Almost Every Company Falls into Three Enterprise IT Archetypes. Surprisingly, each has an AI Complex

As business leaders race to adapt to the AI era, most conversations revolve around what AI can do: automate workflows, improve decision-making, unlock productivity, create entirely new business models, and much more. Yet for many organizations, the bigger question is not what AI can do, but where their enterprise AI development efforts must begin. 

Every enterprise enters the AI era carrying the weight of its existing technology landscape. Some struggle with fragmented systems and scattered information. Others battle shadow workflows that thrive outside official enterprise applications. Still others find themselves trapped within heavily customized platforms that have become increasingly difficult to evolve.

These organizations may operate in different industries: manufacturing, logistics, healthcare, insurance, or banking. But they often exhibit one of three enterprise IT archetypes: Fragmenters, Evaders, and Clingers.

Each archetype is defined by a distinct pain point. More importantly, each requires a different enterprise AI development strategy.

The irony is that while organizations view AI as the destination, for many of these archetypes, AI also becomes the means through which long-standing technology challenges are finally addressed.

Fragmenters: When Information Is Scattered Across the Enterprise

A Fragmenter is defined by disconnected information.

Consider a regional healthcare network that has grown through acquisitions. Patient records reside in multiple applications. Clinical notes exist as PDFs. Billing information sits on another platform. Critical operational knowledge often lives inside email threads or with long-tenured employees. The challenge is not a lack of information but the inability to see the full picture.

What Fragmenters are ultimately seeking is coordination. They want information to flow seamlessly across the organization without embarking on years-long transformation programs.

This is where enterprise AI development can create immediate value. 

1 Intelligent document processing, for example, can extract and standardize information from clinical documents, invoices, reports, and forms, making previously inaccessible information available for analysis.

2 Enterprise knowledge assistants can connect policies, records, manuals, and historical documents into a single conversational experience. Instead of searching through multiple systems, employees can ask questions such as: “What were the recurring causes behind patient discharge delays last quarter?” and receive answers synthesized from information scattered across the organization.

AI does not eliminate fragmentation overnight. Instead, AI development is focused on creating a coordination layer that helps organizations make better use of the information they already possess. While AI can identify inconsistencies, reconcile records, and surface gaps in information, it is most effective when paired with sound data governance practices that improve the quality and reliability of enterprise data over time. 

Evaders: When Systems of Record Become Systems of Reference

Evaders have already invested heavily in enterprise platforms. Their problem is that employees continue to work around them.

Consider a logistics company running a modern Transportation Management System (TMS). Every shipment, carrier assignment, and delivery milestone is recorded within the platform. Yet when a critical shipment is delayed, the TMS alone is rarely enough to determine the best course of action. Because to assess the impact on customer commitments, the planner must gather information from diverse sources: warehouse systems, supplier updates, carrier communications, and internal discussions. Much of this information extraction happens through emails, MS Teams conversations, and spreadsheets maintained specifically to track exceptions and delays. Once the planners make alternative arrangements, they go on to furnish the information in the TMS. 

Thus, the TMS acts merely as the system of record, while the spreadsheet becomes the system of action. This is the typical trait of Evaders. They have OEMs but they continue to operate on shadow workflows dictated by spreadsheets. Over time, this results in not just underutilization of OEMs but costly human errors, poor operational visibility, and ultimately prevents the organization from responding in real-time. 

Enterprise AI efforts for evaders must focus on retaining the flexibility inherent in shadow workflows, while ensuring every action is tracked and trailed in OEMs. 

For example, in the event of delayed shipment,  a cross-system operational copilot can continuously connect information across transportation systems, warehouse platforms, supplier portals, and collaboration tools, giving planners a unified view of a situation without requiring them to manually assemble it. Instead of spending hours gathering updates in spreadsheets, a planner can simply ask:

Will this delay impact customer delivery commitments, and what alternatives do we have?

The copilot can instantly identify affected shipments, available carrier options, inventory implications, and service-level risks. 

But for such a solution to gain adoption, it must be easier to use than the spreadsheets it seeks to replace. Employees often rely on spreadsheets because they trust them—they know how the data is organized, how calculations are performed, and how different scenarios can be tested. Enterprise AI succeeds only when it delivers the same flexibility and confidence while eliminating the manual effort involved in maintaining shadow workflows. 

Clingers: When Customization Becomes a Constraint

Clingers are defined by customization debt.

Consider an insurance provider operating a claims platform that has been customized for over a decade. Every regulatory change, product update, or process improvement requires code modifications, testing cycles, and costly development efforts. The system continues to work, but every change becomes harder than the last.

What Clingers are seeking is agility. They want to adapt to changing business conditions without constantly modifying the overly customized core that keeps the business running.

AI offers a practical path forward. 

An intelligent business rules layer, for example, can externalize evolving policies and decision logic from the core platform. New approval thresholds, claims rules, pricing conditions, or compliance requirements can be managed through configurable and governed policy frameworks rather than hard-coded customizations. While business owners continue to define, review, and approve these policies, AI can help interpret, execute, and operationalize them without requiring modifications to the underlying system. 

AI agents can further orchestrate workflows across systems, coordinating activities and resolving exceptions without requiring deep modifications to the underlying platform.

Over time, organizations can progressively move toward a cleaner core. But AI delivers immediate agility by absorbing change through rules and orchestration layers rather than through additional customization.

Read More: Enterprise AI Solutions Prioritization Roadmap

AI Is Both the Destination and the Means

Fragmenters seek coordination. Evaders seek governed flexibility. Clingers seek agility. Traditionally, solving these challenges required large-scale development programs, extensive modernization efforts, and years of organizational change. AI does not remove the need for transformation. What it does is dramatically reduce the development effort required to overcome the obstacles standing in the way of it.

1. For Fragmenters, AI helps connect information that was previously disconnected. 
2. For Evaders, AI helps eliminate shadow workflows while preserving governance.
3. For Clingers, AI introduces adaptability without increasing customization debt.

In that sense, Enterprise AI development is not merely the destination these organizations are striving toward. It is increasingly becoming the means through which they untangle themselves from the technology challenges that defined them for years.

Which archetype are you?

Gather your business and technology leaders and ask three simple questions:

• Is critical information scattered across systems, documents, and teams?
• Are employees routinely bypassing enterprise systems through spreadsheets and shadow workflows?
• Has customization made it increasingly difficult to adapt systems to changing business needs?

Identify your archetype first. Then determine your AI strategy