one upload, and each card can hold many files. A backfill (your historical months, loaded at once) of several years of monthly snapshots is many files under Loan Snapshots in a single upload. Preparing your data explains which files unlock which results.
Adding files
Section titled “Adding files”- Click a card. A window opens for that file type.
- Drag your files into it, or press Browse files and pick them. You can add several at once.
- The window reads each file’s columns in your browser (“Reading 2 of 5 files”) and closes when it is done. Nothing is uploaded yet.
Added files appear below the cards, grouped under their type, each with the number of columns read. Remove a file with its remove control if you added it by mistake. Continue becomes available once at least one file is present and no file breaks the rules below.
Every file in a group shares one set of columns
Section titled “Every file in a group shares one set of columns”All files under one card in one upload must have the same column names. The order of the columns does not matter; the names do. This is checked the moment you add a file: a file whose columns do not match the rest of its group is listed with the problem named, and Continue stays unavailable until you remove it or fix it:
All Loan Snapshots files must share the same columns. Remove the mismatched file or upload it separately.
The reason is that you make two decisions once per group rather than once per file: which columns hold personal information, and what each column means. Those decisions are recorded against column names, so they can only apply safely to files that carry the same names.
The rule is per upload. Next month’s snapshot can have a different set of columns from this month’s; it goes in its own upload, and Vintage tells you at Column Mapping which columns are new or missing compared with your earlier files. See Uploading month after month.
Every column needs its own heading
Section titled “Every column needs its own heading”A file with two columns of the same name, or a column with a blank heading, is refused at this stage. Vintage records your decisions by column name, so a column it cannot tell apart from another by name cannot be protected or mapped reliably. The message names the file and the problem. Give every column a unique, non-empty heading in your export and add the file again.
CSV and Excel (.xlsx) only
Section titled “CSV and Excel (.xlsx) only”Vintage reads two file types: CSV (.csv) and Excel workbooks (.xlsx). Anything else is refused when you choose it, with a message naming the two supported types.
The older Excel format, .xls, is common from core systems and is also refused. Vintage recognizes an older workbook even when it has been renamed to .xlsx or .csv. When that is what was refused, the message adds the remedy: open the file and save it as .xlsx, or export it as CSV.
Files with nothing in them
Section titled “Files with nothing in them”Two kinds of file are refused at selection because there is nothing to upload:
- An empty file. Check that the export ran and saved before you add it again.
- A file with column headings and no data rows. This usually means the export was filtered to nothing, or the wrong sheet was saved.
If a file cannot be read at all, for a reason Vintage does not recognize (a damaged or partly downloaded workbook, for instance), the message names that file and tells you that it and any files picked after it were not added. Check that the file opens on your computer and finished downloading or exporting, then add the missing files again.
Excel workbooks with more than one sheet
Section titled “Excel workbooks with more than one sheet”Vintage reads the first sheet only of an Excel workbook. When a workbook has more than one sheet, the staged file says which sheet was read and how many were left out:
This Excel file has 3 sheets. Only “Loans” was read; the other 2 sheets will not be uploaded. If your loans are on a different sheet, export that sheet on its own and upload it.
The same note appears again on the Review screen beside the row totals. A single-sheet workbook, or a CSV, carries no note.
CSV or Excel: which to send
Section titled “CSV or Excel: which to send”Excel files are read whole in your browser before anything is sent, so a large workbook is slow to read and heavy on the browser. CSV files stream in and are much faster. When you add an Excel file, Vintage shows a dismissible note recommending CSV for a large portfolio, and a stronger one when the workbook looks very large. For a backfill, send CSV.
| CSV | Excel (.xlsx) | |
|---|---|---|
| Size per file | No limit of its own (see below) | Up to 50 MB |
| Files per upload | No limit | Up to 5 Excel files |
| Total per upload | No limit | Up to 100 MB of Excel |
| How it is read | Streamed | Read whole in the browser |
Crossing any Excel limit is refused when you add the file, with the same advice: export to CSV, or split the files across separate uploads.
A CSV file has no size cap of its own. It is bounded only by what Vintage can address in a single file, about 2 GB, and a file that would cross that is refused while it is being sent, with a message saying so, rather than after every byte has arrived. A monthly snapshot of a very large portfolio is far smaller than that.
A row that does not match the header
Section titled “A row that does not match the header”In a CSV, a value that contains a comma but is not wrapped in quotes (Smith, John) splits into two cells and pushes every later value one column to the right. Such a row cannot be trusted, so Vintage does not import it at all. This is counted rather than caught here: the number of skipped rows is stated at Column Mapping and again at Review, with the fix (correct the export and upload again). Excel files address cells by position, so their rows cannot shift this way. See Mapping columns.
Next: Remove PII
Section titled “Next: Remove PII”When your files are staged, press Continue. The next stage, Remove PII, is where you choose the Loan ID and Tax ID columns and remove personal information, still entirely in your browser. See Removing personal information.