The People-Process-Technology Framework Still Holds, But The Friction Has Grown

When I initially discussed change management, the main challenge was convincing enterprise teams to trust integrated data platforms and adopt new analytics workflows. Now, those same teams are expected to work alongside AI agents that can recommend actions, trigger workflows and operate across systems with less human input.

This is not just technical. It changes work, accountability and decision-making. Previously, process redesign focused on enhancing human work efficiency. Today, it also includes determining when humans should remain involved, when machines operate autonomously and how to manage exceptions. This makes the conversation far more complex, and many change management programs are still designed for an earlier phase of transformation.

Systems Are Moving Closer To Execution

Enterprise platforms are now closer to where software does more than store records or explain what happened.

Snowflake and Databricks are moving beyond analytics and toward operational decisioning. The goal is no longer just to centralize data for reporting but to bring together information, automation and action so the platform can influence how the business operates, instead of only explaining what happened after the fact.

The hyperscalers are heading in the same direction on a broader scale. Microsoft is positioning Fabric as a control layer across data, analytics and AI. AWS is connecting data, infrastructure and models into a tighter operating stack. Google Cloud is pushing further into agent-driven orchestration, moving AI closer to daily execution. Others, such as IBM, are taking a controlled route, positioning watsonX as a governed data and AI layer for hybrid environments where oversight matters as much as speed. These platforms are starting to shape decisions, not just store data.

That is why change management matters more, because employees are adapting to systems that now influence decisions and change how work gets done.

Governance Is Becoming The Operating Constraint

This is where many enterprise AI strategies stall. Not because the models are weak, but because the controls may not be ready. Data quality, ownership, lineage, policy and trust determine whether AI can operate inside the business with credibility.

Informatica is a common starting point for enterprise AI because data quality, lineage and governance determine whether anything built on top can be trusted. When data is inconsistent, poorly defined or hard to trace, AI breaks down quickly. Teradata provides governance for medium to large-scale environments, prioritizing consistency and control. Cloudera supports organizations managing sovereignty, compliance and hybrid complexity, challenges that AI alone cannot resolve. Salesforce brings the risk closest to the customer, where AI operates within live sales and service workflows and mistakes are visible right away.

The question has shifted from whether AI can deliver results to whether the business feels confident enough in its controls to allow that output to influence decisions.
_{KramerERP offers paid services to technology companies, similar to those provided by other tech-industry research and analyst firms. These services include research, analysis, advising, consulting, benchmarking, acquisition matchmaking, video and speaking sponsorships. KramerERP, an ERP industry expert, has collaborated with or is currently working with the companies listed in this article.}

ERP Is Where The Risk Becomes Real

ERP is where change management gets tested because it sits at the center of how the business runs. When ERP changes, finance, supply chain, procurement, operations and customer workflows change with it. SAP, Oracle and Microsoft Dynamics 365 show this at enterprise scale, where automation can surface exceptions, reconcile accounts and move approvals forward, but the business still needs clear controls and review points before those actions affect financial reporting, compliance or executive decision-making.

Infor, Epicor, QAD and IFS bring the same challenge into operationally sensitive environments where execution risk is higher and industry context matters more. In those settings, the closer software gets to execution, the less room there is for error.

In the mid-market, Sage, Acumatica, NetSuite and Odoo make change easier to start, but process discipline still determines whether automation creates efficiency or confusion. Newer AI-native vendors are pushing even further, but the question is still the same: how much operational autonomy is the business prepared to hand over?

Event-driven ERP Changes The Equation

Traditional ERP systems are reactive. They wait for input, process transactions and generate reports. As I wrote about last year, event-driven architecture changes that model. Instead of waiting for a report, a batch job or a manual check, the system responds as soon as an event occurs. A sudden drop in inventory can trigger replenishment. A delayed shipment can trigger an alert and rerouting. A production anomaly can trigger a maintenance response. The system responds first, and human review follows.

This is no longer theoretical. Oracle, SAP, Infor and IFS are all moving toward architectures in which ERP is no longer just a system of record but a system that can respond in real time. That changes the operating model. People are no longer only initiating each action; they are increasingly reviewing actions the system has already prompted.

This calls for a new operating approach, clearer escalation paths and greater trust in automated responses. It also means data platforms and ERP systems need to work much more closely together. Real-time decisions can break down quickly if the data is delayed, incomplete or disconnected.

Industry 5.0 Changes What We Ask Of Employees

In a previous article, I explained that Industry 5.0 brings the employee role back into the center of the transformation conversation. The question is no longer only whether people can use the system. It is what role people should play inside it.

Industry 5.0 puts human well-being, resilience and sustainability alongside efficiency. For ERP and data transformation, that means building systems that reduce cognitive load, surface the right information at the right time and preserve meaningful human involvement. If the system works as designed but people work around it anyway, the transformation has already failed.

This also demands reskilling. The workforce transition is not just an HR issue. It is a change-management and operational issue. If organizations expect employees to work differently, they have to invest in helping them do so.

Mindset Is A Variable That Matters

I have seen technically strong systems deliver average results because the people using them were never convinced the change was worth making. I have also seen simpler systems succeed because the organization was aligned, informed and prepared. Technology is only one part of the equation. Engagement and readiness decide whether it works.

Employee mindset is not soft. It is the variable that determines whether the business case survives contact with reality. Mindset does not change because leadership sends an email about strategic importance. It changes when employees understand what is changing, why it matters and how their role will evolve.

Mindset expert and speaker Ricky Kalmon described mindset as the internal operating system that determines how change is processed, accepted or rejected. His point is simple: advanced technology cannot overcome an organization that is not mentally prepared to change.

Kalmon emphasizes that most resistance to enterprise transformation is not rooted in the technology. It begins with how people process uncertainty, risk and the loss of control. Organizations that understand this spend less time forcing adoption and more time helping people reset how they think, build confidence under pressure and trust what comes next.

That is why the rise of agentic AI makes change management more important, not less. If AI agents are going to execute tasks that employees used to own, people need a clear answer to a basic question: What do I do now? If leadership cannot answer that, adoption gives way to resistance, workarounds or disengagement.

What Enterprises Still Need To Get Right Before The Opportunity Closes

Enterprise technology is entering one of its biggest platform shifts since the ERP wave of the 1990s. Businesses are moving from systems built to record activity to systems expected to respond in real time.

Companies that manage this well should move faster, reduce manual work and make better decisions. That only happens when change management is treated as part of execution, not project overhead.

Vendors can deliver the technology. Partners can help implement it. But change management still belongs to the enterprise. It has to start early and be owned. Define where people stay involved before the system starts acting on its own. Invest in reskilling before employees are forced to react. Build feedback loops early enough to catch resistance before it becomes expensive. Measure adoption with the same discipline as go-live. The constraint is rarely the technology. It is whether the business is ready to absorb what comes with it.