Sift Data cleaning for CRM imports

Sales exports

Clean an Apollo, ZoomInfo, or Sales Navigator export before a CRM import

Exports from Apollo, ZoomInfo, and Sales Navigator are prospecting data, not import-ready files. They carry ALL-CAPS names, "Inc" and "LLC" suffixes on company names, job-title clutter, name columns that are either split or joined the wrong way, and CSV quoting errors that make an importer choke on line 1. Whole paid tools exist just to tidy this up. Sift cleans the format for free, in your browser, with nothing uploaded.

Clean your export in Sift, free →

What comes out of these tools

Apollo throws verbatim errors like "Most rows need a valid full name" and "Illegal quoting in line 1" when the file has a single combined name column or unbalanced quotes. Its field mapping is case-sensitive. ZoomInfo warns you to "expect a 15-25% bounce rate on any export" and its own reviews flag "Outdated Contacts". Sales Navigator hands you names in caps and a full name jammed into one column. There is a reason paid cleaners like Evaboot exist solely to strip "Inc." and "LLC" from company names, format job titles, and split full names.

What the export gives youWhat a CRM import needs
JANE DOE→Jane Doe
"Illegal quoting in line 1"→Re-quoted cleanly on export
ACME INC→Acme Inc
Full name in one column→First name and Last name

The manual way, honestly

You can do most of this by hand. Proper-case the names with a PROPER() formula, then go back and fix the ones it broke (Mcdonald, mismatched acronyms). Find-and-replace the "Inc" and "LLC" casing. Use Text to Columns to split the full-name column, then clean up every row where it split on the wrong space. Chase the quoting error to whichever row has a rogue comma or quote. Or you pay for a dedicated tool that does the caps, suffixes, and name-splitting for you. Both routes work; both take time, and the formula fixes un-fix themselves the next time the CSV is reopened in Excel.

Clean it in Sift

  1. Load the export into Sift. It profiles every column in your browser; nothing is uploaded.
  2. Fix the casing with name-safe rules: ALL-CAPS names become proper case, acronyms like CTO stay uppercase, McDonald and O'Brien are preserved, and legal suffixes are title-cased.
  3. Split the full-name column on a delimiter into First name and Last name, reviewed with a before/after diff so you catch the rows that split wrong.
  4. Re-parse the file with a robust CSV reader, so the "illegal quoting in line 1" error is fixed and the file writes back with clean quoting on export.
  5. Dedupe contacts (exact or fuzzy) and merge duplicates into one golden record instead of deleting rows.
  6. Normalize phones to an E.164-style format, and standardize countries and postcodes.
  7. Map to your CRM template and run the import-readiness check for required fields, types, and allowed values.
  8. Export a clean CSV and import it.
Privacy note: Sift is a static web app with no backend. Your export is processed entirely on your device, which you can verify by disconnecting your internet after the page loads; the cleaning still works. The opposite of pasting a prospect list into a web tool or ChatGPT.

What Sift can't fix

This is the honest limit. ZoomInfo and similar tools warn of a 15-25% bounce rate because contacts go stale, people change jobs, and mailboxes get deprovisioned. Sift fixes the format of an address, whether "Jane.Doe @Acme.com " is trimmed and cased correctly, but it does not verify deliverability and it does not refresh outdated data. If an address is well-formed but the person left the company, Sift can't know that. For bounce risk, run the cleaned file through an email-verification service before you send. Sift gets the file into shape; a verifier tells you which addresses are still live.

Related guides

Clean trade show leads before a CRM import Split a full-name column into first and last Email verification vs list cleaning: which one you need Dedupe contacts into a golden record All guides