Import records

Import a CSV or Excel file of up to 10,000 rows: map columns to fields, check every row before saving, add or update records and download failed rows.

Bring many records in at once from a CSV or Excel file — a customer list from your old software, this month's price list, or a stock count. You match the file's columns to fields, see every problem before anything is saved, and then AxisIQ adds or updates the records in the background.

Before you start: You need Can add & edit (or higher) on Business records, which includes importing. Updating records also needs permission to edit those records. Each row goes through the same rules as a record you add by hand.

Steps

  1. Open the record type's list, for example Customers.
  2. Click More › Import at the top of the list. The dialog is titled Import customers. The Import dialog
  3. Under File, choose a CSV or Excel file (.csv, .xlsx, .xlsm) and click Next.
  4. AxisIQ reads the file and suggests a field for each column, matching on the field's key, its label and common names (for example Mobile → Phone). Check each row of the table:
    • Column in file and Example values show what is in the file.
    • Field is where it goes. Choose Don't import to leave a column out. Each field can take only one column. Required fields have a *.
  5. If the Excel file has more than one sheet, pick it under Sheet.
  6. Under What to do, choose one:
Choice What happens
Add every row as a new record Each row becomes a new record.
Update matching records, add the rest A row that matches an existing record updates it. A row with no match becomes a new record.
Only update matching records A row that matches an existing record updates it. A row with no match fails with "no record matches".
  1. For the two update choices, pick Match rows on: a mapped field (unique fields are listed first and marked (unique)), or Record id when the file has a column headed id — for example a file you exported earlier.
  2. Read the check under the table: 48 ready and, if any, 2 with problems, with a line per problem row such as Row 7: Email: not a valid email address. The check runs again every time you change the mapping. Fix the file and upload it again, or carry on and skip those rows.
  3. Click Import 50 rows (or Import 48 rows (skip 2)).

The dialog shows progress, then 45 added · 3 updated · 2 failed. You can close the dialog while it runs; the import carries on. When it finishes, click Download failed rows to get a CSV of the rows that failed, exactly as you uploaded them, with an error column saying why. Fix them in a spreadsheet and import that file.

Tip: If required fields have no column, the dialog says "Map a column to: … — they are required." and Import stays greyed out. Fields that fill themselves in (an auto-number or a default value) are not counted as missing.

File rules

  • Up to 10,000 data rows and 10 MB per file. Split a bigger file into parts.
  • AxisIQ finds the header row by itself. The line above the table shows what it found, for example customers.xlsx · header on row 3 · 50 rows.
  • Column headings can be labels (Unit price) or keys (unit_price); you choose the field either way.
  • Rows are saved one by one. A bad row fails on its own and never stops the others. There is no all-or-nothing rollback, and no undo — but imported records can be moved to the trash like any other.
  • Duplicate values in unique fields and permission problems are found only as rows are saved, so a row that passed the check can still fail. It then appears in the failed rows.
  • Row numbers in messages count data rows from 1, not counting the header.

How each cell is read

Field type What to put in the cell
string, text, email, url, file The text
number A plain number such as 1250.50
currency 1250.50 INR (amount, space, 3-letter code) or just 1250.50. A missing code uses the field's locked currency, or the workspace default.
bool true or false
date 2026-10-14, or day first: 14/10/2026, 14-10-2026, 14.10.2026, 14 Oct 2026. Excel date cells work as they are.
datetime An RFC 3339 time such as 2026-10-04T09:30:00Z, or a date as above
select One of the field's options, spelled exactly as saved
multiselect A list such as ["red","blue"] or `red
ref The linked record's ID (a UUID), not its name. Use Copy ID in the row menu, or export the linked type.
refs A list of IDs, as ["id1","id2"] or `id1

Note: Dates are read day first, as is usual in India. 03/04/2026 is 3 April 2026. A file with month-first dates should use the 2026-04-03 form.

What happens to empty cells and special fields

  • An empty cell is skipped. On a new record the field's default applies; a required field with no value fails that row. On an update the field keeps its current value — an import never clears a field.
  • Auto-numbered fields get the next number when the cell is empty or the column is not mapped. A number you supply is kept.
  • Computed fields cannot be mapped. They are worked out for you.
  • Every field rule applies, including unique and patterns.
  • Cells that begin with ' followed by =, +, -, @ are read without the leading '. AxisIQ adds this mark when it exports to keep spreadsheets from running a cell as a formula, so a round trip gives you your original value back.
  • Flows, notifications and the activity trail see imported records like any other new or changed record.

Troubleshooting

The file could not be read The file is not a CSV or Excel file, is larger than 10 MB or 10,000 rows, or has no data row. Check it and try again.

Row 3: <field>: must be one of the configured options The cell does not match one of the select options. Match the spelling exactly.

Row 5: a unique field already holds this value That value is already used by another record, or earlier in the same file. To change the existing record instead, choose Update matching records, add the rest and match on that field.

Row 9: several records match "…" More than one record has that value in the match field, so AxisIQ will not guess. Match on a unique field or on Record id.

Row 12: no record matches With Only update matching records, the row's match value was not found. Use Update matching records, add the rest to add it instead.