Migration Guide · Updated 2026

Migrating from Anaplan to Pigment — what it actually involves

We've completed full Anaplan-to-Pigment migrations — model mapping, data hierarchy translation, integration rebuild, parallel testing, and cutover. This guide covers what migration actually involves, what most people get wrong, and how to decide if it's the right move for your organization.

Most articles about Anaplan-to-Pigment migration are written by people who haven't done one. They'll tell you about the platforms' differences, recommend you "develop a migration strategy," and suggest you "engage with an experienced partner." That's not what this is.

We've run these migrations. We've mapped model architectures, rebuilt integrations from scratch, and managed the part that's actually the hardest — getting an entire organization to cut over from a platform they've been running for years. This guide covers what we've learned from doing it.

The short version: migration is absolutely feasible, and in the right situations it's the right call. But it's not a copy-paste job, it takes longer than most people expect, and the technical build is not the hard part.

The question we always ask first: Is the problem your platform, or is it your model architecture? A badly designed Anaplan model won't become a good Pigment model just because you migrate it. Sometimes the right answer is a rebuild on the platform you're already on — and we'll tell you that honestly if it applies to you.

Should you migrate?

Migration is the right move in specific situations. It's not the right move in others. Here's an honest framework for thinking through it.

Quick reference
Migration readiness at a glance
Migrate to Pigment if…
  • Your business has changed significantly and your Anaplan model hasn't kept up — structural changes are expensive and slow
  • Your team is waiting on IT to build or maintain integrations, and that bottleneck is slowing down your planning cycle
  • You want AI actively involved in your planning workflow now — Pigment's AI Modeler is production-ready; Anaplan's is still maturing
  • Your admin costs are accumulating — Polaris fees, middleware licensing, dedicated Anaplan admin headcount
  • Your planning function is growing and you want a platform your FP&A team can run without constant specialist support
Stay on Anaplan if…
  • You have a large, active planning CoE with 10+ analysts deeply fluent in Anaplan — the switching cost will exceed the benefit
  • You're a global enterprise with 100+ legal entities requiring complex multi-currency consolidation that Anaplan handles well
  • Your current model is stable, well-maintained, and your business model is predictable — if it's working, "better" isn't always worth the disruption
  • You're on Polaris and have invested in integration infrastructure — the Pigment advantage in those areas is smaller for you
  • You've accumulated 5+ years of complex custom logic that's deeply embedded in downstream reporting

If you're genuinely on the fence, the most useful next step is a 30-minute call where we'll walk through your specific situation. We have no incentive to push you toward Pigment — we implement both, and we'll give you a straight read.

It's not a lift-and-shift

The most common misconception we encounter is that Anaplan-to-Pigment migration is a matter of exporting your Anaplan model and importing it into Pigment. It isn't. The two platforms have fundamentally different data model architectures.

Anaplan is built around lists, modules, and cells. Everything in an Anaplan model is structured around these primitives. Your model logic, your formula dependencies, your integration points — all of it is Anaplan-specific. None of it transfers directly.

Pigment has a different structural approach with its own formula language, its own data hierarchy model, and its own way of thinking about dimensions and relationships. Migrating means mapping your Anaplan model components to their Pigment equivalents — which requires understanding both platforms deeply, not just one.

There's also a subtler point worth making: some of your Anaplan model complexity shouldn't be rebuilt at all. Over time, Anaplan models accumulate logic that was designed to work around Anaplan's constraints — cell limits, sparsity management, calculation order dependencies. That logic isn't modeling your business; it's managing your platform. When you migrate, those workarounds shouldn't come with you. A good migration is an opportunity to build the model your business actually needs, now that you know more about it.

A note on "we'll just rebuild it in Pigment": Clients sometimes arrive with a mental model of migration as "same thing, different platform." The better framing is: use the migration as an opportunity to rethink how your model is structured from first principles. You'll end up with a cleaner result and a faster build.

How a migration actually runs

Every migration is different in scope, but the phases are consistent. Here's how we run them.

01
Discovery & user mapping
Weeks 1–3

Before we touch a single formula, we map the full scope of who is using Anaplan and what they're using it for. This is the step most migration projects skip — and it's the one that causes the most problems at cutover.

Most organizations have more Anaplan users and connections than they realize. The finance team is the obvious audience, but there are often downstream consumers — operations teams pulling reports, HR running headcount queries, execs with dashboard views — who weren't in the original scoping conversations. When you cut over without knowing they exist, you accidentally disrupt them.

We get every Anaplan user in the room at the start of the project. Not just the primary stakeholders — everyone. This is the single most important thing you can do to protect a smooth cutover.

  • Complete user audit: every person accessing Anaplan and how
  • Integration inventory: every system connected to Anaplan, every data flow
  • Report and dashboard inventory: every output downstream consumers rely on
  • Use case prioritization: what gets migrated first, what comes later
02
Architecture design
Weeks 2–4

With a full picture of the existing model, we design the Pigment architecture — not as a mirror of Anaplan, but as the model your business actually needs. We map list structures to Pigment dimensions, translate module logic to Pigment's data model, and identify which Anaplan complexity carries over versus which gets left behind.

We also make integration decisions here: which Anaplan Connect or ETL connections become Pigment native connectors, and which require a different approach. For most enterprise SaaS companies, the majority of integrations can be rebuilt as point-and-click Pigment connectors — often faster and cheaper than the original build.

  • Pigment data model and dimension structure design
  • Formula translation map: Anaplan logic → Pigment equivalents
  • Integration architecture: native connectors vs. API where needed
  • Output design: recreating dashboards and reports in Pigment
03
Build & data validation
Weeks 4–14 (varies by scope)

We build the Pigment model iteratively, module by module. Integration connections go in early so data is flowing as each section is built — this makes validation faster and catches data issues before they compound. Your team reviews builds in progress, not just at the end.

Data validation runs continuously throughout the build: Anaplan outputs vs. Pigment outputs, number by number, for every key calculation. We don't move to cutover planning until validation is clean.

  • Iterative module builds with your team reviewing at each stage
  • Integration connections built early — data flowing from day one of build
  • Continuous output validation: Anaplan vs. Pigment side by side
  • User acceptance testing with the people who will actually use the model
04
Parallel running
2–4 weeks before cutover

For a defined period, both Anaplan and Pigment run simultaneously. Users do their actual work in Pigment — but Anaplan is still live as a fallback while confidence builds. Outputs are compared in real time. Any discrepancy gets resolved before cutover.

Parallel running is also where your users build muscle memory in Pigment before the pressure of the cutover. The teams that struggle most at cutover are the ones who didn't get enough parallel running time. We push for a minimum of two planning cycles in parallel wherever possible.

  • Both platforms live and fed by the same data sources
  • Users completing real planning work in Pigment, not just testing
  • Daily output comparison and discrepancy resolution
  • Training sessions for all user groups identified in discovery
05
Cutover & decommission
Cutover day + 30-day stabilization

Cutover is a defined event with a clear checklist, not a gradual drift. Anaplan is turned off for active use on a specific date, with all users notified in advance and trained in Pigment. We stay closely involved through the first 30 days — catching anything that surfaces once real-world usage begins at full volume.

Anaplan is typically kept in read-only archive mode for 90 days post-cutover, giving users a safety net for historical reference without it being a live production system. After 90 days of clean Pigment operation, it's decommissioned.

  • Cutover date set and communicated across all user groups
  • Anaplan moved to read-only archive (not immediately deleted)
  • 30-day intensive support window post-cutover
  • 90-day archive period before full decommission

What goes wrong — and how to prevent it

02
Rebuilding Anaplan's workarounds instead of the business logic

Anaplan models accumulate complexity that exists to manage Anaplan's constraints — not to model the business. If you rebuild it all in Pigment, you carry the debt with you. Identify what's business logic and what's platform management before the build starts.

✓ Prevention: Architecture review in Phase 2 — what carries over vs. what gets left behind.
03
Skipping or shortening parallel running

Teams under timeline pressure are tempted to shorten the parallel running period. This almost always creates problems at cutover. Users who haven't done real work in Pigment yet are suddenly dependent on it — with no fallback. Two planning cycles minimum.

✓ Prevention: Build parallel running time into the project plan from day one. Treat it as non-negotiable.
04
Treating migration as an IT project

Migrations that are owned by IT rather than Finance tend to produce technically correct models that Finance teams don't trust or use. The people who will live in Pigment every day need to be involved in building it — in UAT, in parallel running, and in the handoff.

✓ Prevention: Finance leads the project. IT is a stakeholder, not the driver.

Migration isn't the only path forward

If your Anaplan model isn't working well, there are two routes. The right one depends on your situation.

Migrate to Pigment
Move to a platform built for where your business is going

The right path if Pigment's AI capabilities, native integrations, or flexibility are material to how you want to work — and if you've concluded that Anaplan's architecture is the constraint, not just the model design.

  • AI-native planning workflow from day one
  • Eliminates IT integration overhead permanently
  • Removes Polaris and middleware from your cost structure
  • Team can self-serve model changes without specialist support
Rebuild on Anaplan
Clean rebuild of what you know now, on the platform you're on

The right path if Anaplan is genuinely the right platform for your organization — but your current model reflects an older version of your business. A clean rebuild is often more valuable than a migration, and faster than people expect.

  • No platform learning curve for your team
  • Keeps your existing Anaplan ecosystem and integrations
  • Applies years of model-building experience from scratch
  • Often faster to deliver than a full platform migration

Not sure which path is right?

Tell us where you are — current platform, what's not working, and what you're trying to build. We'll give you a straight answer on whether migration, rebuild, or something else makes the most sense.

No sales pitch. We'll tell you honestly if migration isn't the right call.