From prediction to practice: What SAP teams actually did in 2026

We published that report with a set of predictions about how organizations would handle the move off ECC (SAP ERP Central Componentz). Six months later, the deadline hasn’t moved, the pressure has increased, and the honest question is whether intent has turned into action. For the most part, it hasn’t. The gap between what teams said they would do and what they have actually put in place is the story of this year. It’s a big gap, and we need to narrow it before it becomes a chasm.

The problem was never testing

Almost everyone we interviewed was committed to migrating, 93.8% of respondents, but that’s the easy part. It’s easy to say something should be done. It’s harder to turn that intention into action. The debate has genuinely moved from “if” to “how,” and that’s real progress. But there is still a long way to go.

Knowing that you have to do something is a long way from being ready to do it. Fewer than one in five organizations had a mature Quality Engineering practice, the discipline that tells you whether the business will still be running the Monday after go-live. More than half are still on ECC, and the end-of-support date is getting closer.

It isn’t a testing problem or a tooling problem. It’s a timeline, cost, and risk problem, and QE&T is the only discipline that addresses all three at once.

Where intent outran execution

This is where “prediction to practice” earns its title.

Take AI. Six months ago, 25% of organizations were using AI to support cloud migration. That figure hasn’t meaningfully changed, even though 57.2% now report active or partial use of AI somewhere in QA. Adoption has stalled exactly where deadline pressure is highest. Meanwhile, 39.3% still conduct change impact analysis manually. The task AI is best suited to is still being performed manually by nearly four in ten teams.

The reasons are unglamorous: data privacy concerns, cited by 50%; fear of overreliance, cited by 45.3%; cost; skills; and, above all, time. Plans are tight, and people are already overloaded. Nobody wants to introduce a new risk into a program that is already carrying too many.

What has actually happened is ad hoc use. People are reaching for Copilot outside the plan to stay ahead of their own workloads. That trades an unknown amount of risk for an uncertain payoff, and the benefits don’t compound. If your ways of working aren’t mature, can you introduce something new and still have it working in six months? In twelve? And who is checking what employees are sharing with shadow AI tools?

The same pattern repeats elsewhere. SAP BDC adoption stands at 32.1%, but 57.7% of those users know about the AI data-integrity features and haven’t activated them. RISE with SAP has near-universal awareness, but only 29.9% report active use of the methodology. Hybrid testing is the norm at 44.7%, while only 25.5% have achieved full automation. In every case, the capability has been purchased, understood, and left idle.

I have seen organizations buy tools and assign them to an in-flight project, only for no one to have the time to focus on them. When the project ends, so do the funding and investment, and the knowledge gained is lost.

Automation offers the clearest and least controversial entry point for AI. With only a quarter of organizations having achieved comprehensive automation, it’s also where the need is most obvious. That’s the thinking behind the Test Automation Factory and Business Assurance for SAP: shift left from the start, use AI-driven change impact analysis, and work with an independent QA partner who owns the answer.

What it’s worth when you do it

The payoff isn’t theoretical. More than 75% of organizations credit QA and QE integration with improved business outcomes. More than 75% report that test automation has a high or very high impact on business agility in RISE operations. Nearly 60% of AI users see significantly improved QA consistency, and around half say AI has significantly improved the efficiency of security testing. That matters when more than one in three teams are now testing security weekly

Translated from percentages into practical terms, the team stops arguing about which 2,000 tests exist and starts knowing which 200 matter. That’s the difference between a regression cycle you dread and one you can run on demand.

There is a forward-looking signal in the data as well. A total of 72.6% plan to increase accountability for their SIs and OEMs. Ownership is shifting, and clear ownership is what independent assurance is designed to provide.

One thing to do

If you take one practical step from this, use RISE quality gates to validate data readiness at every phase rather than discovering problems at cutover. The goal was never better testing. It’s business assurance: knowing that the business still works.

Download the full 2026 report or talk to the Sogeti SAP Business Assurance team.

Which of your 2026 predictions has actually made it into practice?