Duplicate Leads in Google Sheets? Define the Record Before Removing Rows

Your lead sheet contains the same email address twice. Is that a repeated import, a second request from the same customer, or two records that need a relationship rather than deletion? The cleanup button cannot make that business decision for you.

Before asking AI to identify duplicates or removing spreadsheet rows, define what each row represents. Keep the original export and review a clearly named working copy. This checklist is about preparing a trustworthy cleanup decision, not automatically merging CRM contacts or contacting leads.

Decide whether a row represents a person or an inquiry

A contact list and an inquiry log answer different questions. A contact list might aim to keep one entry per person, while an inquiry log may need several requests from that person. Do not apply one matching rule to both without checking the intended use.

Write a plain-language rule before sorting or deleting anything. For example: “Flag repeated rows with the same source inquiry ID for review; preserve different inquiry IDs even when the email matches.” That is a proposed rule, not a universal requirement. Your source system and business owner must confirm which identifier is reliable.

If no stable reference exists, mark likely matches as candidates. A shared name or address is not enough to invent a missing ID or decide that one inquiry never happened. Keep uncertainty visible so a reviewer can resolve it from authorized source records.

Compare candidate matches on a review copy

Use a small, understandable group first. Compare source reference, request date, requested service and current follow-up status. Preserve the original values rather than asking AI to silently normalize them and then delete the apparent repeats.

Here is a fictional review set, not a customer dataset or a test we performed:

An email-only match groups all three rows. The different reference and request on the third row are reasons to investigate before discarding it. Even the first two rows need a check for differences in notes or status before choosing what to keep.

AI can prepare a comparison table from an approved synthetic sample or authorized dataset. Ask it to explain the fields behind each proposed match and flag conflicts. Do not treat its confidence or fluent explanation as evidence that two records are the same real-world event.

Inspect the selected range and matching columns

Google’s Sheets cleanup documentation describes Data → Data cleanup → Remove duplicates, with choices for the included columns and whether the range has headers. Inspect those choices on your working copy before confirming a removal.

The selected columns matter: a match on email alone is a different comparison from a match that includes an inquiry reference. Keep all relevant fields together in the review range, and ensure the reviewer understands which columns the operation will compare.

Google notes that identical values can count as duplicates despite differences in letter case, formatting or formulas. It also warns that selecting a single cell within a range can apply duplicate removal to the whole range. Do not assume that a highlighted cell means only that cell’s record will be affected.

If the selection, headers or matching criteria are unclear, stop before confirming. Use the approved synthetic example to make the expected result explicit. This guide does not authorize deleting live records, changing CRM matching rules or relaxing suppression controls.

Reconcile the kept records before any downstream import

After an approved cleanup on the working copy, compare the result with the expected review decisions. Did both distinct inquiry references survive? Were any notes, owners or follow-up states lost? A smaller row count is not, by itself, a passing result.

Keep a concise change record: original source, reviewed group, decision, retained reference and unresolved conflicts. Do not count a contact as reviewed or a lead as answered merely because a duplicate-looking row was removed.

Before using the result elsewhere, check the next system’s own import and matching behavior. A clean spreadsheet does not prove that a CRM will update the intended record or avoid creating another one. Keep the import separate until its destination, authority and acceptance check are confirmed.

For a handoff, use specific wording such as: “Repeated WEB-201 rows reviewed; WEB-248 retained as a separate request; conflicting notes remain unresolved.” This gives the next owner a decision they can inspect instead of a blanket claim that the list is deduplicated.

For the next-step process after a valid inquiry is identified, see our lead follow-up guide. If you need help connecting source records, review decisions and CRM imports, contact Business Boomer about a scoped workflow review. Start with a synthetic set whose expected kept records you can name before running cleanup.