CRM Migration Checklist Spreadsheet: Move Cleanly (Templates + Data Mapping)

Stop re-cleaning your CRM after every migration: a practical checklist that actually works

CRM migrations are notorious time sinks. You start with high hopes, export CSVs, run a few imports and then spend days — or weeks — fixing duplicates, broken relationships, bad dates and missing fields. If that sounds familiar, this guide is for you. Below you'll find a tested, step-by-step CRM migration checklist, ready-to-use field mapping templates, deduplication strategies, validation checks, and automation ideas (including Apps Script snippets) that reduce post-migration cleanup.

The 2026 context: why migrations still fail (and what’s changed)

Through late 2025 and into 2026, two trends made migration both easier and riskier. First, AI-assisted schema suggestions and low-code connectors speed up mapping. Second, heightened data governance and privacy rules mean a single overlooked field can create compliance headaches. Combine those with increasing CRM API complexity (GraphQL endpoints, rate limits, strong typing) and you need a rigorous checklist — not a hopeful CSV import. For cost and rate-limit planning, see our guide on cost governance and consumption discounts.

“Automate as much as possible — but validate everything.”

Quick wins: What a clean migration saves you

• Faster sales ramp: accurate records mean reps spend time selling, not fixing leads.
• Lower compliance risk: validated PII fields and consent flags reduce exposure.
• Less IT overhead: fewer rollback/import cycles, smaller incremental costs for API calls.
• Repeatable process: templates and scripts make the next migration predictable.

Top-level migration stages (inverted pyramid: do this first)

1. Audit & snapshot — inventory sources, export samples, capture counts.
2. Field mapping — create a canonical mapping sheet with rules and examples.
3. Deduplicate & normalize — finish dedupe before transforms where possible.
4. Validation & enrichment — run checks and repair rules.
5. Test migration — small batch, reconcile counts and relationships.
6. Full migration — run with monitoring, and keep a rollback plan.

Stage 1 — Audit & snapshot: how to make a migration safe

Before touching the target CRM, create a reliable baseline. Exports should include complete metadata and sample records.

• Export full CSVs and a compact JSON (if available) for each object: Contacts, Accounts, Deals, Tickets, Activities.
• Capture counts and last-modified timestamps. Create a single “snapshot” sheet listing: source_system, object_name, record_count, sample_size, last_exported.
• Run a quick pivot to show distribution (leads by owner, accounts by region). That helps spot anomalies early. For binary artifacts and reproducible snapshots, follow binary-release best practices in binary release pipelines.

Stage 2 — The field mapping template (use this structure)

Mapping is where migrations live or die. Below is a compact, practical template you can copy into Google Sheets or Excel. Each column is a column in your mapping sheet.

• source_system — e.g., HubSpot, Pipedrive
• source_object — Contacts, Companies
• source_field — e.g., email, phone, city
• source_type — string, date, boolean, enum
• sample_values — 3–5 typical values
• target_field — target CRM field name
• transform_rule — formula, script, normalization step
• required — Y/N
• validation — regex or lookup table e.g. email regex
• lookup_table — reference for relationship mapping
• notes — consent/legal flags, default values

Example row:

• source_system: HubSpot
• source_object: Contacts
• source_field: phone_number
• source_type: string
• sample_values: "+1 (415) 555-1234"
• target_field: Phone
• transform_rule: =REGEXREPLACE(A2, "[^0-9+]", "")
• required: N
• validation: ^\+?[0-9]{7,15}$
• lookup_table: —
• notes: international normalization

Field mapping best practices (2026)

• Map types, not just names. Date strings need a parse rule; enums require canonicalization.
• Keep sample values. Sample-driven mapping surfaces hidden formats (e.g., dates like 01-02-03).
• Define ownership. Who approves mapping changes? Add an approver column; see guidance on choosing the right CRM and governance for publishers.
• Prepare lookup tables. For account IDs, company matches, territory codes — create a crosswalk sheet.

Stage 3 — Deduplication: 3 practical methods

Deduplication should mostly happen in the source or in a staging spreadsheet before import. Mix exact-match, fuzzy-match and rule-based strategies.

Method A — Exact-key dedupe (fast)

1. Create a composite key column: e.g., =LOWER(TRIM(email)) & "|" & LEFT(TRIM(name),50)
2. Use UNIQUE() or remove duplicates by the key.
3. Flag records with missing keys for manual review.

Method B — Fuzzy matching (find near-duplicates)

When names/phones are noisy, use fuzzy matching. In Google Sheets, you can add a simple Levenshtein function via Apps Script. For Excel, use Fuzzy Lookup add-in or Power Query fuzzy merge.

// Apps Script: simple Levenshtein wrapper (add to Tools > Script editor)
function levenshtein(a, b) {
  if (a == b) return 0;
  if (a.length == 0) return b.length;
  if (b.length == 0) return a.length;
  var matrix = [];
  for (var i = 0; i <= b.length; i++) matrix[i] = [i];
  for (var j = 0; j <= a.length; j++) matrix[0][j] = j;
  for (i = 1; i <= b.length; i++) {
    for (j = 1; j <= a.length; j++) {
      if (b.charAt(i-1) == a.charAt(j-1)) matrix[i][j] = matrix[i-1][j-1];
      else matrix[i][j] = Math.min(matrix[i-1][j-1] + 1, matrix[i][j-1] + 1, matrix[i-1][j] + 1);
    }
  }
  return matrix[b.length][a.length];
}
  

Use normalized fields (lowercase, trimmed, stripped punctuation) and then calculate distance. Flag pairs under a threshold (e.g., <=2) for review.

Method C — Rule-based merges (business logic)

Apply business rules to decide merges: prefer latest activity record, prefer records with opt-in flag, or merge by account hierarchy. Encode these as priority columns and use SORT + UNIQUE with an index to pick the canonical record.

Stage 4 — Validation checks (pre-import QA)

Run these sanity checks in your staging spreadsheet and fail the migration until they pass. Add a validation column for each rule with Y/N and a reason for failures.

Core validation list

• Record counts: source vs staging vs target (after test import).
• Primary key uniqueness: no duplicates in target key column.
• Referential integrity: all lookup IDs exist in their reference tables.
• Email format & domain checks: regex + MX lookup where possible.
• Phone normalization: E.164 validation regex.
• Date ranges: created_date < closed_date etc.
• Consent & legal flags: GDPR/CCPA required fields populated or marked as redacted.

Sample formulas

• Unique key check: =COUNTIF(key_range, key_cell)=1
• Email regex (Google Sheets): =REGEXMATCH(A2, "^[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Za-z]{2,}$")
• Phone length: =AND(LEN(B2)>=7, LEN(B2)<=15)

Stage 5 — Test migration: small batches, big checks

Always run a pilot run: 1–5% of records that are representative. Validate these after import using pivot tables and reconciliation queries. For migration orchestration patterns and rollback planning at scale, see the Multi-Cloud Migration Playbook.

• Create pivot tables for counts by owner, stage, region — compare source vs target.
• Random sample comparison: pick 100 random IDs and diff fields with VLOOKUP/XLOOKUP.
• Verify relationships: accounts <-> contacts, opportunities <-> accounts are correctly linked.
• Log errors from the target CRM import API and fix the staging CSV; never edit target records directly during testing.

Automation & Apps Script: speed up repetitive checks

Use Apps Script (Google Sheets) to automate validations, generate sample exports, and call APIs for live checks. Below are compact, practical snippets.

1) Auto-flag invalid emails

function flagInvalidEmails() {
  var ss = SpreadsheetApp.getActive();
  var sheet = ss.getSheetByName('Staging');
  var data = sheet.getRange(2,1,sheet.getLastRow()-1,3).getValues();
  var out = [];
  var re = /^[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]{2,}$/;
  for (var i=0;i

function exportSampleJSON() {
  var ss = SpreadsheetApp.getActive();
  var sheet = ss.getSheetByName('Staging');
  var headers = sheet.getRange(1,1,1,sheet.getLastColumn()).getValues()[0];
  var rows = sheet.getRange(2,1,101,sheet.getLastColumn()).getValues();
  var out = rows.map(function(r){
    var obj = {};
    headers.forEach(function(h,i){ if (r[i] !== '') obj[h]=r[i];});
    return obj;
  });
  var file = DriveApp.createFile('sample_import.json', JSON.stringify(out));
  Logger.log('Saved sample to: ' + file.getUrl());
}