A CRM migration checklist for Indian sales teams

Most sales teams don't have a CRM problem. They have eleven versions of one spreadsheet, three WhatsApp chats holding the actual commitments, and one person who knows which row is current. If you run a 5-50 person B2B company and your pipeline review is really a spreadsheet review, the migration is mostly logistics — and it fails on process, not on which vendor you pick.

Audit the spreadsheet before you compare tools

Spend the first week auditing what you already have instead of shopping for software. A sheet that has survived three years contains the answers to most of your scoping questions: how many real accounts you have, how many fields anyone actually fills in, and which stages nobody touches.

Export to CSV and count. In our experience a 4,000-row sheet from an Indian mid-market distributor collapses to 1,400-1,800 unique accounts once you strip duplicates, test rows, and buyers entered under three spellings of the same firm. That number — not your headcount — decides your seat count, your import plan, and whether you need custom fields at all. A few of these cleanups are written up in the case studies.

Define stages by exit criteria, not instinct

Every stage in the CRM needs an exit criterion a rep can point to, or your pipeline report will be fiction. 'Proposal sent' is a feeling. 'Proposal sent and acknowledged in writing by the buyer' is a stage. Write the criteria on one page before you configure anything, and cap yourself at five or six stages — most SMB sheets carry nine to eleven, and at least three are never used.

Two fields deserve to be mandatory: deal value and a dated next step. If a rep can save a deal without either, you have built a diary, not a pipeline. Everything else — lead source, industry, GSTIN for B2B invoicing — can stay optional for the first quarter and be tightened later.

Clean the data in a single sitting

Clean the sheet in one pass with the person who built it, or you will spend a month reconciling two half-truths. The work is unglamorous and mostly mechanical:

• Split it into three: companies, contacts, and open deals. Spreadsheets flatten all three, which is why reporting breaks the moment a second contact joins an account.
• Deduplicate on email and phone, not company name — Indian firms often run several entities under one brand.
• Normalise phone numbers to +91 and strip spaces, so WhatsApp capture later doesn't create a second record for the same buyer.
• Fill every owner column. An unowned row becomes nobody's job by Friday.

Budget a full day for a few thousand rows. It is the cheapest day of the entire project.

Import in two passes, then freeze the old file

Import companies and contacts first, verify them, then import open deals — and make the old spreadsheet read-only on the day you go live. Two passes keep the failure small: a bad contact import is easy to spot and reverse, unlike a thousand deals landing with the wrong stage and owner.

Rename the file something unambiguous like legacy_sheet_2025, keep it for one quarter for reconciliation, then delete it. Under the DPDP Act, 2023 you are accountable for the personal data you hold, so export only contacts you can justify — a live deal, an enquiry from the last 18 months, or recorded consent. Dormant leads from 2019 are a liability, not an archive. If you would rather not run the cutover yourself, implementation support is a reasonable thing to buy.

Train on live deals, then track two numbers

Train reps on their own live deals, not a demo pipeline, and judge adoption by two numbers rather than a dashboard. Book 45 minutes with each rep in week one and have them enter the five deals already on their desk while you watch. You will learn more about your field design in that session than from any amount of feedback.

After that, check two things every Monday: the share of open deals with a dated next step, and logged activities per rep. Six reps doing five calls or emails a day should show roughly 150 activities a week; if the CRM shows 40, the work is happening in WhatsApp and going unrecorded. Below 80% next-step coverage, treat it as a process problem before you treat it as a discipline problem.

Where to start

Pick one team, one region, or one product line and run a 30-day pilot before migrating anyone else. Freeze that team's sheet, import only its open deals, and review the two Monday numbers for four weeks. If the pilot holds, the rest of the rollout is repetition. Maysan Labs does this work for Indian SMBs — CRM migrations, sales dashboards, internal tools — but the checklist above is genuinely runnable on your own with a free-tier CRM and one focused weekend.