Data cleaning for CRM imports

Countries

Standardize country names for a CRM import ("USA" is not a valid Country)

CRMs with a Country picklist reject "USA", "US", or "America" when the allowed value is "United States", so the import fails on rows that look perfectly fine. The fix is to standardize every spelling to the exact value your CRM expects, before you import. Sift does this in your browser: it canonicalizes the common country aliases, lets you value-map any variant to your CRM's allowed value, and flags mismatches per row. The file is never uploaded.

Standardize your countries in Sift, free →

What goes wrong

Each of these rows looks correct to a human, but the picklist only accepts one exact string. Anything else is rejected or imported blank, and a 5,000-row list can fail on hundreds of rows that differ only in spelling.

The manual fix

1. Build a find-and-replace mapping of every spelling you see ("USA", "US", "U.S.", "America") to the one value your CRM allows.
2. Watch the case and punctuation: "usa", "U.S.A.", and "U S A" are all different strings to a plain find-and-replace, and it's easy to miss one.
3. Repeat for every country in the file, not just the United States, because "UK", "Britain", and "England" have the same problem.
4. Redo the whole mapping the next time you import, because the next export brings the same messy variants back.

The Sift fix

1. Drop your CSV or Excel file into Sift. It runs in your browser; the file is never uploaded.
2. Sift canonicalizes the common country aliases automatically: USA, US, U.S., America, and United States of America all become United States; CA and Canada become Canada, and so on.
3. Value-map any spelling that's specific to your list to your CRM's exact allowed value, so nothing is left to guess.
4. Check against the picklist: map your file to your CRM's Country template and Sift compares the column to the allowed values, flagging every mismatch per row before you import.
5. Review each change with a before/after diff, then export a clean CSV ready to import.

What Sift can't fix

Honesty matters here: every CRM picklist has its own exact required spelling, so you still need to map to your target's allowed value. Sift standardizes the common aliases and any variants you configure, but it will not guess an ambiguous or obscure place name. If a value could mean more than one country, or it's a place Sift doesn't recognize, it flags the row for you rather than picking a spelling on your behalf.

Related guides