ERP and CRM Project Management in the Age of Artificial Intelligence
Why Has Planning Become More Difficult Than Ever?
The greatest challenge in ERP and CRM projects today is not technical; it is the gap between the speed created by artificial intelligence and the speed at which people can adapt.
ERP and CRM projects have always carried their own inherent difficulties. Long durations, high costs, multiple stakeholders, changing expectations, and organizational resistance are all part of their nature. However, in recent years, a new reality has become increasingly clear in the field:
Artificial intelligence, while accelerating ERP and CRM projects, is simultaneously making planning more difficult.
At first glance, this may sound contradictory. After all, artificial intelligence accelerates code development, simplifies documentation, shortens analysis processes, and automates many manual tasks. Despite this, especially in large-scale ERP and CRM transformations, accurately estimating timelines, delivery dates, and costs has become more difficult than in the past.
This article is not written to downplay the benefits of artificial intelligence. On the contrary, it is written to discuss which risks must be managed in order for these benefits to become sustainable.
In this article, I aim to evaluate—based on real-world observations—how artificial intelligence has transformed ERP and CRM project management, why it has weakened classical planning approaches, and along which axes project management must be rethought in this new era.
What Was Classical ERP and CRM Project Planning Based On?
Traditional ERP and CRM project management relied heavily on past experience. Projects executed in similar industries, previous customer references, man-day calculations, and phase-based planning habits formed the foundation of time and cost estimations.
The core assumption behind this approach was simple:
“What was done yesterday will largely be done in a similar way tomorrow.”
Analysis, development, testing, and go-live phases were clearly separated. Scope was defined upfront and kept as stable as possible. Of course, changes occurred—but their number and impact generally remained within predictable limits. As a result, project plans could provide a reasonable level of foresight despite uncertainty.
Why Did Artificial Intelligence Disrupt This Balance?
Artificial intelligence introduced two simultaneous effects into ERP and CRM projects. The primary reason planning has become more difficult lies in this duality.
On one hand, artificial intelligence:
- Accelerated code generation
- Simplified documentation
- Shortened prototype and POC cycles
- Enabled analysis outputs to become visible much earlier
On the other hand, it:
- Made it almost impossible to keep scope stable
- Continuously pushed user and management expectations upward
- Strengthened the reflex of “if we can do this, we should also add that”
With copilots, AI-driven analytics, and agent-based approaches, stakeholders now see tangible outputs at much earlier stages. While this increases decision-making speed, it also dramatically increases the number of decisions being made.
And this leads to a critical realization:
Artificial intelligence does not accelerate the work itself—it accelerates decision production.
As decision production accelerates, the need for revisions, redesigns, retesting, and replanning increases naturally.
Expectation Misalignment: Projects Trapped Between Management and Users
One of the most challenging impacts of artificial intelligence on project management is the emergence of unrealistic and mutually conflicting expectations among different stakeholder groups.
Management Perspective: “It Should Be Faster and Cheaper Now”
Senior management and budget owners, influenced by the general AI narrative, increasingly approach projects with the following mindset:
“If AI is writing the code, why are these projects still taking so long and costing so much?”
From this perspective:
- Projects that previously took 12 months are now expected to finish in 5–6 months
- Budgets are expected to shrink significantly
- Consultant and developer headcounts are assumed to be reducible
However, the reality in the field is different. While AI accelerates development, it does not eliminate integration complexity, security requirements, data quality challenges, testing effort, or organizational adaptation costs. In many cases, these areas actually become more complex.
User Perspective: “A Perfect System Should Finally Arrive”
On the user side, a different expectation gap emerges. Artificial intelligence creates the perception of systems that are nearly flawless and fully automated.
With this mindset, users expect:
- Data entry to disappear entirely
- All exceptions to be resolved automatically
- Human approvals to become unnecessary
Yet AI is a powerful assistant—it does not eliminate process ownership or human responsibility. When this distinction is not communicated clearly, disappointment and resistance after go-live become inevitable.
For project teams caught between these two extremes, the first step of project management in the AI era is not technical—it is expectation management.
Between AI Speed and Human Reality
Over the last 5–6 years, particularly after the pandemic, another significant shift has been observed in projects: a transformation in how people perceive work and delivery responsibility.
The widespread adoption of remote work models and the constant reproduction of narratives around “lightweight work” and “unlimited freedom” have made the inherent difficulties and continuity requirements of work less visible.
ERP and CRM projects, by nature, are long-term, detail-oriented, and demand intensive effort at certain phases. Integrations, data cleansing, testing cycles, and pre–go-live critical periods require not only technical competence but also focus, accountability, and—at times—extra effort.
This reality has not changed.
What has changed is the level of mental resilience shown toward it.
While AI accelerates some technical tasks, delivery discipline on both the partner and customer sides has weakened. The perception that “AI will speed everything up anyway” has led to underestimating the workload fluctuations across project phases. As a result, deliveries that are technically feasible begin to slip due to human-side disconnects.
This issue does not stem from rejecting the value of work—but from redefining the role of work incorrectly in life priorities.
AI does not eliminate the need for focus and prioritization in ERP and CRM projects. In fact, in some areas, it makes this need even more pronounced.
The Loss of Human Touch: The Invisible Impact of Remote Work and AI
The transformation driven by AI in project management has not only changed technical processes—it has fundamentally altered human interaction.
Today, consultants often understand user work solely through screens, documents, and online meetings. The daily reality of a warehouse worker picking goods or an accountant processing invoices is no longer directly experienced—it is translated through descriptions.
In the past, time spent on-site enabled teams to:
- Understand why processes worked the way they did
- See the real needs behind steps that appeared illogical on paper
- Identify where improvements truly created value
Remote work and AI-supported analysis can partially fill this gap—but they cannot fully replace contextual understanding.
This directly affects expectation management. When users do not build a human relationship with consultants, their demands shift from negotiable needs to rigid requirements. On the consultant side, lacking emotional context leads to more mechanical and rigid solutions.
When strong human interaction exists, benefit–cost–need trade-offs can be discussed more naturally, mutual acceptance emerges, and the expectation of a “perfect system” gives way to the right system.
Thus, AI-era project management is not only about using technology correctly—it is about consciously redesigning human interaction. Hybrid models, especially physical presence during analysis and design phases, have become critical for both expectation alignment and delivery quality.
Loss of Code Ownership and the “Black Box” Risk
AI-driven code generation introduces a subtle but serious risk: the erosion of code and design ownership.
In ERP and CRM systems, financial calculations, pricing rules, inventory logic, and credit limits must be explicit and traceable. With AI-generated code, teams may gradually proceed without fully understanding why the code behaves the way it does.
When issues arise and AI is again used for troubleshooting, the problem deepens. The code works, tests pass—but the system evolves into something no one truly owns. This severely undermines long-term sustainability.
The core principle must be clear:
AI accelerates—but ownership must remain human.
Information Abundance, Meaning Scarcity
AI has triggered an explosion of content in project management: automated tasks, continuous reports, meeting summaries, and dashboards quickly become overwhelming.
Beyond a certain point:
- Critical information gets lost in noise
- Stakeholders develop selective blindness
- Key risks go unnoticed
In the AI era, project management is no longer about producing more information—it is about filtering and interpreting it.
Conclusion: Not Better Plans, But Better Uncertainty Management
Artificial intelligence has not simplified ERP and CRM projects.
It has made them more complex—but also more valuable. As technical execution accelerates, the true determinants of success—strategic decision-making, ownership, and governance—have become more critical than ever.
AI may complete 60% of the work.
But when the remaining 40%—decision-making, prioritization, negotiation, ownership, and discretionary effort—is missing, timelines still slip and expectations remain unmet.
Successful project management in the AI era is not about centering technology; it is about synthesizing AI speed with disciplined work culture and modern flexibility.
Neither rigid “old-school” planning nor blind faith that “AI will handle it” is sufficient.
Today, what matters in project management is:
- Not fixed schedules
- But fixed and clear decision points
- Not freezing scope
- But governing scope consciously
As someone who has spent years in ERP and CRM projects, I can state this clearly:
AI may accelerate projects—but what keeps them standing is still sound management.
In the AI era, the teams that make the difference are not those who write code the fastest, but those who manage uncertainty best, make timely decisions, and refuse to outsource human responsibility to technology.
Regards.
www.fatihdemirci.net
#ERP # CRM # AI #PM
