Data cleaning for CRM imports
Salesforce
How to clean a CSV before importing into Salesforce
To clean a CSV for Salesforce: open it in Sift, approve the suggested fixes, dedupe by email, normalize phones and countries, pick the built-in Salesforce template in Match & check to map your columns, and run the readiness check. If contacts need parent accounts, generate a deduped accounts file and import it first. Sift runs entirely in your browser, so the file is never uploaded.
Why Salesforce imports go wrong
Most failed or messy Salesforce imports come down to a handful of issues:
- Contacts without accounts, Salesforce links each contact to an Account; importing contacts whose company doesn't exist yet breaks the link.
- Duplicate leads and contacts, the same person twice trips matching rules or creates two records.
- Last Name is required, a blank one rejects the row outright.
- Picklist mismatches, a Lead Source or Industry value that isn't a valid picklist option.
- Format noise, inconsistent phones, countries, dates, casing and whitespace.
Step by step
- Load the CSV into Sift. It profiles every column in your browser.
- Clean: approve trim, casing, fix-emails, normalize-phones, and standardize-countries, each with a before/after diff.
- Dedupe by email (exact or fuzzy) so Salesforce matching rules don't fight your import.
- Generate accounts: in Export, build a deduped accounts file from your contacts and import it first (Salesforce needs the Account to exist before the Contact).
- Map columns: pick the built-in Salesforce · Contacts or Salesforce · Leads template in Match & check.
- Check readiness for blank required fields and wrong formats, then export and import via the Data Import Wizard.
Tip: import the accounts file first (Setup → Data Import Wizard → Accounts), then the contacts file. Sift rewrites each contact's company to the canonical account name so the lookup resolves on import.