Data cleaning for CRM imports
Event leads
How to clean trade show leads before importing to your CRM
Leads contacted within 48 hours of a show are 5x more likely to convert, yet exhibitors routinely spend two weeks sorting through them first. The reason is the badge-scan export: names come back in ALL-CAPS, fields like company or email are missing, and the same person gets scanned at two booths and lands in the file twice. Here is how to clean that list fast, so you are following up inside the window instead of untangling a spreadsheet.
Clean your lead list in Sift, free →
The manual way (fix by hand, or let it roll)
The honest options once the show ends are laid out plainly in one HubSpot-agency guide to trade-show imports: "Fix upper/lower case by hand, de-dupe email addresses by hand... Or just let it roll." Fixing by hand is where the two weeks go, one ALL-CAPS name and one duplicate scan at a time. Letting it roll is faster, but it imports the mess straight into your CRM, where every duplicate and every "JOHN SMITH" now lives permanently. There is a third option: clean the whole list in one deterministic pass before it ever reaches the CRM.
Clean the list in Sift
- Load the badge-scan CSV (or Excel export) into Sift. It profiles every column in your browser; nothing is uploaded.
- Fix casing and trim stray spaces, so "JOHN SMITH" and " john " both become "John Smith", each with a before/after diff you approve.
- Dedupe exact and fuzzy across scanners, merging the same person scanned at two booths into one golden record with survivorship rules, so the phone from one scan and the company from another both survive instead of one row being deleted.
- Normalize phones and countries to a consistent E.164-style and standard country names.
- Tag the whole list with a constant column for the event name and lead source, so every record carries where it came from.
- Map to your CRM template (HubSpot, Salesforce, Pipedrive, Dynamics 365, Zoho) and run the import-readiness pre-flight: required fields, types, and allowed values, flagged before you import.
- Export a clean file in minutes, not two weeks, and import it while the 48-hour window is still open.
What Sift can't fix
Sift is deterministic, not magic. A badge scan that never captured an email or a company leaves a genuinely blank field, and Sift will not invent one. There is no enrichment, no lookup, and no AI guessing. What Sift does instead is flag every empty required field to the needs-your-eyes list, so you know exactly which records to chase or set aside, rather than discovering the gap after a failed import. It also does not verify that an email address actually delivers; it can standardize and dedupe what the scanner captured, but confirming a mailbox is a separate step.