← All practice problems

Fuzzy contact dedupe by normalizing names

working · sql🔒Walkthroughs

NEW

Fuzzy contact dedupe by normalizing names

Group contacts whose full_name is identical after this normalization pipeline: 1. Lowercase. 2. Trim leading/trailing whitespace. 3. Collapse internal multi-space runs to a single space.

For each normalization group, keep the canonical contact (the one with the SMALLEST contact_id).

Columns: canonical_id, norm_name. Sort by norm_name ascending.

Note: This does NOT do edit-distance matching — "Jon" and "John" stay separate groups. True fuzzy matching is its own rabbit hole.

Tables & sample data

contacts

Loading preview…

query.sql

Loading…

Loading editor…