Skip to main content

SF-0218 · Scenario · Easy

What are the steps to UPDATE the data in salesforce using the Data loader?

✓ Verified by Vikas Singhal · Last reviewed 5/17/2026 · Updated for Spring '26

An Update operation modifies existing records. The single rule that distinguishes it from Insert: every row must carry the 18-character Salesforce Id so Data Loader knows which record to change.

Step-by-step

  1. Prepare the CSV. First column is usually Id (the record’s Salesforce ID). Add only the fields you want to change — fields you omit from the CSV are left untouched. UTF-8 saves a lot of pain with apostrophes and accents.
  2. Launch Data Loader and click Update.
  3. Log in to production (login.salesforce.com) or sandbox (test.salesforce.com).
  4. Pick the target object — the API name has to match (Account, Contact, My_Object__c).
  5. Select the CSV file. Data Loader previews the first rows.
  6. Map columns to fields. Id must map to Id. Map every other column you want to change. Click Auto-Match if your header names already match the API names. Save the mapping as .sdl for reuse.
  7. Pick the success/error output directory.
  8. Click Finish. The job runs row-by-row in batches.
  9. Review the outputs.
    • success_<timestamp>.csv — every row that updated, with the Id echoed back.
    • error_<timestamp>.csv — failures with an error column. The first column always tells you why.

How to get the Ids in the first place

The standard workflow is Export → edit → Update:

  1. Run an Export with a SOQL query that grabs Id plus the fields you want to change.
  2. Open the result in Excel, change the values.
  3. Save and run Update against that same file.

This pattern is the bread and butter of admin work — bulk-changing record owners, normalising picklist values, fixing typos across 50,000 records.

Things to verify before clicking Finish

  • Are required fields covered? Update doesn’t need required fields if they already have values, but if you blank one out you’ll get REQUIRED_FIELD_MISSING.
  • Are you running in the right org? A common production mistake — log in twice if you’re not sure.
  • Are triggers/validation rules going to block you? If so, plan accordingly: either bypass them through a “Loader” custom permission your code respects, or fix the data first.
  • Bulk API vs SOAP API? Tick Use Bulk API under Settings → Settings for jobs over ~10,000 rows.

Trap: blank cells vs nulls

A blank cell in your CSV is ignored — the field stays at its existing value. To explicitly null out a field you need to tick Insert Null Values under Settings → Settings. Without that flag your “delete this field” rows do nothing.

Verified against: Data Loader Guide — Updating Data, Metadata API Developer Guide. Last reviewed 2026-05-17 for Spring ‘26 release.