Failure Patterns, Contracts and Official Sources
Moving from Excel or another CRM: a migration plan that does not lose leads
How an Indian small business moves leads and customers from spreadsheets, WhatsApp or an old CRM — clean-up, field mapping, consent data, cut-over, costs and checks.
A CRM migration succeeds when you move less data, cleaner: standardise phone numbers, merge duplicates, keep source and consent information, decide what to leave behind, map fields before importing, and switch over on a fixed date with a short freeze. For a small business moving from spreadsheets and WhatsApp, this is typically one to three weeks of part-time work; from another CRM, add time for exporting activity history and attachments. This guide gives the plan, the checks and an illustrative cost; it is part of why CRM rollouts fail.
Step 1: inventory what you have
| Source | Typical content | Problems to expect |
|---|---|---|
| Excel or Google Sheets | Leads, customers, notes in cells | Several sheets per salesperson, inconsistent columns |
| WhatsApp Business app | Chats, labels, contacts | History hard to export usefully; consent rarely recorded |
| Old CRM | Contacts, deals, activities, attachments | Export limits; custom fields; IDs linking records |
| Accounting (Tally etc.) | Customers with GSTIN and addresses | Different spellings from sales data |
| Marketplace or portal dashboards | Lead history | Export period limits |
Step 2: decide what to move and what to leave
Not everything deserves to move. A practical rule:
- Move: active customers, open deals, leads from the last 12 months, and anyone with a recorded consent or a current relationship.
- Archive (export and store securely, do not import): older closed deals needed for reference.
- Delete: duplicates, junk, contacts with no basis for further contact, and data you no longer need for any purpose.
Leaving data behind is not only tidier; the DPDP Act's principles of purpose limitation and erasure point the same way. See DPDP and consent.
Step 3: clean before importing
| Clean-up task | Rule |
|---|---|
| Phone numbers | One format, e.g. +91 followed by 10 digits; separate landline and mobile |
| Duplicates | Merge on mobile number first, then on email and company name |
| Names and companies | Consistent capitalisation; company names matched to accounting where possible |
| Cities and states | Standard list (matters for territory rules and GST place of supply on quotes) |
| Source | Fill where known (marketplace, website, referral) |
| Consent and opt-out | Record what you know; mark unknown as unknown rather than guessing |
| Owner | Every record assigned to a current user |
Step 4: map fields
Create a mapping sheet: old column → new field, field type, allowed values. Create custom fields in the new CRM before importing. Decide stage mapping for open deals (old "Hot" might become "Quotation sent"). If moving from another CRM, keep the old record ID in a field so you can trace issues later.
Step 5: test import, then full import
- Import 50 records; check duplicates, owners, stages and special characters (names in Indian scripts should display correctly).
- Fix the mapping; delete the test records.
- Import in order: companies → contacts → deals → activities and notes → attachments.
- Reconcile counts against the source: number of contacts, open deals and total open pipeline value.
Step 6: cut-over
- Freeze date: from this date, new leads go only into the new CRM.
- Delta import: move records created or changed between the full import and the freeze.
- Switch sources: point forms, marketplace connectors and telephony to the new CRM (integrations).
- Retire the old route: make old sheets read-only; stop the old CRM's automations so customers do not get double messages.
- Keep the old CRM readable until the export is verified and the contract ends; check the notice period first (contract terms).
Moving from another CRM: extra points
- Request a full export early; some vendors limit exports by tier or need a support ticket.
- Activity history and emails are often the hardest to move; decide whether you need them in the new system or as an archive.
- Call recordings and attachments may need a separate download.
- Re-register nothing on DLT: your entity, headers and templates stay with your business, but update the new CRM or SMS provider with your template IDs, and check that the telemarketer chain is updated.
- Your WhatsApp Business Platform number may need to be moved between providers; plan the timing.
Illustrative migration cost
| Task | Hours | Cost at ₹400/hour (illustrative) |
|---|---|---|
| Inventory and decisions | 6 | 2,400 |
| Clean-up of 8,000 records | 30 | 12,000 |
| Field mapping and setup | 8 | 3,200 |
| Test and full import, reconciliation | 10 | 4,000 |
| Cut-over and source switching | 6 | 2,400 |
| Total internal | 60 | 24,000 |
| Optional partner help (fixed scope) | — | 20,000–60,000 |
Include migration in the one-time costs of your 3-year TCO and the TCO estimator.
Checks after migration
| Check | Pass condition |
|---|---|
| Record counts | Match source after excluding deliberate deletions |
| Open pipeline value | Matches within rounding |
| Random sample of 30 records | Fields, owner, stage and notes correct |
| Duplicates | Below 2% on mobile number |
| Opt-outs | All known opt-outs suppressed |
| Integrations | New leads arriving; calls logging |
FAQ
Should we migrate all historical WhatsApp chats?
Usually not. Keep an export as an archive if you need it, and start fresh conversations in the new system. Importing old chats rarely helps sales and adds personal data you must protect.
How long should we keep the old system?
Until the export is verified and the contract notice period ends. Then delete or securely archive, in line with your retention rules.
Can we do the migration over a weekend?
For small, clean datasets, yes. For messy data, the clean-up takes longer than the import; schedule that first.
Sources
- Ministry of Electronics and Information Technology — DPDP Act and Rules on purpose limitation and erasure
- Telecom Regulatory Authority of India — DLT registrations that stay with the sender
- Tally Solutions — customer master data in TallyPrime
- CERT-In — secure handling of exported data