The Scenario
Your demand-gen team spent the last quarter collecting contacts from three different databases — a legacy CRM export, a LinkedIn campaign download, and a webinar attendee list. The merger landed in a Google Sheet with 600 rows. Some contacts appear two or three times under slightly different names, different email formats, or only a LinkedIn URL in one record and only an email in another. Before this goes anywhere near the new CRM, someone needs to identify and collapse the duplicates. That someone is apparently you.
The bad version:
- Run a VLOOKUP on email column to flag obvious email duplicates — catches about 200 of the 600 rows
- Try to fuzzy-match the remaining records by name and company using a formula — find that "Jon Smith at Acme" and "Jonathan Smith at Acme Inc" don't match and you have to review each one manually
- Spend two days on the dedup exercise and still hand off a sheet with 30 suspected duplicates you weren't sure about, annotated with "check this one"
The CRM migration is blocked on data quality. The data quality problem is blocked on you having enough hours in the week to review 600 rows by hand.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent that lives inside your Google Sheet. It reads the contact data across the rows and uses PDL's person identity resolution to assign a unique person ID to each contact — then flags rows that share the same ID as duplicates.
For each contact row in this sheet, identify the person via PDL using email in column B or LinkedIn in column C, write the PDL person ID to column D, then flag rows that share the same PDL ID as duplicates in column E.
What You Get
- Column D populated with a unique PDL person ID for each row that PDL can resolve
- Column E flagged with "Duplicate of row X" for any row whose PDL person ID matches another row — pointing you to the primary record
- Rows PDL cannot resolve left with a blank in column D and a note in column E so you know which contacts need manual review
- A clean, ID-annotated sheet you can hand to the CRM admin for the final merge
What If the Data Is Not Quite Ready
Some rows have no email and no LinkedIn — only name and company
For rows where both column B and column C are empty, attempt PDL identity resolution using first name in column A and company name from the database source tag in column F. Write the PDL person ID to column D if found, and flag unresolvable rows in column E as "Insufficient signals — manual review."
You want to see which source database each duplicate came from
For each contact, resolve via PDL and write the person ID to column D. Flag duplicate rows in column E and note which source database tab the primary record came from — CRM export, LinkedIn download, or webinar list — in column F.
You want to collapse duplicates into a single row automatically
For each contact row, resolve via PDL and write the person ID to column D. Group rows by person ID. For each group with more than one row, merge the records by keeping the most complete version of each field — prefer rows with a work email over rows without, and rows with a LinkedIn URL over rows without. Write the merged single row to a "Deduplicated" sheet.
Resolve identities, flag duplicates, and score each surviving contact by completeness in one pass
Resolve each contact via PDL identity resolution and write the person ID to column D. Flag duplicates in column E. For each non-duplicate row, score completeness in column F based on how many of these fields are populated: email, LinkedIn, company, title, phone. Scale 0 to 5.
One prompt handles identity resolution, deduplication, and quality scoring — the CRM admin gets a ready-to-import file, not a problem to untangle.
Try It
Get the 7-day free trial of SheetXAI and open any Google Sheet with a merged contact list from multiple sources. Ask it to resolve each contact via PDL identity resolution and flag the duplicates. Then see how to enrich the surviving contacts with job title and company data or go back to the People Data Labs overview.