“After the Migration” Is the Most Expensive Phrase in Enterprise AI

The business logic AI needs is already running. A new SAP platform doesn’t automatically create a better business.

By Gerald Schlechter, Founder and Chief Strategy Officer, enosix
Published November 5, 2026

Key takeaways

AI innovation doesn’t have to start after an S/4HANA migration. Companies can modernize SAP and improve business workflows with AI in parallel.
A new platform doesn’t automatically create a better business. The SAP business logic companies have spent decades building is already valuable today.
Waiting has a cost. Slow quotes, manual order changes and delayed customer responses don’t disappear while the migration continues.

There’s a phrase I keep hearing from companies running SAP. “We want to do more with AI, but we’re waiting until our S/4HANA migration is completed.”

I’ve spent much of my career working with SAP, so I understand the thinking. These are complex environments running critical business operations. Migrating them is a major undertaking, and nobody wants to introduce unnecessary risk into a program that’s already consuming years of effort and tens or even hundreds of millions of dollars.

What I don’t understand is why we’ve decided the migration must finish before the business can start improving.

Why AI is waiting on the S/4HANA migration

Look at what’s driving ERP strategy right now. In SAPinsider’s 2026 ERP Migration and Transformation research, SAP’s AI announcements (43%) edged out the 2027 SAP ECC maintenance deadline (39%) as the biggest outside influence on ERP strategy.

AI is becoming the reason to migrate, and the migration is becoming the reason AI has to wait.

Does that really make sense? I’m not sure it’s a strategy. It looks a lot more like a loop.

Even SAP is bringing AI to ECC and on-premises S/4HANA

Even SAP seems to be reconsidering the idea that innovation has to wait. In 2023, SAP CEO Christian Klein said SAP’s newest innovations would be reserved for its cloud environments. At SAP Sapphire in May 2026, SAP announced that many Joule assistants and agents would also become available to ECC and on-premises S/4HANA customers that have committed to modernizing, typically through RISE with SAP. The conditions matter, but the shift matters more. Even SAP is acknowledging that customers need to realize AI value before their migrations are complete.

Modernizing technology is not the same as modernizing the business

Part of the problem is that we’ve spent so long talking about digital transformation that we’ve started confusing modernizing technology with modernizing the business. They’re connected, of course, but they’re not the same thing.

ISG found in February 2026 that 49% of organizations moving to SAP S/4HANA carry their existing processes forward with little or no re-engineering. Fewer than one in five reimplement both process and technology.

A lot of companies are spending enormous time and money on a new ERP while deliberately keeping the business running the way it already does.

Sometimes that’s the right call. Changing the underlying technology while redesigning every business process is a tremendous amount of risk to take on. But if the processes aren’t fundamentally changing, why are we treating the migration as a prerequisite for improving them?

Your SAP business logic is already valuable

SAP customers have spent decades building business logic into their systems. Customer-specific pricing. Product configuration. Available-to-promise. Credit checks. Order validation. Approval rules. These aren’t simply pieces of data sitting in a database. They represent years of experience, investment and hard-earned knowledge about how a business actually works.

SAP business logic is the set of rules a company has built into SAP to decide how work gets done, from what a customer pays to whether an order can ship.

That knowledge doesn’t become valuable when it lands in S/4HANA. It’s valuable today.

What AI can do with SAP business logic before the migration

Take something ordinary: a customer calls to check on an order. Much of the information needed to answer is already in SAP, along with the rules governing availability, fulfillment and delivery. Yet the person on the phone still clicks through several screens, pieces it together, and then manually kicks off whatever comes next.

AI gives us a way to change that. Not just by generating a better answer, but by working with the actual SAP business logic and transactions to help complete the work: confirming the date, changing the order, starting the return. Connecting AI to SAP is the easy part. Making sure it does the work correctly is the hard part, and that’s exactly what the business logic is there for.

None of that requires a finished S/4HANA migration or a migration at all. And if you design it properly, that investment doesn’t have to be thrown away when you migrate. The underlying SAP environment may change, but the business problem you’ve solved doesn’t disappear. That’s a head start worth having.

The cost of waiting for the migration to finish

What’s especially concerning is how long that postponement can last.

According to ISG, 58% of SAP migrations run both over budget and behind schedule, and only 15% land on time and on budget. ASUG’s 2026 research, reported by CIO, found 61% of respondents name budget constraints as their top challenge. A separate ASUG survey on AI found just 10% have reached an enterprise-scale rollout. ASUG’s research director, Marissa Gilbert, put it plainly: “S/4HANA projects are creating the budget pressures.”

Those numbers don’t prove migration is the reason AI adoption is slow. But they should make us uncomfortable about where the time, money and attention are going. We’re spending heavily to get ready for AI while struggling to use it in the business we already have.

I’ve seen enough of these programs to know the pattern. A migration planned for two years takes three. Budgets stretch. Priorities shift. And the things that could have helped customers, sped up orders or taken manual work off people’s plates sit on the shelf.

We spend a great deal of time calculating the cost of migration. How much attention are we giving to the cost of waiting? Every quarter spent waiting is another quarter of slow quotes, manual order changes and avoidable customer frustration. Some of that has a direct financial cost. Nobody puts “after the migration” on the P&L, but it belongs there.

Run SAP modernization and AI improvement on two tracks

I’m not arguing against S/4HANA. Companies need to modernize, deal with technical debt and get ready for what’s next, and some capabilities will make more sense once that work is done. What I’m questioning is the assumption that modernization and improvement have to happen in sequence.

I’d approach this on two tracks. Keep moving on the migration, clean up the architecture, improve data quality and rethink the processes that genuinely need changing. But don’t make the rest of the business wait.

In parallel, identify workflows where AI can create measurable value using the SAP environment already running. Start with a real business problem, whether that’s slow quotes, order changes, customer service or product configuration.

And be disciplined about how you do it. SAP should remain responsible for the transactions, validations and business rules. AI needs to work within those controls, not invent its own. Otherwise, you risk creating a second version of your ERP, and probably not a better one.

I’ve always believed the real value of SAP isn’t the software. It’s the business knowledge companies have built into it. AI creates an extraordinary opportunity to make that knowledge more accessible and useful, and to turn it into action.

We shouldn’t have to wait years to begin realizing that value.

The next time someone tells me they’re waiting until after the migration to do anything with AI, I have a question.

What is it about the business problem you’re trying to solve that requires you to wait?

Frequently asked questions

Do companies have to finish their S/4HANA migration before using AI with SAP?

No. AI can work with the actual SAP business logic and transactions already running to help complete the work, such as confirming a delivery date, changing an order or starting a return. That doesn’t require a finished S/4HANA migration or a migration at all. Designed properly, that investment doesn’t have to be thrown away when you migrate.

What is SAP business logic, and why does it matter for AI?

SAP business logic is the set of rules a company has built into SAP to decide how work gets done, such as customer-specific pricing, product configuration, available-to-promise, credit checks, order validation and approval rules. It represents years of hard-earned knowledge about how a business actually works. Connecting AI to SAP is the easy part. Making sure it does the work correctly is the hard part, and that’s exactly what the business logic is there for.

What does it cost to wait until after the migration?

Every quarter spent waiting is another quarter of slow quotes, manual order changes and avoidable customer frustration. ISG reports that 58% of SAP migrations run over budget and behind schedule, and only 15% land on time and on budget, so the wait often runs longer than planned.

How can companies run SAP modernization and AI in parallel?

Work on two tracks. Keep moving on the migration, clean up the architecture, improve data quality and rethink the processes that genuinely need changing. In parallel, start with a real business problem, such as slow quotes, order changes, customer service or product configuration, and let SAP remain responsible for the transactions, validations and business rules.

About Gerald Schlechter

Gerald Schlechter is the founder and Chief Strategy Officer of enosix. He has more than 20 years of SAP experience and has advised Fortune 500 companies on complex ERP landscapes and transformation programs. His work focuses on helping companies reuse the business logic they have already built in SAP instead of recreating it for every new application or technology. Connect with Gerald on LinkedIn.

arnold® and enosix® are trademarks of enosix, Inc. SAP, SAP S/4HANA, SAP ECC, RISE with SAP, Joule and other SAP products and services mentioned are trademarks or registered trademarks of SAP SE in Germany and other countries.