PipeDrive's forecast engine is built on a logic tree that few other platforms replicate exactly. When you migrate to Salesforce, HubSpot, or even Orin, deal amounts and close dates copy over cleanly enough. But the forecast math itself —the weighted probability, activity-based stage velocity, and pipeline health signals—doesn't port. At ₹50M ACV, getting this wrong means your forecast flatlines, inflates by 40%, or both. We watched a 120-person sales org leave PipeDrive and rediscover in month three that their new platform's forecast was 18% lower than reality because activity weighting was invisible. They'd migrated the data but not the logic. This post maps what breaks, what survives, and the 90-day validation checklist you need. What migrates cleanly: the core three fields Deal amount, close date, and stage name copy without loss. That's the good news. Deal amount: Numeric field, no transformation needed. Direct 1:1 mapping across any platform. Close date: Date field. Survives the export–import cycle unchanged. Ensure date format is ISO 8601 before export to avoid timezone or locale ambiguity. Stage name: Text field. Transfers, but map it manually. PipeDrive's stages ('Contacted', 'Proposal Sent') may not match your new platform's taxonomy (Salesforce's 'Needs Analysis', 'Negotiation'). Create a crosswalk before import or lose stage history. If you move only these three, you retain deal tracking and basic pipeline visibility. You lose forecast confidence . What breaks: the forecast math trio PipeDrive calculates forecast in three ways. Two of them don't migrate. 1. Weighted probability by stage PipeDrive lets you set a confidence % per stage ('Contacted' = 20%, 'Proposal' = 60%, 'Negotiation' = 85%). The forecast sums deal_amount × stage_probability for each open deal. This is not a field you export. It's a configuration rule stored in PipeDrive's database, invisible in the data dump. When you migrate, every deal defaults to 0% or 100% in the new platform (depending on whether it's open or lost). Your forecast either vanishes or becomes unrealistically optimistic. What you lose: Stage-based confidence weighting. Your new platform may offer it, but you'll rebuild it manually. 2. Activity frequency signals PipeDrive's Activity Forecast feature adjusts stage probability based on recent contact. If a rep hasn't updated a ₹5L deal in 60+ days, PipeDrive flags it as at-risk and lowers the forecast contribution. This logic is computed in real time and doesn't live in any field you can export. After migration, your new platform has no signal that a deal went dark three months ago. It sits at whatever stage weight you assigned, inflating forecast. What you lose: Automatic forecast decay. You'll need to rebuild this as a rule-based workflow or accept that old, silent deals tank your accuracy. 3. Custom forecast formulas and deal splits Advanced PipeDrive users create custom fields that feed into forecast: 'Deal Confidence Score' (1–10), 'Shared Deal %' (if one opportunity touches multiple reps), 'Probability Adjustment' (for seasonality or rep experience). These export as flat values, not formulas. Your new platform won't know how to recalculate them if a source field changes. What you lose: The living logic. A custom formula in PipeDrive stays dynamic; after migration, it's a static snapshot. The ₹50M forecast reality: where the 15–40% miss happens At ₹50M ACV with 80+ sales reps and hundreds of open deals, these three breakages compound: Shared deals inflate by 18–22%: A rep owns 60% of a ₹2L deal, another owns 40%. PipeDrive's weighting gives each rep credit for their share in forecast. If you migrate without split tracking, both get 100%, and that one deal counts twice. Stale deals stay 'active': Without activity decay, 90+ day-old deals (which close only 8–12% of the time) sit in forecast at full value. A typical org sees 12–18% of forecast bloat here. Stage weight gets reset: Your new platform's default 'Proposal' stage might be 50% probability; PipeDrive used 60%. Multiply that 10-point shift across 200+ deals, and forecast swings 4–7%. Together: 18% + 12% + 5% = forecast overstates by 35%. At ₹50M, that's ₹17.5L of fictional upside. Field-by-field migration checklist: what to export and how to validate Before you leave PipeDrive, extract these fields: deal_id, deal_name, deal_amount, close_date, stage_name: Core. Use PipeDrive's export; map stage names to your new platform's taxonomy. created_at, updated_at: Critical. Tells you deal age and whether reps are engaged. Use `updated_at > TODAY() - 60 days` to flag active vs. stale deals. user_id (deal owner), custom_field:forecast_confidence (if you have it): Ownership and any custom weighting. Export both. Activity log (call, email, meeting count in last 60 days): PipeDrive's API includes this; use it to flag stale deals post-migration. Lost deals with loss_reason: Export and archive separately. Do not import into your new platform's active pipeline; they'll