Importing data: CSV upload and column mapping
The Setup > Data Import screen: pasting from Excel or uploading a CSV, column mapping, preview and the mistakes people make most.
Where is the screen?#
Import lives under Setup > Data Import and is open to admins only. The flow is always the same: you supply the data, the system maps the columns, Preview shows you what will happen, then you press Import. No record is created before you have seen the preview.
Two ways in: paste or pick a file#
- Paste: select and copy your data in Excel, then paste it into the text box. The first row must be the column headers.
- Pick a file: upload a
.csv,.tsvor.txtfile from the file field. Files are read as UTF-8.
The separator is detected for you: tab, semicolon or comma. Quoted values are understood too.
Action: Insert or Update#
| Action | What it does |
|---|---|
| Insert | Creates a new record from every row |
| Update (by Id) | Updates existing records; the file must contain an Id column |
The easiest way to get the Ids you need is a round trip: export the list first, keep the Id column, then upload it back together with the fields you corrected.
Which objects?#
The object list holds the standard objects licensed in your org (Account, Contact, Lead, Opportunity, Case, Task and the objects of your active apps) plus the custom objects you defined in the Object Manager. Objects of a disabled app do not appear.
Column mapping#
Your headers can be in any of three shapes:
- the field name:
name,amount,email - a familiar API name:
Name,StageName,AnnualRevenue - the field label as shown in the interface
Custom fields you added in the Object Manager map by their key or label as well. Columns that link to a company (Account) are matched by the account's name, which must match exactly. The preview prints what each header mapped to: mapped columns become green chips, unmapped ones red. Unmapped columns are simply ignored and do not block the import.
Preview and row statuses#
| Status | Meaning |
|---|---|
| OK | The row will be imported |
| Required field missing | One of the object's required fields is empty (a name on an Account, for example) |
| Id not found | In update mode, the given Id does not exist in the org |
| Validation | One of your validation rules rejected the row |
| Duplicate | Your duplicate rule says this record already exists |
Duplicate rows are skipped by default. To bring them in deliberately, tick Import duplicates too. The number on the button tells you how many rows will actually be written.
The mistakes people make most#
| Symptom | Fix |
|---|---|
| Dates come in empty | Write dates as YYYY-MM-DD (2026-09-15). In Excel, set the column to Text before saving. |
| Picklist values do not stick | Values must match your org's picklist exactly. Copy the correct spelling from Object Manager > Picklists first. |
| The leading zero of a phone number disappears | Format the phone column as Text in Excel; in a number format Excel drops the leading zero on its side. |
| Amounts are wrong | Currency symbols and thousand separators are stripped; use a dot for decimals. |
| Records do not attach to their company | The account name must match letter for letter; import accounts first and contacts afterwards. |
ohana360.com/blog/excel-customer-tracking. Its columns are named to suit the import.Exporting#
Every list view has an Export menu offering two formats: Excel (.xlsx) and CSV (.csv). The file takes the filters and columns currently on screen. To download your org's entire data in one package, use the full data export on the compliance screen.
After the import#
What to do with the data you imported (merging duplicates, the recycle bin, backups) is covered in the data management article.