The Scenario
It's your third week running sales ops cleanup. Someone merged two lead lists into an Excel workbook — 800 rows where column A has names pulled from Salesforce and column B has names from a purchased lead list — and column C is supposed to say whether each pair is the same person. Nobody built that. The deduplication task was assigned three months ago and marked done without being done.
The bad version:
- Write a fuzzy-match formula that checks for exact and partial string overlap, realize it flags "Jonathan Smith" vs "Jon Smith" as different and "David Miller" vs "David Miller Jr." as the same
- Manually review the 200 edge cases row by row, find that your judgment for row 50 contradicts your judgment for row 150 because the cases look subtly different
- Export to a Python fuzzy-matching library, get results that conflict with the formula results, now you have two competing verdicts and neither is authoritative
You're supposed to be building the routing logic, not reconciling two broken dedup attempts against each other.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Excel workbook. It uses the Tisane entity comparison tool to evaluate each name pair and write a verdict directly back into your workbook.
Use Tisane to compare person names in 'CRM Name' vs 'Import Name' columns and add a 'Match' column with a similarity verdict and any identified differences
What You Get
- A 'Match' column filled with one of three values: "Same", "Different", or "Uncertain"
- "Uncertain" is used for cases where Tisane detects partial overlap — a nickname vs. a full name, or a name with a suffix — rather than forcing a wrong binary answer
- Rows where either column is blank get a blank in 'Match', not an error
What If the Data Is Not Quite Ready
The name columns have inconsistent formatting — extra spaces, all-caps entries, mixed case
Before comparing names, normalize both the 'CRM Name' and 'Import Name' columns by trimming whitespace and converting to title case, then run Tisane entity comparison and write the verdict into 'Match'
Only rows marked 'Unverified' in a status column need comparison
For rows where the 'Status' column says 'Unverified', compare the names in 'CRM Name' and 'Import Name' using Tisane entity comparison and write the verdict into 'Match' — leave all other rows untouched
You need a brief explanation for 'Uncertain' rows
Compare names in 'CRM Name' vs 'Import Name' for all 800 rows using Tisane, write the verdict into 'Match', and for any row where 'Match' is 'Uncertain' add a short note in a 'Match Notes' column describing the identified difference
Full pipeline: compare, flag uncertain rows, and move them to a review worksheet
Compare names in both columns using Tisane for all 800 rows, write the verdict into 'Match', then copy all rows where 'Match' is 'Uncertain' to a new worksheet called 'Needs Review' for manual follow-up — sorted alphabetically by 'CRM Name'
Comparison, triage, and routing in a single prompt.
Try It
Get the 7-day free trial of SheetXAI and open any Excel workbook with two name columns you need to reconcile, then ask it to run entity comparison across all rows. To continue cleaning your lead data, see how to detect language per row or strip HTML from a text column. The full Tisane overview is at the hub.