---
title: "Twenty CRM to Salesforce Migration: Field Mapping & Rollback"
description: "Migrate Twenty CRM to Salesforce with FlitStack AI: field-level mapping, a sample run you approve, full history and one-click rollback."
url: "https://flitstack.ai/migrations/crm/twenty-crm/to/salesforce-sales-cloud"
---

# Twenty CRM to Salesforce migration

We migrate Companies, People, Opportunities, Tasks, and Notes from Twenty CRM into Salesforce Sales Cloud with the required import sequence enforced, relationship lookups preserved through foreign key resolution, and custom objects mapped to Salesforce custom objects before any records are written to the destination.

Short answer

FlitStack AI migrates accounts, contacts, opportunities, tasks and notes from Twenty CRM to Salesforce, with their history and the associations between records. Every field is mapped and validated, a sample run is checked with you before the full migration, and the whole move can be rolled back in one click. Typical timeline: 3-5 weeks.

![](https://flitstack.ai/images/authors/aditya-kumar.webp)By [Aditya Kumar](https://flitstack.ai/about/aditya-kumar) Updated November 5, 2026

[Get a quote ](https://flitstack.ai/contact) [See what moves →](https://flitstack.ai/migrations/crm/twenty-crm/to/salesforce-sales-cloud#what-moves)

![](https://flitstack.ai/images/illustrations/pair-stage.webp)

Source Twenty CRM ![](https://flitstack.ai/platforms-trim/twenty-crm.webp)

Destination Salesforce ![](https://flitstack.ai/platforms-trim/salesforce-sales-cloud.webp)

Account Contact Opportunity Task Note

Mapped field by field · sample run · one-click rollback

## What moves from Twenty CRM to Salesforce?

Moving from Twenty CRM to Salesforce Sales Cloud is a migration that respects a strict dependency order: Companies first, then People with companyId resolved, then Opportunities with both company and person references satisfied, then Tasks and Notes with their relational targets resolved.

Read the full summary

Twenty's PostgreSQL-backed object model enforces this sequence through foreign key constraints, and Salesforce requires it through its own lookup integrity rules. We use Salesforce REST and Bulk APIs with chunking, rate-limit handling, and parent-record lookup resolution to preserve the relationship graph across the wire. Workflows, automations, and custom view configurations do not migrate; we deliver a written inventory of every active workflow requiring rebuild in Salesforce Flow. Soft-deleted records in Twenty require explicit handling because the platform's uniqueness check runs against all records including those in the deleted-view, and importing a matching email or domain triggers a restore rather than a create.

1:1

### Company → Account

Twenty Companies map directly to Salesforce Account.

How it maps

The company domain field becomes the Account Website field and serves as the dedupe key during import. We use the domain as an external ID on Account to prevent duplicate creates when matching records exist in Salesforce. Account must be the first object imported because both People (via companyId) and Opportunities (via companyId) reference it through foreign key lookups.

1:1

### People → Contact

Twenty People map to Salesforce Contact.

How it maps

The email field serves as the unique dedupe key and external ID. We resolve the companyId foreign key to a Salesforce AccountId at migration time by querying the Account external ID mapping created during the Account phase. Contacts are inserted with AccountId populated for every record where a matching Account exists; contacts without a resolved Account are held in a reconciliation queue for the customer admin to resolve before continuing.

1:1

### Opportunity → Opportunity

Twenty Opportunities map to Salesforce Opportunity.

How it maps

We resolve both companyId (to AccountId) and the personId relations (to PrimaryContactId via OpportunityContactRole) at migration time. Stage names from Twenty map to Salesforce StageName values that we configure as a Sales Process before migration begins. Amount, closeDate, and probability percentage transfer directly to Opportunity fields.

1:1

### Task → Task

Twenty Tasks map to Salesforce Task.

How it maps

We resolve the relational target (linked to People, Companies, or Opportunities) to a Salesforce WhatId or WhoId at migration time by querying the ID mapping tables created during the Company, People, and Opportunity phases. Status, Priority, ActivityDate, and description migrate directly. Tasks without a resolvable parent record are imported with no WhatId and flagged for manual reassignment.

1:1

### Note → Note

Twenty Notes map to Salesforce Note records.

How it maps

We resolve the relational target to a Salesforce record ID (Contact, Account, or Opportunity) and link via ContentDocumentLink. Note body content migrates as rich text. Attachments within notes migrate as ContentDocument records linked to the parent via ContentDocumentLink. Notes without a resolvable parent are flagged for manual reassignment during reconciliation.

1:1

### Custom Object → Custom Object

Twenty Custom Objects (user-defined entities on Organization or Pro tier) map to Salesforce custom objects of equivalent API name.

How it maps

We pre-create the destination schema in Salesforce including all custom fields, field types, and lookup relationships before any data import. Custom object records are migrated last because they frequently reference People, Companies, or Opportunities through foreign keys that must resolve to Salesforce record IDs.

******What moves 13 objects

Twenty CRM Salesforce

- Company Account 1:1
- People Contact 1:1
- Opportunity Opportunity 1:1
- Task Task 1:1
- Note Note 1:1
- Custom Object Custom Object 1:1

\+ 7 more in the mapping below

## How Twenty CRM and Salesforce differ

The differences that decide how each record is mapped. These are what a CSV export and a native importer don't resolve for you.

| Dimension               | Twenty CRM                                                                                                                | Salesforce                                                                                                                                      |
| ----------------------- | ------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------- |
| Data model              | Five standard objects (People, Companies, Opportunities, Tasks, Notes) with unlimited Custom Objects on Organization tier | Standard objects (Contact, Account, Opportunity, Task, Note) with Custom Objects available from Professional tier ($80/user) without setup fees |
| Import dependency order | Enforced: Companies before People, People before Opportunities                                                            | Lookup integrity: Account must exist before Contact insert, Contact or Account before Opportunity                                               |
| Custom object support   | Unlimited on Organization ($19/user), capped at 10 on Pro ($9/user)                                                       | Available from Professional ($80/user) without edition-based caps                                                                               |
| API rate limits         | 200 req/min on Organization cloud tier, self-hosted has no enforced limits                                                | Varies by API type: Bulk API for large loads, REST API at 100 req/min per connected app by default                                              |
| Soft-delete uniqueness  | Uniqueness checks run against all records including soft-deleted ones visible in the deleted view                         | Recycle Bin records do not block insert of matching external IDs; dedupe uses External ID field logic                                           |
| Open-source license     | AGPL-3.0 with full source code on GitHub (40.6K stars)                                                                    | Proprietary SaaS; no source code access, no self-hosting option on standard tiers                                                               |
| Email sequencing        | No native sequencing or cadence tools; requires third-party outreach platform                                             | Sales Engagement available via Sales Cloud and third-party AppExchange apps; sequences are a licensed add-on                                    |

## What to watch for in a Twenty CRM to Salesforce migration

The issues we plan around on this pair: limits and data-model gaps that the official import tools leave to you.

high

### Import sequence is enforced and must be Companies before People before Opportunities

Twenty requires CSV imports in a specific sequence: Companies first, People second (with companyId references), Opportunities third, and Custom Objects last.

How we handle it

We follow this ordering rigorously because foreign key constraints block import of any record referencing a parent that does not yet exist. Skipping this sequence silently fails or creates orphaned relations in the destination CRM. In Salesforce, the same principle applies: Account must exist before Contact insert, and Contact or Account must exist before Opportunity insert. We enforce both constraints simultaneously during migration design.

high

### Export limited to 20,000 records and visible columns only

Twenty's built-in export function caps output at 20,000 records per operation and includes only columns currently visible in the active view.

How we handle it

We pre-configure the export view to expose all fields before extracting, and chunk migrations exceeding 20K records into multiple exports by date range or record ID window to avoid silent truncation. For self-hosted Twenty instances, we extract directly from the PostgreSQL database to bypass the 20K UI export limit entirely and ensure all fields are captured.

medium

### Soft-deleted records trigger restore instead of create on unique match

Twenty's uniqueness check runs against all records including soft-deleted ones visible under See deleted records.

How we handle it

If we import a record with the same unique value (email for People, domain for Companies) as a soft-deleted record, Twenty restores the deleted record instead of creating a new one. We detect this condition during scoping and coordinate with the customer on whether to purge deleted records or map to existing ones before migration. This gotcha is specific to Twenty and does not apply when writing to Salesforce, but it affects which Salesforce records receive which Twenty data during migration.

medium

### API rate limits cap at 200 req/min on Organization tier

The Twenty cloud tiers enforce per-minute API quotas: 100 req/min on Pro, 200 req/min on Organization.

How we handle it

For migrations above 10,000 records, we distribute writes across throttled request batches with exponential backoff to respect limits without triggering 429 responses. Self-hosted Twenty deployments do not enforce these quotas, which can accelerate migration timelines for teams running their own infrastructure. Salesforce Bulk API is used for large engagement migrations to bypass individual API call limits.

low

### No native email sequences migrate; cadence tools require separate platform

Twenty lacks built-in email sequencing or sales engagement features.

How we handle it

Teams migrating from Outreach, Salesloft, or other sales engagement platforms to Twenty do not find equivalent cadence tools inside Twenty. When migrating away from Twenty to Salesforce, this gap in the source platform means that any cadence or sequence history is not present in Twenty to begin with. We flag this gap during discovery so customers understand that sequence history will not exist in the source regardless of the destination chosen.

## How the migration runs

1. ### Discovery and scoping

   We audit the source Twenty instance across tier (Self-hosted, Pro, or Organization), object count, custom object schema, active workflows, and record volume per object.

   What this step covers

   We identify soft-deleted records in the deleted view that may conflict with active records on unique fields. We confirm the Salesforce destination edition (Professional at $80/user or higher) and verify that all required Salesforce features are available in the chosen tier. The discovery output is a written migration scope document covering object counts, conflict status, and a recommended import sequence.

2. ### Source data extraction with visibility pre-configuration

   For cloud Twenty instances, we pre-configure the export view to expose all fields before extraction, then export in chunks of up to 20,000 records per object.

   What this step covers

   For self-hosted Twenty instances, we extract directly from PostgreSQL using a read-only connection to bypass the UI export limit and capture all fields including those hidden in the active view. We extract in dependency order: Companies first, then People, then Opportunities, then Tasks, then Notes, then Custom Objects last.

3. ### Destination schema pre-creation

   We create the Salesforce destination schema before any data is written.

   What this step covers

   This includes custom objects with matching API names, custom fields with Salesforce-appropriate types, Salesforce Account as the first standard object for the Company mapping, Contact for the People mapping, and Opportunity for the Deals mapping. We configure the Sales Process and stage values to match the source pipeline stages before migration begins. Custom object fields are created in Settings before import because CSV import creates records, not fields.

4. ### ID mapping and owner reconciliation

   We build ID mapping tables that associate each Twenty record ID with its corresponding Salesforce record ID as records are created.

   What this step covers

   This mapping is required to resolve lookups: People.companyId requires the Salesforce AccountId, Opportunities require both AccountId and OpportunityContactRole resolution, and Tasks require WhatId or WhoId resolution. We reconcile Twenty Owners against Salesforce Users by email match, placing any unmatched owners in a queue for the customer's admin to provision before record import continues.

5. ### Production migration in dependency order

   We run production migration in record-dependency order: Accounts (from Twenty Companies), Contacts (with AccountId resolved from the mapping table), Opportunities (with AccountId and OpportunityContactRole resolved), Tasks (with WhatId and WhoId resolved), Notes (with ContentDocumentLink resolved), and Custom Objects (last because they often reference standard objects).

   What this step covers

   Each phase emits a row-count reconciliation report before the next phase begins. We use Salesforce REST API for standard objects and Bulk API 2.0 for large activity volumes with batch chunking and exponential backoff.

6. ### Cutover, validation, and workflow inventory delivery

   We freeze writes to Twenty during the cutover window, run a final delta migration of any records modified during migration, then enable Salesforce as the system of record.

   What this step covers

   We deliver a written inventory of every active Twenty workflow, automation, and custom view configuration for the customer's admin to rebuild in Salesforce Flow. We do not migrate workflows as code or provide post-migration admin support for Flow rebuild as standard scope; that work is handled separately by the customer's admin team or a Salesforce partner.

![](https://flitstack.ai/platforms/twenty-crm.svg) Twenty CRM ![](https://flitstack.ai/platforms/salesforce-sales-cloud.svg) Salesforce

1. Connect
2. Map
3. Sample
4. Migrate
5. Verify

- **Contacts** 48,210 / 48,210 Done
- **Companies** 9,134 / 9,134 Done
- **Deals** 6,377 / 11,920 In progress

Example run

******Restore points · Salesforce Rollback ready

1. Before sample run Restore point 01 Restore
2. Before full migration Restore point 02 Restore
3. After verification Restore point 03 Current

Example

## How long it takes, and what drives the quote

**3-5 weeks** Typical, smaller accounts

**8-14 weeks** Large or heavily customised accounts

A quote scales with record volume, custom object count, and the number of relationship dependencies requiring lookup resolution. Every quote is scoped with you; there is no self-serve pricing.

[Get a quote for Twenty CRM to Salesforce →](https://flitstack.ai/contact)

## Why teams move from Twenty CRM to Salesforce

- Teams outgrow Twenty's v1.0 operational maturity. The CTO explicitly held off promotion until reaching v1.0, meaning the platform has a shorter production track record than established CRMs. For enterprises requiring SOC 2 Type II, multi-year SLA data, and vendor stability backed by institutional reference customers, Salesforce's 20-plus years of operational history provides a risk profile that Twenty's recent release cannot yet match.
- Salesforce's ecosystem of over 4,600 AppExchange apps and 2,000 consultants provides integration depth that Twenty's developer-first API approach cannot yet match. Teams requiring native ERP connectors (NetSuite, SAP), CPQ tools (Conga, Apttus), or revenue intelligence platforms find that Salesforce's certified integrations reduce implementation risk compared to building custom API connections to a v1.0 CRM.
- Enterprise revenue operations teams cite Salesforce's reporting engine as more flexible for multi-dimensional pipeline analysis across territories, product lines, custom fiscal periods, and non-standard attribution models. Twenty's reporting is functional for core views but lacks the advanced analytics capabilities and Einstein AI-powered insights available in Salesforce's higher tiers. Teams with complex forecast models move to Salesforce for the reporting depth.
- Twenty's workflow automation is functional but limited in complexity for sophisticated multi-step sales motions, according to early users. Salesforce Flow supports record-triggered, scheduled, screen, and autolaunched flows with a mature builder ecosystem and extensive documentation. Teams with complex multi-step sales processes requiring advanced branching logic, time-based escalations, or outbound API call actions find Twenty's automation insufficient and Salesforce's mature Flow builder a stronger fit.
- Self-hosting on Twenty is described as free but ignores real infrastructure and DevOps costs. PostgreSQL hosting, backups, security patching, and ongoing maintenance add $20-$100 per month in cloud infrastructure costs plus significant admin time. Salesforce's cloud-only SaaS model has a higher per-seat price but eliminates DevOps overhead entirely, and for teams without dedicated infrastructure engineers, the all-in cost comparison often favors the predictable Salesforce subscription over the hidden-cost self-hosted model.

Questions, answered

## Twenty CRM to Salesforce migration questions

How long does a Twenty CRM to Salesforce migration take?

Most migrations land between three and five weeks for accounts under 15,000 People and 3,000 Opportunities with no custom objects and no soft-delete conflicts to resolve. Migrations with custom objects, large task and note volumes, self-hosted Twenty instances requiring direct database export, or multiple soft-delete conflicts move to eight to fourteen weeks because of schema pre-creation complexity, lookup queue management, and Salesforce validation rule coordination.

Do Twenty Workflows and automations migrate to Salesforce?

No. Twenty Workflows and Salesforce Flow are different automation models with different trigger types, action libraries, and execution semantics. We do not migrate them as code. We deliver a written inventory of every active Twenty Workflow with its trigger, conditions, actions, and a recommended Salesforce Flow equivalent. The customer's admin or a Salesforce partner rebuilds them post-migration. Custom views and saved filters also do not migrate; we document the view configurations for manual rebuild in Salesforce List Views and Reports.

How do you handle Twenty's soft-delete uniqueness constraint during migration?

We pre-query Twenty's deleted-view for People and Companies before migration begins to identify any records with email or domain values that match active records. The customer chooses between two approaches: permanently purge the soft-deleted records from Twenty before migration begins, or accept that the restored record will be the migration target. We cannot write a duplicate Contact or Account with the same unique field as a soft-deleted record because Twenty's uniqueness check triggers a restore rather than a create. This decision is made before any data is written to Salesforce.

Can FlitStack AI migrate from a self-hosted Twenty instance?

Yes. For self-hosted Twenty instances, we connect directly to the PostgreSQL database to extract data, which bypasses Twenty's 20,000 record export limit and UI-visible-columns constraint. We use a read-only database connection with credentials provided by the customer, extract all records across all objects, and apply the same dependency-ordered migration sequence into Salesforce. Self-hosted instances also have no API rate limits, which can reduce migration timeline compared to cloud tier instances.

What happens to relationship data (People linked to Companies, Opportunities linked to People) during migration?

We preserve the relationship graph through explicit lookup resolution. For each Twenty People record, we extract the companyId, resolve it to the corresponding Salesforce Account ID using the mapping table created during the Account phase, and write Contact.AccountId at insert time. For Opportunities linked to People, we create OpportunityContactRole records linking the Salesforce Contact ID to the Salesforce Opportunity ID. Tasks and Notes with relational targets receive resolved WhatId or WhoId values. Each relationship is resolved before the child record is written, not after.

Do historical task and note timestamps preserve during migration?

Yes. We set the Salesforce ActivityDate on Task and Event records to the original creation timestamp from Twenty, preserving the chronological ordering of the activity timeline. Note body content migrates as Salesforce Note records with the original created date. The full timeline is queryable in Salesforce after migration with the correct chronological sequence intact.

Can we migrate from Twenty and keep our existing Salesforce org configuration intact?

We migrate into the customer's existing Salesforce org without overwriting existing configuration. We write records to standard and custom objects without modifying page layouts, validation rules, or existing data. The customer's admin should temporarily disable validation rules and required-field checks during the migration window to prevent record rejection, then re-enable them after migration completes. We coordinate this with the customer's Salesforce admin during the scoping phase.

## Related migrations

[![](https://flitstack.ai/platforms-trim/salesforce-sales-cloud.webp) Migrate to **Salesforce** Every source we move into Salesforce, native options and the checklist. Guide, checklist & sources](https://flitstack.ai/migrations/crm/salesforce-sales-cloud)

### [Salesforce implementation](https://flitstack.ai/services/salesforce-implementation)

[Set up Salesforce and migrate into it as one engagement.](https://flitstack.ai/services/salesforce-implementation)

[See the service →](https://flitstack.ai/services/salesforce-implementation) [**Blackbaud to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/blackbaud/to/salesforce-sales-cloud) [**ActiveCampaign to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/activecampaign/to/salesforce-sales-cloud) [**Freshsales to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/freshsales/to/salesforce-sales-cloud) [**Highrise to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/highrise/to/salesforce-sales-cloud) [**Jobber to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/jobber/to/salesforce-sales-cloud) [**SuiteCRM to Salesforce** Mapping · history · rollback](https://flitstack.ai/migrations/crm/suitecrm/to/salesforce-sales-cloud) [**Twenty CRM to HubSpot** Mapping · history · rollback](https://flitstack.ai/migrations/crm/twenty-crm/to/hubspot) [**Twenty CRM to Zoho CRM** Mapping · history · rollback](https://flitstack.ai/migrations/crm/twenty-crm/to/zoho-crm) [**Twenty CRM to Dynamics 365** Mapping · history · rollback](https://flitstack.ai/migrations/crm/twenty-crm/to/dynamics-365-sales)

## Sources: official documentation

1. [Custom Field Attributes — Salesforce Help (External ID)](https://help.salesforce.com/s/articleView?id=platform.custom_field_attributes.htm)
2. [Increase Limit of Salesforce Big Object Storage Allocation — Salesforce Help](https://help.salesforce.com/s/articleView?id=000380434\&type=1)
3. [Uploading Activity data to Salesforce — Salesforce Stack Exchange](https://salesforce.stackexchange.com/questions/15694/uploading-activity-data-to-salesforce)
4. [Sandboxes — Staging Environments for Customizing and Testing](https://resources.docs.salesforce.com/latest/latest/en-us/sfdc/pdf/deploy_sandboxes.pdf)
5. [Migrating from Other CRMs — Twenty Documentation](https://docs.twenty.com/user-guide/data-migration/how-tos/migrating-from-other-crms)
6. [Data Migration — Twenty Documentation](https://docs.twenty.com/user-guide/data-migration/overview)

Get a quote

## Moving from Twenty CRM to Salesforce?

Tell us roughly how many records and objects you have. We scope the mapping and the Salesforce setup with you, then send a quote.

[Get a quote](https://flitstack.ai/contact)
