ERP Migration 2023 Edition: What Changed

I've led three major ERP migrations in the last four years. The first was in 2019—SAP to a cloud native system. The second was in 2021, and the third wrapped up two months ago. Each one taught me something different. By July 2023, I've realized that ERP migration in 2023 is materially different from 2019. The fundamentals are the same. The details have shifted.

Let me walk through what's actually changed, and what that means if you're planning a migration.

The Old Playbook (2019)

Four years ago, ERP migration looked like this:

1. Long planning phase. 6-12 months of requirements gathering, vendor selection, and customization planning before the data even moved.
2. Big bang cutover. You picked a date, you migrated all your data on that date, and you hoped nothing broke.
3. Manual data cleansing. Teams of people manually cleaning your legacy data to fit the new system. This took weeks.
4. Heavy customization. Your new ERP was heavily customized to match your old workflows. You paid a lot for this.
5. Post-go-live chaos. First month after cutover was always chaotic. Workarounds, emergency patches, fires.

It worked. But it was expensive, risky, and slow.

What's Different in 2023

The first major change is cloud-native ERP is no longer niche. Most migrations I see now are to cloud systems—Netsuite, Microsoft Dynamics 365, or specialized cloud ERPs. That changes the game fundamentally.

Cloud-native systems are designed for rapid deployment. They have good APIs. They have standard integrations. You don't need to customize as much because the system is built to be flexible without customization. This sounds obvious, but it means the planning phase is shorter. You're not spending 6 months arguing about customization. You're configuring the system to match your process, but you're also willing to adapt your process to the system.

The second change is data migration tooling. In 2019, data migration was mostly manual or expensive consulting. Now, there are actually good tools. IPaaS platforms like Informatica, Dell Boomi, or open-source tools like Talend can handle complex data migrations. More importantly, they can handle iterative migration. You don't have to do one big bang. You can migrate incrementally, validate, fix, and repeat.

The third change is AI-assisted validation. Once your data is migrated, you need to validate it. Did you lose anything? Did numbers add up? In 2019, you had analysts going through reports checking. Now, tools can run anomaly detection automatically. AI models can compare legacy and migrated data, flag inconsistencies. This is faster and catches things humans miss.

The fourth change is phased cutover strategies. Instead of one big cutover date, more companies are doing wave-based migrations. Migrate one department or business line, get them stable, then migrate the next. This reduces risk and spreads the chaos over time.

What This Means Practically

Here's how a 2023 migration actually looks:

Months 1-2: Configuration and process readiness. You're not spending months customizing. You're configuring the system and deciding where you'll adapt your process to the system. You're identifying integrations you need. You're building your data migration plan.

Months 2-4: Iterative data migration. You run data migration pilots. Small datasets. You validate them thoroughly using automated tools. You find problems early. You have time to fix them. By month 4, you've migrated and validated all your data. You feel confident.

Month 4-5: Testing and training. Real user testing with actual production data. Training rollout starts. You're not training on assumptions. You're training on the actual system.

Month 5-6: Phased cutover. First wave goes live. Usually a smaller department or region. You work out the kinks. Second wave goes live with lessons learned. By month 6, you're fully live.

Month 6+: Stabilization and optimization. You have 3-6 months of post-go-live support. During this time, you're fixing issues, optimizing, training stragglers.

That's a 6-month migration. In 2019, this would've taken 12-18 months. The compressed timeline is possible because the technology is better and because cloud-native systems are simpler.

The New Risks

The timeline is better. The risks are different. Because you have less customization, you're more exposed to system limitations. If the system doesn't handle your workflow perfectly, you have three choices: customize (expensive), adapt your process (sometimes impossible), or build integrations (ongoing cost). You need to identify these early.

Also, because migrations are faster, the planning phase is compressed. You have less time to think through edge cases. You need to be more disciplined about documenting requirements upfront.

The third risk is data quality. If your legacy data is garbage, cloud migration isn't going to fix it. You might migrate garbage into a shiny new system. The tools are better at validating, but they can't fix fundamentally bad data. Data cleansing needs to happen in the legacy system before migration.

What This Means for Vendors Like Arvension

For us, this changes what we build. We're focused on the things that are harder in 2023:

1. Data quality and validation. Before migration is the critical moment. Tools that help companies assess and improve data quality are valuable.

2. Integration complexity. Cloud-native systems are great, but integrations are complex. Tools that make integration easier are valuable.

3. Workflow adaptation. Companies need help figuring out whether to customize or adapt. We're building better tools for process re-engineering.

4. Post-go-live optimization. The first three months after cutover are when you're running both systems. Tools that help companies sync data and avoid cutover chaos are valuable.

The Honest Assessment

2023 ERP migrations are faster, cheaper, and lower-risk than 2019 migrations. That's real progress. But they're not easy. They still require discipline, good planning, and willingness to adapt. The companies that succeed are the ones that respect the migration, not the ones that think "cloud-native ERP" means easy.

If you're planning a migration in 2023, you have better tools than ever. Use them. Plan well. Validate obsessively. And don't believe the vendor's claim that this will take four months. It'll take longer. It always does. But at least you're starting with better technology than anyone had in 2019.