Database ReactivationResource guide

How to Prepare a CRM Database Before a Reactivation Campaign

Reactivation should not begin with "upload list, send campaign." It should begin with understanding, cleaning, verifying and segmenting the database first. This guide walks through the practical steps — duplicates, consent, opt-outs, contact sources, CSV imports and workflow testing — before any message is sent.

A business opens an old spreadsheet and finds 2,800 contacts. That number sounds impressive. But the list may contain duplicates, old email addresses, disconnected phone numbers, previous opt-outs, incomplete records, customers mixed with leads, contacts with no known source and records collected for completely different purposes.

The question is not "How many contacts do we have?" The better question is: "How many contacts do we actually understand well enough to use appropriately?"

Database size is not the same as database quality.

Why Clean the Database Before Reactivation?

Potential benefits of cleaning first include:

  • fewer duplicate messages
  • clearer segmentation
  • better understanding of contact history
  • easier opt-out management
  • better CRM organisation
  • fewer irrelevant messages
  • more reliable reporting
  • easier human follow-up

Cleaning does not automatically increase revenue or conversion rates — it improves the reliability and appropriateness of what you send.

Step 1: Back Up the Original Database

Before editing or importing anything, create a secure backup of the original data — for example a CRM export, CSV or a secure spreadsheet copy. Keep the original unchanged where practical.

If something is accidentally deleted, merged incorrectly, remapped or overwritten, the original source still exists. Do not store customer databases in insecure public folders.

Step 2: Identify Where Every Group of Contacts Came From

Possible sources include:

  • website enquiries
  • previous customers
  • quote requests
  • bookings
  • imports
  • event registrations
  • referrals
  • marketing forms
  • older CRM systems
  • manually entered records

Do not simply use "Database" as the source. Where possible preserve useful provenance such as original source, date added, original enquiry type and previous customer status.

Source helps determine context.

Step 3: Separate Customers, Leads and Other Contacts

Do not treat every record identically. Useful broad categories might include:

  • Previous Customer
  • Previous Enquiry
  • Previous Quote
  • Active Lead
  • Dormant Lead
  • Marketing Subscriber
  • Supplier / Vendor
  • Staff / Internal
  • Unknown

Do not include employees, suppliers or unrelated contacts in a marketing campaign merely because they appear in CRM.

Step 4: Remove or Merge Duplicates

Common duplicates include:

  • same person under two emails
  • same phone number entered twice
  • old and new CRM imports
  • form submitted multiple times
  • spelling variations

Duplicate problems can cause repeated messages, conflicting stages, confusing reporting and separate conversation histories.

HighLevel currently provides contact deduplication controls. Its current documentation states matching during CSV updates or imports can use identifiers such as Contact ID, Email or Phone depending on configuration.

Do not merge two records unless you are confident they represent the same person.

How HighLevel Handles Duplicate Contacts

Current HighLevel functionality includes Contact Deduplication Preferences. Businesses can configure whether duplicate contacts are allowed and which field is primarily used to identify existing contacts.

Depending on configuration this may include:

  • Email
  • Phone
  • with an optional secondary lookup

HighLevel can update existing contacts during imports rather than creating unnecessary duplicates when matches are identified. Review your deduplication settings before a major import.

Step 5: Standardise Contact Information

Review fields such as:

  • First Name
  • Last Name
  • Phone
  • Email
  • Company
  • Suburb / Location
  • Lead Source
  • Service Interest

Examples of poor data: JOHN SMITH, john smith, John, 0412 345-678, +61 412 345 678.

The objective is not cosmetic perfection — it is consistency that helps matching, segmentation, workflow logic and staff understanding. Do not overwrite useful original data without reason.

Step 6: Check Phone Numbers

Potential issues:

  • invalid formatting
  • disconnected numbers
  • missing country code
  • duplicate number assigned to several contacts
  • landline/mobile confusion
  • very old numbers
  • recycled numbers

An old phone number may eventually belong to someone else. Do not assume a phone record that was correct five years ago is still correct today. If data quality is uncertain, be conservative.

Step 7: Check Email Addresses

Potential problems:

  • obvious typos
  • duplicate addresses
  • old work addresses
  • role addresses
  • bounced addresses
  • unsubscribed addresses

Do not repeatedly send to addresses that are known to be invalid. Do not interpret a deliverable address as evidence of marketing consent.

Step 8: Protect Opt-Out Information

Existing opt-out information should not be deleted simply to make the list larger. Preserve SMS opt-outs, email unsubscribes, Do Not Disturb status where applicable and relevant channel preferences.

An unsubscribe record is useful data — not clutter to remove.

Do not re-import opted-out contacts in a way designed to bypass suppression.

Before marketing communication, understand:

  • how the information was collected
  • what the person was told
  • whether express consent exists
  • whether inferred consent may genuinely apply
  • whether the communication is relevant
  • whether an opt-out already exists

For Australian commercial SMS or email, current ACMA guidance requires appropriate consent, and the sender should be able to substantiate consent. Do not attempt to manufacture consent retrospectively. This is general information, not legal advice.

Useful CRM information may include, where genuinely available:

  • Consent Status
  • Consent Source
  • Consent Date
  • Original Form / Process
  • Channel Preference
  • Opt-Out Status

Do not fabricate consent dates or statuses because the old system did not record them. If consent is genuinely unknown, mark it as unknown.

Unknown is better than invented.

Step 10: Remove Obviously Irrelevant Records

Examples:

  • suppliers
  • staff
  • test records
  • spam submissions
  • obvious fake records
  • completely unrelated contacts

Do not simply delete potentially important historical customer information without considering retention requirements and business needs. This article is about campaign preparation, not legal data-retention advice.

Step 11: Deal With Contacts With Unknown Sources

If a business cannot determine where the contact came from, why it was collected, the previous relationship or communication expectations, do not automatically place the record into the marketing segment.

A safer operational category might be: Source Unknown — Review Required. This keeps uncertain contacts separate.

Step 12: Segment the Database

Potential useful segments:

  • Previous Customers
  • Previous Enquiries
  • Previous Quotes
  • Specific Service Interest
  • Recent Dormant Leads
  • Older Contacts Requiring Review
  • Known Marketing Subscribers
  • No Marketing / Opted Out
  • Source Unknown

Do not create dozens of segments unless the business can actually use them.

Tags vs Custom Fields

Tags

Useful for labels or categories such as:

  • previous-customer
  • plumbing-enquiry
  • quote-request
  • 2026-reactivation-review

Custom Fields

Useful for structured information such as:

  • service type
  • suburb
  • lead source
  • original enquiry date
  • relevant business attribute

HighLevel allows tags to be added during CSV imports and tags can also trigger workflows. But do not let adding an import tag accidentally launch a live marketing workflow before the database has been reviewed.

Be Careful With Workflow-Triggering Tags

Suppose reactivation-ready starts an SMS workflow. If that tag is applied to 2,000 contacts during import, the workflow could potentially start for all qualifying contacts.

Therefore:

  • build and review the workflow first
  • keep it unpublished or off where appropriate
  • import the data
  • verify the contacts
  • apply the sending segment only after review

Do not accidentally combine database cleaning and live campaign activation.

Step 13: Decide What Data You Actually Need

Do not copy every historical column into the new CRM. Potential useful fields:

  • name
  • phone
  • email
  • source
  • enquiry type
  • customer status
  • last meaningful interaction
  • relevant tags
  • opt-out status

Potentially unnecessary:

  • obsolete internal codes
  • outdated system IDs
  • unused columns
  • duplicate notes
Import useful context, not historical clutter.

Step 14: Map CSV Fields Carefully

When importing into HighLevel, review each CSV column. Example:

CSV ColumnCRM Field
First NameFirst Name
Last NameLast Name
MobilePhone
Email AddressEmail
ServiceService Interest custom field
SourceLead Source
SegmentTag

HighLevel's current documentation advises checking mappings carefully before finalising the import. Unneeded columns can remain unmapped.

What File Format Does HighLevel Use for Contact Imports?

Current HighLevel documentation states that contact imports use CSV. Do not upload .xlsx directly as the standard contact-import format. If the business has an Excel or Google Sheets file, convert or export the relevant sheet to CSV first.

Current HighLevel guidance also says to:

  • use a header row
  • ensure required identifiers are available
  • check mappings
  • review data before final import

Step 15: Do Not Overwrite Good CRM Data With Blank Values

An old import might contain empty fields where the live CRM contains current information. Careless imports could potentially replace or interfere with useful information depending on settings and process. Review update behaviour carefully.

HighLevel currently offers import controls related to handling empty values. The principle:

New import does not automatically mean newer information.

Step 16: Preserve Useful Historical Context

Where appropriate retain:

  • last enquiry type
  • previous customer status
  • meaningful notes
  • previous quote
  • known source

This may help staff understand a reply. But do not import unlimited notes or sensitive history just because it exists — only retain information legitimately useful to the business.

Step 17: Identify the Reactivation-Eligible Segment

After cleanup, create a deliberate group for review. For example:

Database → Cleaned → Source Known → No Existing Opt-Out → Relevant Segment → Consent / Communication Basis Reviewed → Reactivation Candidate

This should not automatically mean "send immediately." It means "ready for final review."

Step 18: Define the Campaign Before Building Automation

Answer:

  • Who is this message for?
  • Why are we contacting them?
  • Why is it relevant now?
  • Which channel will be used?
  • What should the recipient do?
  • What happens if they reply?
  • What happens if they opt out?
  • When does the sequence stop?

If those questions cannot be answered clearly, the workflow is not ready.

Step 19: Keep the Workflow Simple

Example:

Approved Segment → Message → Wait → Reply? YES → Stop workflow → Notify human. NO → Optional limited follow-up → End

Do not begin with a complicated 20-step automation. See our guide on practical business automation for the broader trigger-and-action model.

Step 20: Build Stop Conditions

Stop when:

  • the recipient replies
  • the recipient opts out
  • a booking occurs
  • the opportunity is closed
  • the contact becomes inappropriate for the campaign
  • staff manually remove the contact where needed
Every automated campaign needs an exit.

Step 21: Check Sender Identity and Unsubscribe

For Australian commercial SMS or email, ensure applicable requirements are addressed. Current ACMA guidance requires commercial messages to generally:

  • have consent
  • identify the sender accurately
  • provide correct contact information
  • provide a clear unsubscribe method

Current ACMA guidance says unsubscribe requests generally need to be actioned within 5 working days, and the unsubscribe method should remain functional for at least 30 days after the message is sent.

Step 22: Test With Internal Contacts First

Before contacting actual customers, test the entire flow. Check:

  • Is the right contact entering?
  • Does the message display correctly?
  • Does personalisation work?
  • Is sender identification correct?
  • Does the reply arrive in CRM?
  • Does a reply stop the workflow?
  • Does STOP or unsubscribe work as intended?
  • Is the correct staff member notified?
  • Does the pipeline update properly?
  • Are duplicate messages prevented?

Never assume a workflow is correct simply because it saved successfully.

Step 23: Start With a Controlled Segment

Do not necessarily launch to the full eligible database immediately. A controlled initial segment can help expose:

  • bad numbers
  • unexpected reply types
  • formatting issues
  • workflow mistakes
  • incorrect CRM tags
  • staff-capacity problems

Do not prescribe one universal batch size. The correct size depends on the business and its ability to respond.

Step 24: Make Sure Humans Can Handle the Replies

This is critical. If a campaign produces conversations, someone needs to:

  • read them
  • respond
  • answer questions
  • book appointments
  • prepare quotes
  • update CRM
  • handle unusual cases

Do not launch outreach at a volume the business cannot reasonably service.

The campaign is not finished when someone replies. That's when the real conversation begins.

Step 25: Monitor the Database During the Campaign

Potential operational signals:

  • failed deliveries
  • opt-outs
  • replies
  • wrong numbers
  • outdated records
  • duplicate conversations
  • campaign outcomes
  • contacts requiring manual correction

Use what the campaign reveals to improve the database.

Create a Data-Cleanup Loop

Campaign → New Information → Update CRM → Better Data → Better Future Segmentation

Examples:

  • contact says number is wrong
  • customer has moved
  • customer is no longer interested
  • email no longer works
  • enquiry becomes active

CRM should become more accurate over time.

Don't Keep Bad Data Forever Just Because It Exists

Old records may eventually stop being useful. Businesses should have sensible processes for reviewing, correcting, archiving and deleting where appropriate and lawful.

Do not prescribe universal retention periods. Different legal and business requirements may apply.

What About Sensitive Information?

Be particularly cautious. Do not create marketing segments based unnecessarily on:

  • health conditions
  • financial hardship
  • other sensitive personal information

If a CRM contains sensitive information, businesses need to consider applicable privacy requirements carefully. For dental and health businesses especially, do not import clinical history into a generic marketing CRM merely to run a campaign. Use OAIC guidance where applicable.

Common Database Preparation Mistakes

  • importing before backing up
  • no source field
  • leaving obvious duplicates
  • deleting unsubscribe information
  • treating unknown consent as YES
  • mixing customers and suppliers
  • using one campaign for every contact
  • importing unnecessary fields
  • applying workflow-triggering tags accidentally
  • sending immediately after import
  • no test contacts
  • no reply stop condition
  • no staff ready to respond
  • assuming a huge list is a good list

A Practical CRM Database Preparation Checklist

Before Import

  • Original database backed up
  • Contact sources understood
  • Customers, leads and other records separated
  • Obvious bad records removed
  • Duplicates reviewed
  • Phone and email formatting checked
  • Opt-outs preserved
  • Consent status reviewed
  • Unknown records separated

During Import

  • CSV correctly formatted
  • Columns mapped properly
  • Deduplication settings reviewed
  • Existing good values protected
  • Tags reviewed
  • No live workflow accidentally triggered

Before Campaign

  • Segment relevant
  • Sending basis reviewed
  • Message relevant
  • Sender identified
  • Unsubscribe works
  • Reply stop condition works
  • Internal testing completed
  • Staff ready for replies

Where JMG Nitro's Reactivation & Follow-Up System Fits

A practical setup may connect:

Clean CRM Database → Approved Segment → SMS / Email Workflow → Reply → Human Follow-Up → CRM Pipeline

See our Reactivation & Follow-Up System for how this is built. No sales results are promised.

Database Health Can Be Part of a Wider Growth Audit

When assessing a business process, database health may reveal:

  • duplicates
  • missing lead sources
  • enquiries with no next action
  • unclear pipeline stages
  • opt-out issues
  • dormant contacts requiring review
  • disconnected enquiry channels

These are process and data-quality opportunities — not guaranteed revenue leaks. This may later support JMG Nitro's planned Free Local Growth Audit.

For the foundational concept, see our guide on what database reactivation is, our comparison of database reactivation vs cold outreach, our guide on what a CRM is and whether a small business needs one, and our guide on which small business tasks to automate. To explore a managed reactivation campaign, see our database reactivation service.

This article provides general information only. Businesses should review current ACMA and OAIC guidance and obtain professional advice where appropriate.

Explore Database Reactivation

A reactivation campaign is only as reliable as the database behind it. See how JMG Nitro reviews, cleans and segments existing contacts before any message is sent — and how replies flow into a controlled follow-up process.

Frequently Asked Questions

Keep reading

Related Resources

Other practical guides connected to this topic.

For Australian businesses

See the System in Action

Book a free demo to walk through how JMG Nitro's systems could fit around your existing business process.

0485 099 000

Back to Resources