The Enterprise Can't Decide: Why Strategic Decision Debt Is the Real Barrier to AI Transformation

[Fred Champlain]

Here is an emerging paradox worth sitting with: The prevailing narrative around AI in the enterprise focuses on putting humans in the loop — primarily to make decisions. However most organizations lack a formalized approach to structuring or managing the decision-making process itself.

Ask any enterprise leadership team what decisions they need to make and the conversation immediately becomes fuzzy: dependencies, delays, budgets, caveats, politics. In most organizations decisions intentionally flow upwards to executives, where they struggle to make well-informed decisions on a timely and disciplined basis — not because they lack intelligence or intent, but because they are structurally removed from the operational context where the implications of those decisions are best understood. The people closest to the work hold the critical knowledge, while the people holding decision authority are often furthest from it. 

Meanwhile, AI is compounding the problem it was supposed to solve. In an era when AI provides factual support in real time, humans are being overwhelmed by facts — not empowered by them. The volume of data, analysis, and scenario modeling now available to decision-makers is producing decision flooding: more inputs than any governance structure was designed to process. The result is counterintuitive but increasingly observable. AI is actually slowing down human decision-making rather than speeding it up, at precisely the moment when AI-driven markets are accelerating.

This is not a technology problem. It is a portfolio governance problem. And it is creating what I call “strategic decision debt” — the accumulated weight of unmade, unclear, or inconsistent portfolio-level decisions that prevent enterprises from translating AI capability into AI outcomes.

A Taxonomy of Enterprise Liabilities

The enterprise understands technical debt — shortcuts in code that compound over time, increasing fragility and reducing agility. Portfolio leaders have encountered zombie projects — the walking dead of the portfolio, consuming budgets and talent while delivering little strategic value. In 2015, Steve Blank introduced organizational debt — the accumulated structural compromises in roles, policies, and processes that organizations make to "just get it done."

Strategic decision debt is a distinct liability. It does not accumulate in the codebase, the project portfolio, or the org chart. It accumulates at the governance layer — where capital allocation, architecture standards, value stream prioritization, and transformation strategy are decided, or more precisely, where they should be decided but aren't. A single zombie project traps capital in one initiative. Strategic decision debt traps capital across the entire portfolio by preventing the reallocation decisions that would free it.  (See my earlier article: “Untrapping Capital: Build A Value Management System with ServiceNow, SAFe, and Agentic AI”)

McKinsey's State of Organizations 2026 report found that two-thirds of leaders admit their organizations are too complex and inefficient to execute effectively, with 20 to 30 percent of operating expenses lost to structural inefficiency. That complexity is, in significant part, the residue of unresolved strategic decisions — each one reasonable to defer in isolation, unreasonable in aggregate.

Three Forms

Strategic decision debt takes three forms:

Unmade decisions: which AI initiatives to fund, which legacy investments to sunset, which value streams get priority access to scarce talent — these unmade decisions stall everything downstream. Strategy-to-execution is blocked by unmade decisions, especially those that go undetected. 

Unclear decisions: this category of decisions creates the illusion of progress. Leadership announces a direction but never specifies what it means for portfolio prioritization or funding, and business units interpret the ambiguity differently. For example, three competing AI initiatives are launched. None align with enterprise architecture standards for AI that don't yet exist because that decision hasn't been made either.  Cascading decisions and the interdependencies between them create complex decision spaces that quickly become difficult to understand for humans.

Inconsistent decisions: this type of decision is the most corrosive — different parts of the organization make contradictory choices about the same strategic question, producing conflicting priorities and misalignment of funding further contributing to strategic decision debt.  Inconsistent decisions fundamentally bog down the ability of an organization to compete effectively in accelerating markets.

The Systems Dynamic

Through a systems thinking lens, strategic decision debt is self-amplifying. Unmade decisions create ambiguity. Ambiguity increases coordination cost as teams work around the gap. Increased coordination cost consumes leadership bandwidth. Reduced bandwidth leads to more unmade decisions. The loop accelerates.

A second loop runs in parallel. When teams see portfolio decisions consistently deferred or contradicted, they stop waiting for guidance and start making local decisions that optimize for their own context. This is rational behavior at the team level, but it fragments the portfolio at the enterprise level. The fragmentation demands new decisions from leadership — decisions that are now harder because the landscape has diverged.

This is where BANI — Brittle, Anxious, Nonlinear, Incomprehensible, the framework Jamais Cascio created as a successor to VUCA becomes directly relevant. Decision environments in the AI era are not only volatile and uncertain. They are nonlinear — small decisions cascade into disproportionate consequences  (the butterfly effect) and increasingly incomprehensible to human-centric governance structures designed for linear, predictable investment cycles.

The Meta-Decision

The biggest strategic decision debt most enterprises carry is the decision about how they will make decisions in an AI-driven world. Their governance models and authority structures were designed for a slower rate of change. Those structures are now generating strategic decision debt faster than any team can pay it down.

Making this meta-decision means confronting a structural root cause: most enterprise governance is organized around cost centers, functional hierarchies, or project deliveries — not around the flow of value to customers. When governance is misaligned with value delivery, every strategic decision must traverse organizational boundaries that exist for administrative reasons rather than value creation reasons.

The alternative is to structure portfolios around value streams and customer journeys, an essential change when adopting a Product Operating System, but relevant to any organization. When portfolios align to value streams, decisions naturally align to value delivery. Decision latency drops. Governance becomes a value-enabling flow system rather than a checkpoint process. Lean Portfolio Management operationalizes this by organizing lean budgets around value streams, establishing portfolio-level decision cadences, and creating guardrails that enable decentralized decision-making within strategic boundaries.  Enterprises can only respond to change as fast as they can reallocate capital.

But even this structural shift addresses only part of the challenge. The deeper question is whether we are setting the wrong strategy for leveraging AI in decision-making entirely, and how to successfully avoid compounding the growth of strategic decision debt.

What Comes Next

The questions that emerge from strategic decision debt lead outward in several directions: How do we model decisions explicitly so that AI can learn what effective and ineffective decisions look like, and what role do decision graphs and formal decision architectures play in structuring complex decision spaces? How does enterprise architecture modeling need to evolve to enable AI-augmented governance? What is the relationship between strategic decision debt and the zombie projects that refuse to die — and what would a value management system look like that treats decision flow as a first-class metric? What happens when we stop treating decisions as an executive asset and start treating them as a liability to be distributed throughout the organizational topology? And perhaps most provocatively: should we be enabling AI not just to support human decisions, but to make certain categories of complex strategic decisions that humans are structurally incapable of making well at enterprise scale?

These are all facets of my underlying hypothesis of strategic decision debt: that the enterprise decision-making architecture is the critical infrastructure of AI transformation, and most organizations haven't built it yet.

Where to Start

For organizations ready to address strategic decision debt, ServiceNow Strategic Portfolio Management and Enterprise Agile Planning provide the foundation for making decision flows visible, structuring portfolios around value streams, connecting strategic intent to execution, and establishing the continuous feedback loops that surface decision debt before it compounds.

The newly announced ServiceNow Context Engine deepens this further — connecting relationships, policies, and decision history across the enterprise so that AI agents operate with the full context of how the business works, not just the data it generates. Built on ServiceNow's Service Graph, Knowledge Graph, and data inventory, Context Engine captures the why behind decisions, not just the what, compounding intelligence with every human and agent decision made.

ServiceNow Enterprise Architecture extends this into the structural layer — modeling the relationships between capabilities, applications, technologies, and value streams that constitute the enterprise as a system. When enterprise architecture models are connected to live portfolio and workflow data, they become the foundation for a digital twin of the business: not a static blueprint, but a dynamic, queryable representation of how the organization actually operates and where its structural constraints lie.

Together, Enterprise Architecture and Context Engine provide the structural and contextual layers required to move from treating decisions as isolated events toward modeling them as a persistent, interconnected substrate of the enterprise. Combined with ServiceNow AI Agents that detect weak signals of value erosion and decision stagnation continuously, and the disciplined operating cadences of Lean Portfolio Management practices, enterprises can begin building the SPM decision architecture that AI transformation demands.