Today’s enterprise runs on more than human workers: Service accounts automate workflows. Bots execute tasks. AI agents process requests, access systems and take action around the clock. They have no badge, no login, no seat.

Industry data puts non-human identities at anywhere from 25 to 50 times the number of human users in an average enterprise. Those systems do real work and consume real resources. They require governance, but none of them has a seat license.

Most enterprises still budget for software based on head count and planned employee growth. But agents — not humans — are now the fastest-growing group in most environments. That creates a structural problem: The activity driving real value inside the platform is not the activity the pricing model is designed to measure.

When you price software by head count and the workforce increasingly has no head, you get a disconnect between what you pay for and what you actually get. That assumption is collapsing.

Computerwoche: Many companies are facing a generational shift, as baby boomers, including numerous SAP experts, are retiring. How great is the risk of this vital enterprise knowledge being lost?

Uwe Hartmann: You have to look at the topic from different perspectives. The knowledge of experienced employees is not limited to technical operation, but above all includes an understanding of how a company’s processes function and how they are controlled in the SAP system.

Generating an order confirmation, for example, is relatively simple. But when this results in a delivery, special labels with barcodes are printed, financial accounting is integrated, and receivables or liquidity planning is initiated, then we are moving into a significantly more complex environment.

These connections aren’t understood overnight. It requires years of experience and in-depth knowledge of the company. The problem I currently see in many companies is that key personnel are retiring, and often there’s no successor yet.

Does the fact that many companies are in the midst of their S/4HANA transformation increase the pressure?

Absolutely. According to current figures, more than half of companies have not yet finished their migration to S/4HANA. If experienced employees leave, significant risks arise.

Studies by Gartner and McKinsey show that a large proportion of S/4HANA projects fail to meet their original cost or timeline targets. Every company faces a real challenge here.

What about after a successful S/4HANA migration? Will SAP experts leaving the workforce have less of an impact then?

No, not at all. The migration to S/4HANA alone won’t solve the problem. Many companies see it primarily as a prerequisite for continued maintenance and support from SAP. But this is similar to a smartphone update: You have the latest version and remain technically up-to-date, but the real added value doesn’t automatically materialise. This is precisely the question many companies are asking themselves: What does S/4HANA offer us beyond the mere maintenance requirement?

In my opinion, the real work begins here. Looking at the evolution of artificial intelligence, one can see how rapidly requirements are changing. Two years ago, AI was still an abstract topic for many. Today, it is increasingly being used productively in business processes. At the same time, SAP is also continuously developing its platform and has invested heavily in AI technologies in recent months.

But it is crucial to use these opportunities judiciously. AI only unfolds its full potential when it is meaningfully integrated into processes. Business processes in companies will very likely look completely different in three or four years than they do today. S/4HANA creates a solid foundation for this, but nothing more. Companies must build on this foundation. This will continue to require experienced employees who understand business processes and can effectively utilise new technologies with the necessary expertise.

Kristian: I want to be clear—if someone has been using Autodesk software without the appropriate licenses, that usage needs to be addressed. Installation and use of Autodesk software requires a valid license.

Resolving these situations often includes purchasing genuine Autodesk software to cover non-compliant use and, in some cases, past usage as well. That’s an important part of the conversation, and we want to be transparent about it.

But our goal isn’t to punish customers or create unnecessary hardship for them. It’s to understand what happened, explain what we’ve identified, and work toward a resolution that brings the customer onto genuine Autodesk software.

There are usually two types of situations we see—first, cracked software. It’s not legitimate Autodesk software, may not work as intended, and often includes malware and viruses. The other may involve genuine Autodesk software, but it’s not properly licensed.

It might be surprising, but the use of non-valid software isn’t always intentional. In reality, some customers purchase software through channels they believe are legitimate. They may not realise there’s an issue until much later.

Every situation is different, which is why we work directly with customers to understand the circumstances and identify an appropriate path forward. Our focus is to help customers become properly licensed users while finding a fair and practical resolution whenever possible.

Support deadlines have a way of focusing attention. They also have a way of narrowing choices. When a critical platform approaches the end of vendor support, the instinct is often to move quickly toward the recommended next step. In some cases, that may be the right decision. In others, it can lead to organisations into unnecessary cost, disruption and loss of flexibility before they have fully evaluated the alternatives.

The risk for IT leaders is treating the vendor product roadmap as if it were automatically the enterprise roadmap. The two may align, but that alignment should be proven, not presumed. A vendor’s timeline is not, in itself, a business case for broad infrastructure change.

A vendor’s priorities are shaped by product strategy, recurring revenue, portfolio simplification and platform consolidation. An enterprise’s priorities are shaped by uptime, security, cost control, application performance, operational continuity and business agility. Those priorities can overlap, but they are not identical and treating them as interchangeable can result in decisions that solve for the vendor’s direction more than the customer’s needs.

This is where CIOs and infrastructure leaders need to step back and ask a more fundamental question: what business outcome are we trying to achieve, and does this change materially advance it?

If the answer is improved resilience, better automation, simplified management, stronger security or a more cloud-like operating model, then modernisation may be justified. But, if the answer is simply maintaining continuity or preserving support status, organisations should examine the full range of viable paths before committing to a major platform shift. The right response is not always the most disruptive one.

Your cloud provider probably spreads your month across services, regions, and pricing dimensions. Almost none of the line items carry names a department head would recognise. You won’t get anywhere reading it as-is, but all three platforms include a view that translates the raw feed into something you can question.

In AWS, you can do this in Cost Explorer, grouping the month’s spend by service to see where the money concentrates. Azure keeps the equivalent in the cost analysis view inside Cost Management. On Google Cloud, the built-in billing reports chart spend by project, service, SKU, or location. Consider pulling the last full month and sorting by cost, then working down the top 10 entries until you can say what each one is, which workload it serves, and who would notice if it disappeared. (Fair warning, the first sort may produce results that are a bit startling.)

Once you’re done, put anything you can’t explain on the audit list and work from there.

Clients often assume an audit letter means the publisher suspects something. Usually it does not. Audits are a revenue channel, run on a schedule, and the volume is substantial.

Flexera’s 2026 State of ITAM Report, drawn from a survey of ITAM professionals conducted in early 2026, found that nearly half of organisations — 48 percent — received a software audit in the prior twelve months. Microsoft was the most frequently reported auditor at 64 percent of those audited, with Oracle close behind at 50 percent. Oracle’s reported audit activity jumped roughly fourteen percentage points year over year. Adobe and VMware activity also rose.

The cost side is where it stops being a compliance annoyance and becomes a balance-sheet event. Forty-four percent of organisations reported spending more than $1 million on software audits over a three-year period. In earlier Flexera data, the share of companies reporting audit costs above $10 million nearly doubled year over year, from 7 percent to 12 percent.

Two conclusions follow. First, receiving a letter is not evidence that you did anything wrong. Second, the process that follows is expensive enough that it deserves the same seriousness as any other bet-the-budget dispute — from hour one.