TL;DR
In late 2023 we published a Pipedrive to HubSpot migration guide that walked through the basic steps. It was accurate but incomplete. Two major changes have hit this specific migration since then. Pipedrive moved to token based API rate limiting in December 2024, and deprecated its V1 API effective January 1, 2026, with full shutdown scheduled for July 31, 2026. Any migration script or iPaaS connector built before those changes needs re architecting to V2. Meanwhile, HubSpot’s data model has matured with custom objects, richer workflow associations, and Breeze AI, which means the target environment is substantially more capable than it was in 2023. As HubSpot Gold Solutions Partners and Pipedrive Partners who have run this migration for mid market B2B teams, this post is the technical playbook we actually use today.
Key Takeaways
- Pipedrive V1 API was deprecated on January 1, 2026 with full shutdown scheduled for July 31, 2026. Migration scripts built before December 2024 rate limit changes need re architecting to V2.
- CSV imports fail for anything beyond flat contact lists. Real migrations that preserve object associations (Contact to Company to Deal to Activity threads) require API based ETL.
- Pipedrive Activities are the biggest landmine. Post March 2024, a Pipedrive activity can link to only one object at a time. HubSpot Engagements can link to multiple. That single mapping decision defines the migration architecture.
- Deduplicate in Pipedrive before the migration, not after. Duplicate records in Pipedrive become duplicate records in HubSpot with broken associations, and cleaning them up after the fact is meaningfully harder than cleaning them up before.
- Property parity is table stakes. Create every HubSpot custom property before any data comes in. Migrating into a sparse portal produces sparse data, and re running the migration is not a viable fix.
- Standard scope migrations run 8 weeks end to end. Complex builds with large historical datasets, multiple pipelines, or heavy activity history run 12 to 16 weeks. Anything faster than 4 weeks is either lying or losing data.
- We are HubSpot Gold Solutions Partners and Pipedrive Partners. This post is the actual playbook we run for client engagements, not a theoretical walkthrough.
Introduction
In late 2023 we published a step by step guide to migrating from Pipedrive to HubSpot. It was accurate for the environment as it existed then. It is now incomplete. Two structural changes have hit this specific migration since our original post, and both matter for anyone planning one in 2026.
Pipedrive moved to token based API rate limiting in December 2024, with full rollout completing by May 2025. Then Pipedrive deprecated its V1 API effective January 1, 2026, with full shutdown scheduled for July 31, 2026. Any migration script or iPaaS connector built before those changes needs re architecting to V2. If you built a migration plan in 2023 and shelved it, that plan no longer runs.
On the HubSpot side, the target platform has matured significantly. Custom objects moved from an enterprise novelty to a standard tool. Workflow associations got richer. Breeze AI became a first class part of the platform. What you are migrating into is not the HubSpot of 2023.
We wrote this update from the seat of a firm that has done this migration for dozens of mid market B2B teams. Mountainise holds accreditation as a HubSpot Gold Solutions Partner and a Pipedrive Partner. That dual accreditation matters here because half of the work in a Pipedrive to HubSpot migration is understanding what your Pipedrive data actually looks like before you touch a HubSpot property. This post is the technical playbook we run internally.
FAST FACT: Pipedrive V1 API was deprecated effective January 1, 2026 with full shutdown scheduled for July 31, 2026. Any migration built on V1 endpoints will stop working entirely after that date. Source: ClonePartner Pipedrive to HubSpot Migration Guide, 2026.
Why the Migration Is Materially Different
Three shifts make the Pipedrive to HubSpot migration in 2026 a different technical exercise than it was.
Pipedrive API changes: The V1 deprecation is the most consequential. V2 uses cursor based pagination instead of offset pagination, requires different authentication headers, and returns some fields in different structures than V1. iPaaS connectors and custom scripts that worked in 2023 will silently fail in 2026 if not rebuilt. We audit for this on day one of any migration engagement.
HubSpot custom objects went mainstream: In 2023, most mid market HubSpot portals used only the standard objects (Contact, Company, Deal, Ticket). In 2026, custom objects for Subscriptions, Contracts, Projects, Assets, and industry specific entities are common. That changes the target model. Fields that used to be dumped into deal properties can now live as first class objects with their own associations and reporting. Migration architecture decisions in 2023 that flattened everything into deal fields are worth revisiting.
Breeze AI changed the property design conversation: HubSpot’s AI features rely on well populated, well described custom properties. A migration that dumps Pipedrive data into unnamed custom fields produces a HubSpot portal that Breeze cannot help with. Property naming and description matter more than they did three years ago.
For the broader HubSpot side of what teams are migrating into, see our HubSpot Revenue Engine and Pipedrive CRM services practices.
The Object Model Translation (This Is Where Migrations Break)
Every Pipedrive to HubSpot migration reduces to object model translation. Get this right and the rest is mechanical. Get it wrong and every downstream problem traces back to this decision.
Pipedrive People to HubSpot Contacts: Standard fields (name, email, phone) transfer cleanly. Custom Person fields need to be created as HubSpot Contact properties before import. Owner field maps to HubSpot owner. Marketing status has no direct equivalent and needs a rule set (usually driven by email opt in status and last activity date).
Pipedrive Organizations to HubSpot Companies: Similar clean mapping for standard fields. The Contact to Company association needs to be explicitly preserved during migration, not assumed. Pipedrive supports multiple People per Organization; HubSpot supports the same via primary and associated Contact relationships, but the primary must be identified during the mapping phase.
Pipedrive Deals to HubSpot Deals: Pipeline stage names will not match. Each Pipedrive stage needs to be manually assigned to the corresponding HubSpot stage, including any custom stages your team built. Deal amount, currency, close date, and owner map directly. Deal products (line items) require separate handling since HubSpot uses a Line Item object model.
Pipedrive Activities to HubSpot Engagements: This is the landmine. Post March 2024, a Pipedrive activity can link to only one object at a time (deal, lead, or project). HubSpot Engagements (notes, emails, calls, meetings, tasks) can link to multiple objects. The migration must decide, activity by activity, which HubSpot objects each Pipedrive activity should associate with. Most orgs skip this decision and end up with orphaned activity history in HubSpot.
Our CRM Implementation practice runs the object model translation as a formal deliverable before any data extraction begins.
The 7-Step Migration Sequence We Actually Run
Every migration we run follows the same sequence. This is what our Analysis-First methodology looks like applied to Pipedrive to HubSpot specifically.
Step 1: Data audit and snapshot: Freeze the Pipedrive data on a specific snapshot date. Quantify total records by object. Identify duplicate contacts (same email), duplicate companies (same domain), and orphaned records. Estimate how much history is worth migrating versus archiving to read only Pipedrive access.
Step 2: Field mapping workbook: Every Pipedrive field mapped to its HubSpot equivalent, with resolution logic for fields with no direct mapping. Flag fields containing multi value data, concatenated strings, or free text that will need transformation. This workbook is your source of truth for the entire migration.
Step 3: HubSpot portal configuration: Create all custom properties, pipeline stages, and lifecycle stages before any data comes in. Configure lifecycle stage automation, deal pipelines, and required properties. Set up owner accounts and team structures. Migrating into a sparse portal produces sparse data and you cannot easily fix it after.
Step 4: Pre migration deduplication: Run deduplication in Pipedrive before extracting anything. Duplicate records in Pipedrive become duplicate records in HubSpot with broken associations. Cleaning them up post migration is meaningfully harder than cleaning them up pre migration.
Step 5: Test migration: Migrate 500 to 2,000 records across all object types into a HubSpot sandbox or test portal. Validate with actual sales users, ideally the two reps and one manager whose data you migrated. Fix mapping issues and re run until the data looks clean and associations are intact.
Step 6: Production migration: Extract via V2 API, transform in the mapping layer, load into HubSpot with associations preserved. Rate limit aware, with retry logic for failed records and a full log of every record that was migrated, skipped, or errored.
Step 7: Validation and cutover: Reconciliation against source counts. User acceptance testing with named sales users. Training on HubSpot workflows. Cutover of daily operations from Pipedrive to HubSpot with a formal decommission date for Pipedrive write access. Our HubSpot Implementation Services practice runs this full sequence for client engagements.
FAST FACT: Post March 2024, a Pipedrive activity can link to only one object at a time. HubSpot Engagements can link to multiple. That single object model difference is the biggest source of migration errors we see. Source: Integrate IQ Pipedrive to HubSpot Migration Technical Guide.
Considering a Pipedrive to HubSpot migration?
Before you scope internally or engage a vendor, our two minute RevOps AI CRM Audit maps the exact object model complexity in your Pipedrive instance and estimates migration scope, timeline, and risk. No sales call required to see the output.
Start Your Free RevOps AI CRM Audit →
The Migration Landmines We See Most Often
Almost every migration we audit had the same set of avoidable problems. Here are the ones that come up most often, in rough order of damage caused.
Orphaned activity history: Activities migrated without deliberate association mapping end up in HubSpot linked to nothing useful. Sales reps open a contact record and see empty timelines. Then they stop trusting the system. This one destroys adoption faster than any technical problem.
Deal stage mismatches: Migrating deals into HubSpot with automatic stage mapping produces deals stuck in stages that no longer make sense. Every deal should be reviewed for stage validity, and stalled deals older than a threshold (usually 90 days past expected close) should either be moved to closed lost or archived.
Custom field type mismatches: Pipedrive multi value fields concatenated into single HubSpot properties break click to dial, click to email, and reporting. Dropdown fields with values that do not exist in HubSpot get dumped as free text. These issues surface weeks after go live when someone tries to run a report and the data does not aggregate.
Missing owner assignments: Pipedrive deals owned by users who never got HubSpot licenses lose their owner and appear as unassigned. This is a straightforward pre migration fix (reconcile the user list) that gets skipped when timelines slip.
File attachments left behind: Files attached to Pipedrive deals and activities do not migrate automatically. Every file needs to be either migrated separately (via API) or accepted as read only reference data in the archived Pipedrive instance. Deciding this upfront prevents scope creep during cutover.
Our Business Process Consultation practice handles the operational side of preventing these issues before they become adoption problems.
Native Connector vs Full Migration: When to Choose What
Not every team needs a full migration. The right answer depends on why you are moving off Pipedrive and what data you actually need in HubSpot.
The native HubSpot Pipedrive connector syncs contacts, companies, and basic deal data in real time between the two systems. This is the right choice when you want to run both CRMs in parallel for a defined period, when marketing is on HubSpot and sales stays on Pipedrive, or when you have a subsidiary or acquired company on Pipedrive that needs to feed into a HubSpot mothership.
Full migration is a one way move: extract from Pipedrive, transform, load into HubSpot, then shut Pipedrive down. This is the right choice when the goal is standardization on a single CRM, when Pipedrive licensing costs no longer justify the parallel operation, or when data governance requires a single source of truth.
The middle path we see most often: full migration of active records (deals in progress, contacts touched in the last 24 months) combined with read only archived Pipedrive access for historical reference. This reduces migration scope and cost while preserving auditability.
The native connector has documented limits around dropdown fields, custom objects, and activity types. If your Pipedrive instance has heavy customization, the connector will not capture all of it. Our Marketing Automation practice helps clients decide between connector and migration based on the actual data model and business drivers.
Timeline Reality: What Actually Takes 8 Weeks
Standard scope migrations run 8 weeks end to end. Complex builds run 12 to 16 weeks. Here is where the time actually goes.
Weeks 1 to 2: Discovery, audit, and field mapping: Data snapshot. Duplicate identification. Field mapping workbook. Custom property design. This phase determines everything downstream. Skipping it or rushing it is the single most reliable predictor of migration failure.
Weeks 3 to 4: HubSpot portal configuration and test migration: Create all custom properties. Configure pipelines and lifecycle stages. Migrate 500 to 2,000 test records. Validate with named users. Fix mapping issues. This is where most timeline overruns happen because test migrations surface issues nobody anticipated.
Weeks 5 to 6: Production migration and reconciliation: Extract via V2 API in batches, respecting token based rate limits. Load into HubSpot with associations preserved. Reconcile counts against source. Fix orphaned or failed records.
Weeks 7 to 8: User training, cutover, and stabilization: Train sales teams on HubSpot workflows. Cutover daily operations from Pipedrive. Monitor for issues in the first two weeks post go live. Decommission Pipedrive write access on a defined date.
Vendors promising migrations in 2 to 4 weeks are either running much smaller scopes than they market, or they are cutting corners on the audit and mapping phases that determine data quality. Neither is a good outcome.
The Work After Migration That Actually Determines Adoption
Technical migration is only half the job. The other half is what happens in the first 90 days after go live. Almost every failed migration we have audited was technically successful and behaviorally unsuccessful.
User training on the workflow, not the tool: Sales reps do not need a HubSpot walkthrough. They need to know how to do their specific daily tasks (log a call, update a deal stage, assign a follow up) in the new system. Training organized around user tasks, not HubSpot features, produces meaningfully better adoption.
Data hygiene automation: HubSpot workflows should be configured to catch missing required properties, stale deals, and unassigned records automatically. Without this, the clean data you migrated in gets dirty within a quarter.
Reporting that shows continuity with Pipedrive metrics: If your team was reporting on “deals moved this week” in Pipedrive, they need the equivalent view in HubSpot on day one. If the numbers change or the reports look different, trust in the migration collapses regardless of technical accuracy.
Executive dashboard parity: Whatever the CEO looked at in Pipedrive weekly needs to exist in HubSpot before cutover. Migrating to a system where the executive report is broken produces political problems that outlast the technical wins.
Our Insight and Analytics practice builds the reporting continuity layer that keeps executive trust intact through the transition.
Summary
The Pipedrive to HubSpot migration is materially different from what it was in 2023. Pipedrive V1 API sunsets July 31, 2026. Rate limiting became token based in late 2024. HubSpot’s data model matured with custom objects and Breeze AI. Migration scripts and playbooks built before those changes need re architecting. On the technical side, the work reduces to object model translation, field mapping, association preservation, and disciplined pre migration deduplication. On the operational side, the outcome depends more on user training and reporting continuity than on technical accuracy.
Every failed migration we have audited was technically successful and behaviorally unsuccessful. The sales team did not trust the new system, the executive reports looked different, and adoption collapsed within a quarter. Getting the migration right means getting the audit and mapping phases right in the first two weeks, then executing carefully against a plan that treats user experience as a first class deliverable. If your team is planning a Pipedrive to HubSpot migration in 2026, or you already started one and it has stalled, that is a conversation we have almost every month.
Planning a Pipedrive to HubSpot migration?
Mountainise is a HubSpot Gold Solutions Partner and Pipedrive Partner. We run Pipedrive to HubSpot migrations end to end for mid market B2B teams, from data audit through post migration adoption. Our Analysis-First methodology and Lucrative RevOps Framework are how we deliver.
Book a Strategy Session with Mountainise →
Frequently Asked Questions
Standard scope: 8 weeks end to end. Complex builds with large historical datasets, multiple pipelines, or heavy activity history: 12 to 16 weeks. Anything faster than 4 weeks is either cutting corners on the audit and mapping phases, or running a much smaller scope than they market. Ask any vendor for a specific week by week breakdown of what happens when. If they cannot produce one, that is a signal.
For simple contact lists, yes. For anything preserving deal history, activity threads, and object associations, no. CSV imports flatten the relationships that give your CRM data its meaning. HubSpot’s own imports handle Contacts and Companies well, Deals with limits, and Activities poorly. Real migrations run through API based ETL. If your Pipedrive instance is under 1,000 records and you do not need historical activity, CSV is defensible. Above that, it is not.
Best practice: maintain read only Pipedrive access for a defined archive period (usually 6 to 12 months post cutover). This gives you a rollback option if something surfaces post migration, and it preserves audit access to historical data that was not worth migrating. Formally decommission Pipedrive write access on a specific date so there is no ambiguity about which system is now the source of truth.
Depends on scope and internal expertise. Under 2,000 records with minimal customization: internal team can usually handle it with a well documented plan. Above that, or with custom objects, complex pipelines, or heavy activity history, specialist support meaningfully reduces risk. The bigger question is opportunity cost. Your engineering team can usually build a migration script, but their time is often better spent on product work. Our HubSpot RevOps practice runs these migrations so internal teams stay focused on their core responsibilities.
Skipping the audit and mapping phase. Every failed migration we have reviewed rushed the first two weeks of work. The technical migration then executed cleanly against a bad plan and produced a HubSpot portal that sales teams did not want to use. If you cut anything from a migration timeline, cut it from execution weeks, not from planning weeks. The planning phase is where the outcome is actually determined.
