A published standard format and request pack
Works for any vendor, even one with no connector or API.
Move off your current systems with discovery, mapping, proof and a safe cut-over.
AI Migrate is one guided module for moving to AI from your current CRM, CPQ, incentive, ERP, HR, contract and IT systems. It discovers what you have, proposes mappings with AI that a person approves, loads in rehearsal, proves the result and supports a safe cut-over. Migration is not charged.
Works for any vendor, even one with no connector or API.
Each system is a pack on one engine, tested on recorded samples and honestly labelled by verification status.
Counts, totals and balances tie out automatically, and an unexplained difference blocks cut-over.
Rehearse as often as you like in test, and roll back any batch in reverse order.
Migration is not charged, and AI never approves a mapping or loads Production.
Search or filter. Planned items are labelled; everything else is available now.
One guided project takes you from first look to cut-over, with every step recorded and every load reversible.
24 featuresA published, versioned format and a ready-made request for your outgoing vendor, so migration works for any system before a connector exists.
12 featuresFind out what is in the old system, get an evidence-backed assessment, and map it with AI proposals that a person approves.
11 featuresCounts, totals and balances checked automatically, with a report you can sign.
5 featuresRead from files, databases, data warehouses and lakes, on the cloud or on premises, with cost controls.
15 featuresEach vendor is a data pack on one engine, so new systems are added without new code paths. Packs are tested on recorded samples and show an honest status.
19 featuresMigration is not charged and does not slow your users.
6 featuresMove customers, contacts, leads, opportunities, activities and history from your current CRM.
12 featuresMove cases, knowledge, entitlements and contact-centre history.
6 featuresMove products, price rules, bundles, discount schedules and quotes into AI CPQ.
8 featuresMove plans, quotas, transactions and closed-period history into AI ICM.
10 featuresMove contacts, consent evidence, campaigns and history from your marketing platform.
6 featuresMove ledgers, receivables, payables, assets and master data from your ERP or accounting system.
23 featuresMove suppliers, purchase documents and invoices.
8 featuresMove plans, subscriptions and invoices.
3 featuresMove workers, jobs, organisation and pay history.
19 featuresMove contracts, documents and obligations.
8 featuresMove incidents, configuration items, catalogue, knowledge and software assets.
7 featuresMove projects and resourcing.
1 featureMove warehouse master data and stock.
2 featuresRules, approvals and pricing logic come across as reviewable drafts with a side-by-side test. Code is described and flagged for manual rework, never auto-ported.
12 featuresRun both systems side by side, then switch over with a measured runbook and a way back.
7 featuresIt prepares the work. A person confirms it.
The same four building blocks, specific to boss AI Migrate.
Approvals protect the move: a mapping is approved by someone other than its author, finance and incentive loads in Production need a second approver, and go or no-go is a Workflow approval.
Readiness, progress and reconciliation are visible per project.
Datasets
Dashboards
Migrate has its own API and read tools for AI, and embeds in each hub's Integration area.
/v1 migration resources and admin migrate routes
Events
Connectors
This is the migration module: packs by family are listed on this page.
Approvals and the what-needs-me inbox work on the phone through the one AI app; project planning and mapping are on the web.
crm: accounts, contacts, deals, history
cpq: price books, products, quotes
icm: plans, quotas, history
hcm: workers and org
clm: contracts
itsm: incidents and configuration items
workflow: approvals
integration: connectors and agent
No. Migration usage is excluded from billing and allowances, with a cut-over date required and a grace period after it.
Use the standard data specification and request pack: templates, a draft request letter, how-to guides and a secure delivery link, then the intake validator checks what you receive.
No. AI proposes and explains. A person approves every mapping and every Production load.
Yes. Batches roll back in reverse dependency order, and a rollback is refused if the records have been used since.
Packs are tested on recorded samples and each pack shows its status: unverified, fixture-tested or live-verified.
Rules and approvals become reviewable drafts with a side-by-side test. Custom code is described and flagged for manual rework, not auto-ported.
Yes, as often as needed in test, with automatic reconciliation each time.