Switching from Salesforce to Lightfield: what moves, what you rebuild, and what to settle first

Lightfield

The short version

Lightfield is an agentic CRM. It captures every customer interaction, organizes that history so agents can work from it, and gives agents the tools to do consistent, good work. If you run Salesforce today, a move comes down to two questions: which data comes across, and which processes you rebuild.

We can migrate agreed Salesforce records, custom objects, fields, and relationships into a Lightfield data model. The logic around those records is assessed separately: formulas, automation, validation, approvals, permissions, and integrations. Importing the data doesn’t recreate any of them.

This guide covers migrating data and rebuilding selected processes in Lightfield.

Why teams look at moving

Salesforce was designed around records that people maintain by hand, so an AI layer added on top inherits whatever is incomplete, stale, or contradictory underneath. In Lightfield, records update from captured calls, emails, and meetings, and custom objects and relationships can follow how your business works.

Which Salesforce data can move

“Supported” means product guidance supports it. “Scoped” means it depends on an agreed export, configuration, or implementation scope.

DataStatusWhat to settle

Accounts

Supported

Keep source IDs and agreed field mappings. Matching on name alone can create errors

Contacts

Supported

Decide how to handle missing, shared, or duplicate email addresses

Opportunities

Supported

Resolve the account, stages, owners, amounts, and the actual deal contacts. Include closed deals, renewals, and expansions if they’re in scope

Custom field values

Supported

Map the values and their types. Field configuration and business logic don’t travel with the values

Custom objects

Scoped

Model the objects and relationships in Lightfield, then migrate the agreed records. This is implementation work, and it doesn’t copy a schema automatically

Relationships and hierarchies

Scoped

Needs source IDs and explicit mappings. Test association accuracy as well as record counts

Notes, tasks, and files

Scoped

Decide which types are included. The source has to provide the files and their associations

Logged emails and activity history

Scoped, with limits

Standard exports generally don’t include full logged-email history. Keeping the content doesn’t establish native timeline, author, or audit-history fidelity

Original creation dates

Kept in dedicated fields

Lightfield’s system Created At shows when the record was created in Lightfield, which for an import is the import date

Connecting email and calendar adds interaction context. It doesn’t rebuild every custom field, stage, owner, association, or historical change from the old CRM. Connecting a mailbox and importing a CRM are separate steps.

Imports run through the agent in chat. You attach your export, review the plan for field mappings and record matches, and approve it. For a large import, you review a saved test batch before the rest is written. Keep your source IDs, account domains, contact emails, and the columns that link records together, because they’re how relationships are preserved.

What you rebuild

What you rebuild
Salesforce requirementLightfield capability to look atWhat to qualify

Flows and process automation

Automations, workflows, record-change triggers, schedules, webhooks, tasks, and integrations

Rebuild the agreed process

Formula fields

Calculated fields

Check formula compatibility and dependencies

Roll-up summaries

Relationship-based calculation

Rebuild and test them. Importing related records doesn’t create the calculation

Validation rules

Validation rules, which hold a record back until required fields, files, or linked records are in place

Lightfield's validation rules can enforce requirements like these, such as required fields, files, or linked records before a record is saved or advances a stage. List yours early so they can be assessed

Approval processes

Review of supported drafts, and tasks for human decisions

We can set this up as an automation or a chain of automations. List each approval early so it can be assessed

Which of these features you can use depends on your plan, so confirm it when we scope the work. A feature existing doesn’t mean it behaves like its Salesforce counterpart.

Formula and roll-up values

An exported value and the rule that computes it are different things. The value is a snapshot. It doesn’t carry the formula, its dependencies, or its recalculation. For each formula or roll-up, you have three options:

  1. Keep the source value in a normal field.
  2. Rebuild the calculation in Lightfield where that’s supported.
  3. Keep the calculation in another system and sync its result under an agreed integration design.

We don’t promise to translate all Salesforce formula syntax or roll-up behavior directly.

Settle these before you import

  • Agree which objects and how much history is in scope.
  • Get complete exports, including closed deals, renewals, expansions, and every relevant pipeline.
  • Keep source record IDs and association tables.
  • Map object types, field types, select values, currencies, pipelines, and stages.
  • Decide how owners map, including people who’ve left and users who aren’t in Lightfield yet.
  • Keep original business dates in clearly named fields.
  • Decide which formulas and processes you’ll rebuild and which values you’ll keep as snapshots.
  • Review duplicates and shared-email cases.
  • Define the overlap, the cutoff, the final delta, and your rollback or archive plan.

Running both systems during the move

Running in parallel is supported through scoped deployments. A pilot can use one-way or two-way sync. Before an overlap starts, we define which records and fields sync, which system owns each field, how conflicts are handled, and what has to be true before cutover. Edits made in Salesforce don’t appear in Lightfield unless a sync is scoped. A common pattern is an initial export plus a final delta close to cutover.

When to keep Salesforce for now

A team with heavy Salesforce-specific governance can reasonably stay on Salesforce until it has confirmed Lightfield can meet those requirements. List the Flows, validation rules, approvals, permissions, and integrations you couldn’t run without, and test each one before you plan a cutover.

How long a migration takes

The average enterprise deployment takes 28 days. We confirm your timeline after reviewing your data model, volume, historical content, integrations, and testing needs.

Your plan separates source preparation, model setup, the first import, historical content, testing, corrections, integrations, and the final delta and cutover. Managed migration is available on request, with scope agreed before work starts.

You don't pay until deployment is done. We only start billing once your deployment is complete.

FAQ

Can Lightfield import our Salesforce custom objects?

Yes, as scoped implementation work. We model the objects and relationships in Lightfield, then migrate the agreed records. A Salesforce schema isn’t copied automatically.

Do Salesforce Flows move over?

No. Importing records doesn’t recreate Flows. We rebuild the agreed process with Automations, workflows, record-change triggers, schedules, webhooks, tasks, and integrations.

What happens to validation rules and approvals?

We assess them separately from the data. Lightfield has validation rules that require fields, files, or linked records before a record saves or changes stage, but we don’t claim they match every Salesforce rule or approval chain.

What about territory management and multiple Salesforce orgs?

Lightfield supports multiple pipelines within a single workspace, but not multiple instances. Most customers get the functionality they need from multiple pipelines, since running separate Salesforce orgs was often a workaround in the first place. If you're consolidating several orgs, each one needs its own assessment: source IDs, schema conflicts, stage mappings, access boundaries, ownership, and which system controls each value. For territory management, we'll scope your setup with you before promising anything.

What happens to formula and roll-up fields?

You can keep the value as a snapshot, rebuild the calculation where supported, or keep it in another system and sync the result.

Can we run Lightfield next to Salesforce during the move?

Yes, through a scoped deployment that can use one-way or two-way sync. We agree the records and fields in scope, field ownership, conflict handling, and cutover criteria before the overlap starts.

Does Salesforce stage history come across?

Not as a general replay. Source stage and property history can be archived, and original dates can be kept in dedicated fields.

How long does a migration take?

We estimate it after reviewing your data model, volume, history, integrations, and testing needs.

Does Lightfield offer managed migration?

Yes, on request, with scope agreed before work starts.

What should we have ready before the first call?

Complete exports, including closed deals, renewals, expansions, and every relevant pipeline, plus source record IDs, association tables, and a list of the formulas and processes you want rebuilt.

Next step

Book a demo and bring your Salesforce object list, plus the Flows, validation rules, and approvals you rely on.