Data Loader
A Salesforce data loader that runs in your browser
No desktop install and no Java. Paste rows straight from Excel or Google Sheets, map your columns to fields, see exactly what will happen, and run insert, update, upsert, or delete with per-row error reporting.
Data loader
Insert, update, upsert, or delete records from pasted spreadsheet data.
- OperationUpdate · Account
- Data412 rows
- 3Mapping5 of 6 mapped
- 4Review & runBatch & monitor
| Pasted column | Target field | Type | ||
|---|---|---|---|---|
| Account Id e.g. 0015f00000AbCdEAAV | Id | id | ||
| Account Name e.g. NORTHWIND HOLDINGS PTY LTD | Name | stringrequired | ||
| New Owner Email e.g. [email protected] | Owner.Email (related by external id) | lookup by external id | ||
| Region e.g. NSW / ACT | Region__c | picklist | ||
| Segment e.g. Broker — Tier 2 | Segment__c | picklist | ||
| Notes e.g. moved from Kerr, Mar 2026 | — Skip this column — |
Mapping looks good — every check passed.
What you can do
- Four operations: insert, update (matched by Id), upsert (matched on an external ID field you choose), and delete
- Paste tab-separated rows straight from Excel or Google Sheets, paste raw CSV, or upload a .csv / .tsv file
- A four-step wizard (Operation → Data → Mapping → Review & run) so nothing runs before you have seen it
- Column mapping with a skip option, plus relate-by-external-ID for lookup fields
- Validation flags missing required fields before the run can start
- The object picker only offers objects your Salesforce user can actually write to
- Results stream back with per-row errors, so you fix the eleven bad rows and keep the rest
- Up to 10,000 records per run, every one audit-logged
How it works
Choose the operation
Insert, update, upsert, or delete, and the object, filtered to what your user can write.
Bring your rows
Paste from a spreadsheet, paste CSV, or upload a file. No formatting ritual required.
Map and validate
Match columns to fields, relate lookups by external ID, and let validation catch missing required fields first.
Review, then run
The review step spells out exactly what will happen. Then results stream in row by row.
Why it's different
No install, and it shows you first
The classic Data Loader is a desktop app with a Java dependency and a mapping file. This is a browser tab: paste, map, review, run. The review step is the point. You see the operation, the object, the row count, and the mapping before a single record moves. Writes require owner or admin, and production DML stays switched off unless your deployment explicitly enables it, so sandboxes are the default playground.
Try asking
- Bulk-update 400 account owners from a spreadsheet the sales director sent
- Upsert a product catalogue keyed on an external ID
- Clean up test records in a sandbox after a demo
- Load a one-off migration without installing anything
- Fix the twelve rows that failed, without re-running the other nine hundred
Data Loader — questions
Works even better with
Data Tables
A spreadsheet over live records, parent fields included.
Learn moreDuplicate Manager
Find the same customer entered four times, see both records side by side, and merge only what a person approved.
Learn moreData Seeding
Real data into sandboxes with PII masked on the way in. Refuses production outright.
Learn moreRead next
- Salesforce sandbox seeding with PII masking
A fresh sandbox is empty and useless. Copying production data wholesale is a privacy incident waiting to happen. The path between the two.
Put Data Loader to work on your org
Every AI feature on your own provider key, from A$19.95 per user per month.
- 14-day free trial
- No credit card required
- Cancel any time