Cleanup your Salesforce inherited org with Dedupely

Stepping into a Salesforce org you didn’t build is never simple. What you’re really inheriting is years of habits, imports, integrations, and fixes layered on top of each other. And somewhere in all of that, you’ll find duplicates.

The hardest part is knowing what’s safe to touch: Some records may be linked to Opportunities, others may be feeding in from marketing systems, and some might look identical until you dig into the details. That uncertainty is what makes the first steps feel overwhelming.

This guide gives you a way forward. Instead of trying to solve everything at once, you’ll learn how to break the problem into clear stages: understand what you’ve inherited, surface obvious duplicates, define logic that fits your org, and build a repeatable process. Dedupely provides the tools so you can show progress early and keep Salesforce reliable over time.

Free resource: Salesforce inherited org cleanup plan using Dedupely

Download our step-by-step checklist to work alongside this guide. It outlines the exact actions to take in Dedupely, from setting up your first Search Pads to running bulk merges and scheduling ongoing automation, so you can follow a clear process as you clean up your org. Find it below this guide.

Identify the challenges inside an inherited org

Inheriting a Salesforce org often means working with duplicates that are deeply tied into the system. They can appear in custom fields, record types, and integrations, which makes the work more complex than simply merging records.

These are some common red flags to watch for include:

• Duplicate rules that are broken, inactive, or contradictory
• Hundreds of custom fields spread across multiple record types
• Inconsistent standards for how data is entered
• Ownership split between sales, marketing, and operations
• Syncs and integrations are generating duplicates in the background

Each of these issues makes deduplication more complicated, but they’re also clues: They show you where duplicates are most likely to come from, and where you need to tread carefully before making changes.

Avoid the common mistakes when addressing duplicates

The fastest way to make an inherited Salesforce org harder to manage is by rushing into merges without context. Before you take action, keep these common mistakes off your list:

• Running auto merge before you understand the data
• Deleting records you assume are "safe"
• Trusting weak match logic like First Name + Last Name alone
• Merging records that belong to teams you haven’t spoken to
• Starting with the riskiest objects, such as Accounts tied to open Opportunities

Avoiding these pitfalls protects both your data and your credibility.

Define what a duplicate is for your org

Every Salesforce org has its own history, quirks, and field usage, which means there’s no single definition of a “duplicate.” The key is to decide what it means in your system and then document that logic so it’s consistent.

Start with the fields you know you can trust:

• Email: Strong indicator for Contacts and Leads
• Phone: Useful when formats vary, especially with fuzzy matching
• Domain root: Reliable for grouping Accounts by company

Stay cautious with weaker indicators:

• Names alone often create false positives. Similar names can belong to different people or businesses.

Leverage what’s unique to your org:

• Domain root works well for consolidating Accounts that use different web addresses under the same company.
• Custom fields like Source, Owner Type, or Integration Tag often carry context about why a record exists.

Pro tip: Document everything: Use consistent naming conventions for your Search Pads in Dedupely (e.g., Lead Match – Email Exact). This makes it easier to share logic with teammates and avoids re-inventing criteria later.

The more deliberate you are here, the more predictable your results will be when you start merging.

Begin with obvious duplicates

The best way to build confidence in an inherited org is to start with obvious duplicates. Leads are usually the right place to begin; they’re lower risk than Contacts or Accounts, and you’ll see quick results without affecting critical business processes.

In Dedupely, create a Search Pad with simple, reliable logic:

• First Name: Exact match
• Last Name: Exact match
• Email: Exact match

Add filters to narrow the scope:

• Created date > 90 days ago (removes recent activity)
• No owner or assigned to a test user (avoids disrupting active work)

When you run the scan:

• Review the matches carefully
• Skip anything that looks uncertain
• Take note of patterns you notice for future logic

The goal at this stage is to learn what duplicates look like in your org and establish a process you can repeat with more complex objects later.

Create a structure so your deduplication process holds

Once you’ve started running scans, it’s essential to create a framework you can rely on over time. Deduplication works best when it’s organized and repeatable.

Here’s how to set it up:

• Create one Search Pad per object: Start with Leads, then expand into Contacts and Accounts.
• Name each Search Pad clearly: For example: Lead – Email Exact – Old Records. This keeps logic visible and prevents confusion later.
• Begin with manual review: Look at groups before you commit to merges.
• Skip risky records and tag them: You can return to them later once you’ve aligned with stakeholders.
• Share your logic before automation: Walk through your match setups and merge rules with the teams that own the records.
• Set a schedule: Weekly or bi-weekly reviews keep the process moving without overwhelming you or the users who rely on the data.

With a structure in place, you can gradually scale deduplication, involve the right people, and avoid surprises as you move toward automation.

Track progress in practical ways

In a large Salesforce org, success is measured by the clarity you bring to the data and the predictability of your process.

Look for these progress markers along the way:

• You’ve defined match logic and Search Pads that work consistently
• You’ve reviewed at least 20 matched duplicates without mistakes
• You’ve built a list of records you’re confident about merging
• You’ve shared your deduplication strategy with your team
• You’re tracking skipped records and field inconsistencies for later review

These checkpoints show that your process is working, even if the total duplicate count hasn’t dropped dramatically. They also give you tangible wins to share with stakeholders and proof that the org is moving toward cleaner, more reliable data.

Move forward one step at a time

Inherited orgs are common, what matters is taking a clear first step and building from there.

Dedupely gives you structure to work with: defined match logic, targeted filters, and visibility into match groups so decisions are consistent and repeatable.

Start small, stay consistent, and don’t skip the thinking part. The more you work this process in Dedupely, the faster deduplication moves.