Moving from Excel to a CRM Without Losing Your Customer History

By Recenvy Team ·

Move customer data from Excel into a CRM while preserving notes, dates and ownership. A practical migration checklist for South African operations teams.

Moving from Excel to a CRM Without Losing Your Customer History

To migrate Excel to CRM without losing history, separate customer details from past interactions, give each record a stable reference and test the import before switching systems. Keep an unchanged source copy, preserve original dates and authors where available, and check that notes, documents and follow-ups still belong to the right customer.

1. Decide what history you actually need

A spreadsheet migration is not just a contact upload. Your workbook may contain years of promises, complaints, quote references and ownership changes. Moving names and telephone numbers alone leaves staff with a directory, not a usable customer record.

Start by listing every source: shared workbooks, individual sales files, archived sheets and linked document folders. Nominate one migration owner and ask sales, service and finance to identify information they need to keep working.

Divide the information into three groups:

  • Current customer details: organisation name, contacts, billing information, account owner and customer status.
  • Historical activity: dated calls, meetings, enquiries, complaints, quote references and previous account owners.
  • Open work: promised callbacks, unresolved issues, active opportunities and outstanding approvals.

For example, a Durban supplier might have one row containing a buyer’s details, a note about a disputed delivery and a promise to call on Friday. Those are three different things: a contact, historical context and an actionable task.

Not everything should move into the live CRM. Agree what must remain accessible, what belongs in a controlled archive and what can be removed under your retention policy. Migration is not permission to retain personal information indefinitely. Use the POPIA compliance checklist to review access, purpose and retention before copying sensitive records.

2. Clean the workbook without erasing evidence

Save an unchanged, access-controlled copy of every source file before editing. Work in a separate preparation copy and keep a simple change log covering renamed columns, merged records and excluded rows.

Give each customer a stable legacy ID, such as CUST-001247. Where organisations have several contacts, give each contact a separate ID and retain the organisation reference. Do not use a person’s name or a shared email address as the only matching key.

Then apply a few basic cleaning rules:

  • Standardise formats: trim extra spaces, agree date formats and store phone numbers and identifiers as text to preserve leading zeros.
  • Separate fields: split combined cells such as “Thandi, accounts, call after 14:00” into appropriate contact and instruction fields.
  • Review duplicates: compare names, registration details, addresses and contact information before merging.
  • Resolve ownership: map former employees to a current owner while retaining their historical attribution.
  • Keep unknowns honest: leave an unavailable date blank or mark it unknown rather than inventing one.

Two branches with similar names are not necessarily duplicates. Nor are two people using accounts@ at the same organisation. Put uncertain matches in an exception list for a staff member to review.

Check hidden sheets, filtered rows, comments, hyperlinks and formulas. A CSV export generally carries cell values, not workbook comments, attachments or formatting. If colour means “do not contact”, turn that meaning into an explicit field before exporting.

Preserve dates carefully. An entry such as 04/05/2025 needs confirmation before conversion to an unambiguous format such as 2025-05-04. For timed activity, document whether timestamps represent SAST. A date without a time should not acquire a made-up call time.

3. Map relationships, not just columns

Write a field map before importing. It should show the source column, destination field, conversion rule and what happens when a value is missing. Confirm which record types and historical fields the destination CRM can actually import.

Excel informationIntended destinationHistory safeguard
Company and legacy IDCustomer recordRetain the source reference for matching
Buyer and email addressLinked contactKeep the correct organisation relationship
Dated call noteActivity or historical notePreserve event date, author and source
Promised callbackAssigned taskRetain due date and responsible person
Quote number and file linkRelated reference or documentCheck access and customer association

Import parent records before dependent ones, following the CRM’s supported process. Typically, that means customers first, then contacts and related history. Maintain a cross-reference between legacy IDs and new CRM IDs so that later imports attach to the correct records.

A historical interaction date is different from the date the CRM record was created. If the system cannot backdate activities, use a clearly labelled historical note or supported custom field. For example: “Original interaction: 2024-08-12. Recorded by: previous account manager. Source: service workbook, record HIST-0081.”

Do not present imported notes as a complete system-generated audit trail. They are migrated source records with known limitations. Preserve the underlying evidence where required.

For documents, test whether links remain available after staff folders are retired. Move necessary files to approved storage and check permissions. Avoid recreating old VAT invoices as new transactions merely to make them visible in the CRM. Keep issued documents and references intact, with finance approving any financial migration.

4. Test, reconcile and control the switch

Run a pilot using representative records, not only the cleanest ones. Include a customer with several contacts, a long note, a duplicate candidate, an inactive account and a document link. Include an open complaint and an overdue callback so operational work gets tested too.

Ask the staff who use those records to check the results. Can a service agent understand the last conversation? Can a sales manager identify the owner and next action without opening Excel?

  1. Reconcile counts: compare source rows, approved exclusions, merged duplicates, successful imports and rejected records for each record type.
  2. Inspect content: check note lengths, special characters, dates, telephone numbers and customer relationships.
  3. Test access: verify that ordinary users see what they need without exposing restricted information.
  4. Check side effects: prevent historical imports from triggering unintended welcome emails, reminders or customer notifications.
  5. Rehearse recovery: agree how the pilot batch can be identified, corrected or removed before retrying.

A successful upload message is not reconciliation. If notes were truncated or attached to the wrong branch, the migration still needs correction. Set acceptance criteria before the full import, including no unexplained missing records and correct ownership of open work.

Schedule a clear editing cutoff, stated in SAST. Either pause spreadsheet updates or keep a controlled change log for a final update import. Then make the old workbooks read-only and tell staff where new information belongs.

Keep the source archive under your retention policy, but avoid maintaining two live customer lists. For broader supplier checks, the CRM buyer’s checklist covers migration testing and implementation questions.

How Recenvy fits the migration

Recenvy’s CRM (Customers) sits alongside sales pipelines, quotes and VAT invoicing, helpdesk, projects and reporting. The practical aim is to make the customer record the starting point for ongoing work, rather than another list staff must update separately.

Before migrating, confirm the supported import fields, contact relationships, historical-note handling, document approach and permissions with the implementation team. Use your pilot workbook to agree what will transfer directly, what needs preparation and what should remain in an accessible archive.

Frequently asked questions

Can we import everything from one Excel file?

Possibly, but separate sheets or files may be needed for customers, contacts and activities. Follow the destination system’s templates and use stable IDs to preserve relationships.

Will our original notes keep their dates?

Only if the import supports those dates or you preserve them explicitly in historical fields or note text. Test this before importing the full dataset.

Should we delete duplicate customer records first?

Review and merge confirmed duplicates in the preparation copy. Retain a record of the source IDs and ensure that no related notes or open tasks disappear.

When can we stop using the spreadsheet?

After reconciliation, user checks and the final update import are complete. Make the source read-only once the migration owner approves the switch and staff know the new process.

A good migration preserves context, not just rows. Start with a controlled source copy, prove the mapping on real records and switch only when your team can trust the result. To discuss moving your customer data into Recenvy, book a demo.

More articles