CSV Import¶

Overview¶

CSV import lets you create records from CSV files directly in the console. You can import into an existing collection or create a new collection from the CSV headers and data.

Starting an Import¶

CSV import is available from two places:

• Collections list — click the dropdown menu and select "Import CSV"
• Collection records page — click the dropdown menu and select "Import CSV"

When starting from the collections list, you can choose to create a new collection or import into an existing one. When starting from a collection's records page, the import targets that collection.

Import Wizard¶

Step 1: Upload¶

Drag and drop a .csv file onto the upload zone, or click to browse. Files must be CSV format, up to 10 MB.

Step 2: Preview¶

The preview shows:

• Column headers with detected type badges (text, number, date, or boolean)
• First 50 rows of parsed data
• Row and column count summary
• Delimiter selector — auto-detected by default, but you can switch between comma, semicolon, or tab if the auto-detection is wrong

Warnings appear for:

• Duplicate column headers — automatically renamed with a suffix (e.g., "Name" becomes "Name_1", "Name_2")
• Row limit — imports are capped at 50,000 rows; excess rows are ignored with a warning

Step 3: Configure¶

Importing into an Existing Collection¶

A column mapping table shows each CSV column alongside a dropdown of collection properties. Columns are auto-mapped by matching header names (case-insensitive). You can adjust mappings or skip columns you don't need.

At least one column must be mapped to proceed.

Creating a New Collection¶

Enter a collection name (auto-populated from the filename). Each CSV column becomes a collection property with:

• Property name — editable text input, pre-filled from the CSV header
• Type — dropdown with Text, Number, Date, or Boolean, pre-filled from type inference

Type Inference¶

Types are inferred from the first 100 data rows:

Empty columns default to text.

Step 4: Importing¶

Records are sent to the backend in batches of 500. The screen shows:

• A progress bar with percentage
• A real-time counter: "{processed} of {total} records"
• A cancel button — you can stop the import mid-process and keep records that were already created

Step 5: Summary¶

After the import completes (or is cancelled), the summary shows:

• Imported — number of records successfully created
• Failed — number of records that failed validation
• Total — total records attempted

If any records failed, a scrollable table shows the first 50 failures with row number and error message. Expand a row to see field-level validation details.

Click Download Failed Rows to export all failures as a CSV file with __error and __row_number columns appended, so you can fix and re-import them.

Limits¶

Supported Delimiters¶

CSV files are auto-detected for delimiter type. If auto-detection is wrong, you can manually select:

• Comma (,)
• Semicolon (;)
• Tab

The delimiter can be changed in the preview step, which re-parses the file immediately.