# Duplicate Candidate Records: How Recruitment Agencies Can Prevent and Merge Them
**URL Slug:** `/recruiting/duplicate-candidate-records-recruitment-database`  
**Primary Keyword:** duplicate candidate records  
**Related Keywords:** prevent duplicate candidates recruitment database, merge duplicate candidate profiles, candidate deduplication ATS, duplicate CV records, candidate identity resolution  
**Meta Description:** "Duplicate candidate records waste recruiter time and corrupt reporting. Here's how to prevent, detect and merge duplicates without losing candidate history."

The same candidate arrives three times: once from a job board, once from a LinkedIn scrape, once from a referral. Three records, three email addresses, three inconsistent histories. The recruiter contacts them twice, the client sees two submissions, and the reporting says the agency has three candidates when it has one.

Duplicate records are not a candidate behaviour problem. They are a **data model problem**.

The direct answer: prevent duplicates with **layered identity matching at the point of entry**, detect them with **periodic scans**, and merge them with **full history retention** — not deletion.

## Why Duplicates Happen
- **Multiple intake channels:** job boards, LinkedIn, referrals, inbound email, CV uploads.
- **Email-only matching:** a candidate changes jobs and email addresses; the old record and the new record never connect.
- **Name variations:** "Katie" vs "Katherine," married names, transliterations.
- **No entry-level check:** the ATS creates a new record before checking for an existing one.
- **Recruiter workarounds:** a recruiter cannot find the existing record, so they create a new one.

## The Layered Identity Rule
Relying on email alone is the most common mistake. A more resilient hierarchy:
1. **Exact email match:** highest confidence.
2. **Name + phone:** strong confidence.
3. **Name + employment-history fingerprint:** matches candidates who changed email but have the same career history.
4. **Name + LinkedIn URL:** strong when present.
5. **Fuzzy name + location + employer:** for manual review only, never auto-merge.

Each level should carry a confidence score. High-confidence matches can auto-merge; medium-confidence matches go to a review queue; low-confidence matches are flagged but not merged.

## Prevent at Entry, Not After the Fact
The cheapest duplicate is the one that never gets created. At the point of CV upload, inbound email or manual entry, the system should:
- Run the identity check **before** creating the record.
- Show the recruiter the potential match and the existing history.
- Offer "add to existing record" as the default action.
- Only create a new record if the recruiter confirms it is a different person.

This is the pattern used by agency portals that check phone number or email against existing candidates before allowing submission.

## Merge Without Losing History
A merge should produce **one surviving record** that keeps:
- Every communication (email, SMS, call notes).
- Every vacancy and submission.
- Every document and CV version.
- The earliest source and the most recent consent.
- A merge audit trail.

Merging by deleting the "losing" record destroys exactly the history the agency needs. The correct model is **merge, not delete**.

## The Scale of the Problem
Duplicate rates commonly run **18-25%** in UK and European databases and can reach **60-80%** in large global enterprise databases. At 20,000 records, a 20% duplicate rate means 4,000 wasted records — and an unknown number of duplicate candidate contacts.

## The Business Impact
- **Recruiter time:** no repeat outreach, no manual reconciliation.
- **Reporting integrity:** candidate counts, source attribution and pipeline metrics become trustworthy.
- **Client trust:** no duplicate submissions.
- **Candidate experience:** one consistent conversation.
- **Compliance:** consent and retention apply to one record, not three.

## How BILIVINE Fits
BILIVINE builds identity resolution and deduplication into the data layer that sits alongside the existing ATS — so duplicates are prevented at entry and existing duplicates can be merged with history intact. It is a prerequisite for database search and matching to work reliably.

**If your recruiters have stopped trusting database search results, duplicates are usually the reason. BILIVINE's free audit can quantify the duplicate rate and map a merge plan. [Explore recruiting automation →](https://bilivine.pages.dev/recruiting)**