How to Review Duplicate Businesses in a Lead CSV

Example records: one repeated location and one separate branch

The downloadable exercise contains three fictional records. The business name, branch labels and reserved .example website are illustrations, not collected leads or actual customer data.

Record A: Illustrative Studio, Example branch A, studio.example; first search.
Record B: Illustrative Studio, Example branch A, studio.example; overlapping search.
Record C: Illustrative Studio, Example branch B, studio.example; a different location.
Review decision: A and B are candidates for a merge after checking the source. Keep C separate if your work is organized by branch.

Keep originals and add a source reference

Preserve each downloaded file before editing it. In your working spreadsheet, add the source filename, research date and a row reference. That lets you trace a decision back to its original record. Keep any existing status or contact preference attached to the business when combining information; a cleaner spreadsheet should not erase earlier research.

Compare identity fields together

Start with the business name, full address and official website. A shared website can belong to a chain with multiple branches, while similar names can belong to unrelated businesses. Differences in capitalization or spacing are useful review clues, but do not prove that records should be merged. Check the source website or official branch listing when the identity is uncertain.

Same name and same location: review as a possible repeated listing.
Same name but different address: check for separate branches.
Different names but a shared website: investigate the relationship before merging.

Handle missing values as unknowns

Leave absent phone numbers or emails blank. Two blank phone fields are not evidence that rows represent the same location. If one version has more contact information, confirm that it belongs to the same business and branch before moving it into the retained row. Record where and when you checked it instead of treating the more complete row as automatically correct.

Review missing or published business email details

Choose and record the action for each group

Use three simple outcomes in your own review column: keep separate, merge after verification, or needs review. In the fictional exercise, A and B describe the same example location, so they are merge candidates. C has a different branch label and should remain a separate record for branch-level work. If the evidence is incomplete, leave the group unresolved rather than deleting it to reduce the row count.

Merge details without losing research history

For a verified duplicate, retain one working row and carry over checked details, source references and useful notes. Preserve the original files so you can undo an incorrect decision. Resolve conflicting statuses or contact preferences before planning follow-up. If you move the file into a CRM, check that system’s own matching rules; it may group records differently from your spreadsheet.

Check the output before importing it

Compare the number of original rows with the retained records and documented merge decisions. Spot-check a retained branch and a merged pair. Make sure phone formatting, column order and source references survived. Test a small import before uploading the full file and inspect the resulting records. A lower row count is useful only when the right business locations and research details remain.

Check the export and CRM import workflow