CRM Data Mapping: Field-by-Field Migration Strategy
Someone told sales ops to export the old CRM and import the new one. Source and target fields never match perfectly. Ad-hoc CSV mapping truncates notes, orphans picklist values, and turns currency into text, then the damage only shows up when forecasts break.
We build the field-mapping workbook and transformation rules that keep a clean cutover.
Sound Familiar?
These are the CRM migration mapping pains sales ops and CRM admins bring us after a "simple" CSV weekend:
- Leadership asked sales ops to "just export and import" into the new CRM over a weekend
- Source and target fields never match: text vs picklist, multi-select vs single, dates in three formats
- CSV mapping drops truncated notes, orphaned picklist values, and currency that lands as plain text
- Type mismatches fail silently; reps discover blank stages and wrong deal values weeks after go-live
- Post-cutover cleanup burns 60 to 120 hours of ops time while forecasts stay untrustworthy
Without proper validation, 5 to 15% of records are lost or corrupted, and 67% of organisations discover major data quality issues mid-migration. Teams also underestimate mapping effort by about 40%. Treating schema mapping as a spreadsheet exercise is how silent data loss becomes months of cleanup.
What Field-by-Field CRM Migration Mapping Actually Does
Audit schemas → sign the mapping workbook → trial load → cut over with transformation rules intact.
Inventory Both Schemas
Source and target field lists with types, picklists, custom fields, and unused columns to drop
Write Transformation Rules
Picklist remaps, type casts, truncation guards, and defaults for every mismatch
Trial Load in Sandbox
Sample batches prove counts, associations, and picklist coverage before production
Signed Cutover
Production load against the workbook; cleanup measured in hours, not months
Everything You Need for Reliable Schema Mapping
Field-by-Field Mapping Workbooks
Every source field mapped to a target property with data type, required status, and owner sign-off before a single record moves.
Transformation Rules
Picklist value remaps, date normalisation, currency casting, multi-line text length rules, and default handling for empty required fields.
Schema Mismatch Detection
We surface Salesforce-to-HubSpot, Dynamics, and Pipedrive incompatibilities before load: formula fields, lookup IDs, and custom objects.
Trial Loads and Spot Checks
Sample batches into a sandbox first. Record counts, picklist coverage, and association integrity get signed off before production cutover.
Orphan and Truncation Guards
Values that do not exist in the target picklist, notes that exceed field limits, and flattened relationships get flagged instead of silently dropped.
Cutover Runbook
Freeze window, rollback path, and parallel-run checklist so sales keeps selling while mapping quality is proven.
Platforms We Map Between
From 90 Hours of Cleanup to Under 8
How a 14-person B2B sales ops team stopped a Salesforce-to-HubSpot weekend CSV import from becoming months of silent data loss.
The Ad-Hoc CSV Plan
- Leadership asked ops to export Salesforce and import HubSpot over a long weekend
- Column matching by eye: StageName to dealstage, Amount to amount, notes into a short text field
- Picklist values that did not exist in HubSpot imported as blanks; deal values showed as text
- About 4% of sample records failed spot checks after go-live, in line with manual ETL error bands
- Ops spent roughly 90 hours over six weeks fixing stages, amounts, and truncated notes
Signed Field Mapping
- Full schema audit of Salesforce and HubSpot properties before any production load
- Field-mapping workbook with picklist remaps, type casts, and truncation guards signed by sales ops
- Sandbox trial load reconciled record counts, associations, and picklist coverage
- Production cutover against the workbook; source left read-only until spot checks passed
- Post-cutover validation finished in under 8 hours with record errors below 0.5%
Before vs After Field Mapping
How It Works
From first conversation to signed cutover in 3 to 6 weeks for most mid-market schemas.
Schema Audit
Export field inventories from source and target: types, picklists, custom fields, and which ones sales actually uses.
Mapping Sign-Off
30-minute scoping call, then a field-mapping workbook with transformation rules for every mismatch your ops lead signs.
Trial Load & Validate
Sandbox import with real samples. We reconcile counts, picklists, associations, and truncated text before production.
Cut Over Clean
Production load with the signed workbook. Source stays read-only until spot checks pass, then cleanup is measured in hours not months.
Frequently Asked Questions
Why can we not just export CSV and import into the new CRM?
Source and target schemas never match field-for-field. Picklist labels differ, multi-select becomes single-select, notes truncate, and lookups flatten into orphaned IDs. Industry benchmarks put manual migration error rates at 1 to 5% of records, and rushed imports without validation lose or corrupt 5 to 15% of data. The CSV path looks free until sales spends weeks fixing blank stages and wrong amounts.
What is CRM data mapping and schema mapping in practice?
Field mapping is a signed workbook that lists every source field, its target property, data type, and any transformation rule (for example "Closed Won" becomes "Closed won", or Salesforce StageName maps to HubSpot dealstage). Schema mapping covers objects and relationships: contacts to companies to deals. Without both, associations break and reports lie.
How long does a field-by-field migration mapping take?
Most mid-market mapping engagements take 3 to 6 weeks from schema audit to signed workbook and trial load. Simple contact-and-deal moves sit toward the shorter end. Heavy custom fields, picklist sprawl, and multi-object associations take closer to 6 to 8 weeks, which is still faster than months of post-cutover cleanup.
Which CRM field type mismatches cause the most damage?
The usual failures are picklist values that do not exist in the target (records import with blank stages), long-text fields truncated to short text, currency or number fields imported as text, date formats that shift by a day, and lookup or master-detail IDs that arrive as dead strings. Salesforce formulas, HubSpot calculated properties, and Dynamics choice sets each need explicit rules, not a best-guess CSV column match.
How much does bad CRM data and a botched mapping cost?
Gartner puts the average organisational cost of poor data quality near R213 million a year (about $12.9 million at R16.54 to the dollar). Closer to home, a DIY mid-market migration typically burns 60 to 120 hours of internal time, roughly R74,000 to R298,000 in loaded labour before remediation. The 1-10-100 rule still applies: fixing a bad record after cutover costs about ten times verifying it during mapping.
How much does WebFootprint field-mapping work cost?
Focused field-mapping workbooks with transformation rules and trial loads typically start from around R25,000. Full migration mapping with multi-object associations and parallel-run support usually lands between R40,000 and R90,000. Against 60 to 120 hours of DIY cleanup and the risk of 5 to 15% silent data loss, most clients see payback inside one to two quarters.
Stop Betting the Cutover on a CSV Weekend
If source and target fields do not match, and leadership still wants an export-import, you need a signed mapping workbook before records move.
Tell us which CRM you are leaving, which you are joining, and how many custom fields and picklists sit in the middle. We will show you exactly how field-by-field mapping and transformation rules would work for your cutover.