How to deduplicate in Salesforce: a step-by-step guide

Duplicates get into Salesforce from web forms, imports, and manual entry. Reps end up calling the same lead twice, automations trigger off stale records, and reports stop matching what's actually in the pipeline.

Salesforce's built-in tools can find and merge duplicates without any extra software, and the first half of this guide walks through that setup. When volume, custom objects, or ongoing prevention push past what those native tools handle, the second half covers how Plauti takes over that work inside Salesforce, without exports or scripts.

Before you start, you'll need:

• Salesforce admin permission
• Access to Leads, Contacts, and Accounts, plus any custom objects you care about
• A sandbox for testing — strongly recommended before you touch production
• Optional: a small set of known duplicate records to use for a trial run

Find and merge duplicates with Salesforce's native tools

Every Salesforce org ships with duplicate detection and a merge tool built in. Here's how to turn them on and put them to work.

Step 1: Turn on Duplicate Rules

Salesforce's built-in duplicate detection has to be switched on before it does anything.

• Go to Setup and search for "Duplicate Rules."
• Activate the standard rules for Leads, Contacts, and Accounts.
• For finer control, create or tune Matching Rules — for example, match Contacts on email plus phone.

Combining fields (email, name, and company together) catches more true duplicates than matching on any single field.

Step 2: Decide whether to block or alert

This setting controls what happens when a user tries to save a duplicate through the Salesforce UI.

• Each Duplicate Rule can Block (refuse the save entirely) or Alert (warn the user, who can still override and save).
• Those settings apply only to records created directly in the Salesforce UI.

Records arriving through Web-to-Lead, the Data Import Wizard, Data Loader, or the API follow a different rule. Salesforce hard-blocks a detected duplicate outright: there's no alert, no override, and no separate configuration for these channels — blocking applies even if your rule is set to Alert in the UI. That has a real cost. If someone submits your web form twice, only the first submission gets created, and whatever changed between the two submissions is lost.

Step 3: Find existing duplicates

Next, track down the duplicates already sitting in your org.

• Go to the Leads, Contacts, or Accounts tab.
• Add the "Potential Duplicates" column to a list view — it appears once rules are active.
• Open a flagged record to see its suspected matches side by side.

Step 4: Merge duplicates

Consolidate the duplicates into a single record.

• From a record with potential duplicates, choose Merge, or use the Merge tool from a list view.
• Select up to three records to merge at once — that's the native limit.
• Choose the master record and pick which field values to keep.
• Confirm the merge.

Salesforce can often restore a deleted record after a merge, but double-check before you confirm one. Prevention is cheaper than cleanup.

Where native tools fall short

• Fuzzy matching is limited: variants like "Jon" and "John," or "ABC Corp" and "A.B.C. Corporation," can slip past the rules.
• Merging tops out at three records at a time, so working through a large backlog means doing it record by record.
• Custom objects and cross-object matching need heavy configuration, or aren't supported at all.
• There's no dashboard for duplicate trends — tracking data hygiene over time means building your own reporting.
• Keeping rules current and reviewing alerts is ongoing work for whoever administers the org.
• There are no advanced merge rules: every field value has to be chosen by hand.

Scale deduplication beyond Salesforce's native limits

Plauti brings the same idea — find, merge, and prevent duplicates — to a scale native tools weren't built for.

What it adds

• Exact, fuzzy, and phonetic matching across fields, objects, and locales.
• Bulk merges of hundreds or thousands of records in minutes instead of days.
• Support for standard and custom objects across the org.
• An audit trail for every merge, with a confidence score to review before approving.
• Dashboards that track duplicate trends and data health over time.
• Runs entirely inside Salesforce — nothing gets exported to merge it.

Set it up in about 20 minutes

Test it first: install Plauti in a sandbox, run a scan on a small sample (a Lead segment works well), and review the matches and merge scenarios before touching production.

From there:

1. Install and prepare (5 minutes). Install Plauti from AppExchange, assign permission sets to admins and power users, and open the app.
2. Find and merge existing duplicates (5-10 minutes). Run a duplicate search on your target object, review the exact and fuzzy match groups, and bulk-merge using rules you've tested first. Use the audit trail to check the results.
3. Set up ongoing prevention (5 minutes). Turn on prevention rules for user-facing warnings or blocks, schedule recurring scans to catch new duplicates automatically, and add dashboards to track duplicate rates by object or source. Set incoming records — from a form, import, or API — to dedupe automatically as they arrive; that's the step that actually keeps an org clean rather than just cleaning it once. AI-assisted suggestions can also recommend which field values to keep when you merge by hand.

Before and after

Before:

• Reps call the same lead twice.
• Pipeline and campaign attribution numbers are inflated and unreliable.
• Merging duplicates one by one takes days.

After:

• One record per customer, with its full activity history intact.
• Reports and forecasts that leadership can trust.
• Bulk deduplication in minutes, with an audit trail behind it.
• Incoming duplicates get caught automatically, without losing data.