Abstract
Artificial intelligence is expected to make strategy easier. Yet many leadership teams are finding the opposite. As analytical capability becomes more powerful and widely available, analysis can support multiple credible futures simultaneously—a condition we call analytical abundance. The result is a paradox: more analysis can make strategy harder. We argue that analytical abundance shifts the primary constraint in strategy from generating insight to creating choice and sustaining commitment. To address this challenge, we introduce a three-cycle architecture that separates strategic intelligence, strategic choice, and strategic change. Competitive advantage depends not only on insight, but also on closure and commitment.
Keywords
Key Recommendations
When More Analysis Keeps Multiple Futures Alive
The CEO of a large European industrial company and her executive committee faced a pivotal decision—whether to divest a capital-intensive business unit that had become increasingly difficult to justify within the company’s portfolio. (The example is a composite drawn from multiple real-world situations.) The strategic rationale appeared straightforward. Divesting could free resources for growth opportunities and improve overall performance. Yet the decision carried significant implications for customers, employees, investors, suppliers, and regional governments.
The challenge was not a lack of analysis. Internal teams, external advisors, and increasingly sophisticated analytical tools generated a continuous stream of evaluations, scenarios, valuations, and operating models. Some analyses supported full divestment. Others favored a partial separation. Still others suggested an operational turnaround. Depending on the assumptions, each option appeared defensible.
As the process continued, the executive committee became divided. The chief financial officer favored divestment to improve capital allocation. Business leaders argued that the unit remained strategically important for key customers. Regional executives emphasized employment and political considerations. The strategy team highlighted long-term uncertainties that could change the attractiveness of each option.
Additional analysis did not resolve these disagreements. Instead, it generated increasingly granular alternatives: staged divestments, joint ventures, partial carve-outs, partnership structures, and various turnaround scenarios. Each option came with its own supporting evidence and business case. What initially appeared to be a decision among three alternatives gradually evolved into a growing portfolio of competing paths.
The executive committee postponed the decision twice in the hope that further analysis would create greater certainty. Instead, new analyses reinforced competing viewpoints and expanded the range of viable options. The challenge was no longer understanding the business. It was choosing a path despite the continued existence of credible alternatives.
When the company eventually announced a partial separation, the debate did not end. Senior leaders, business-unit managers, and regional executives continued to generate and commission new analyses supporting their preferred alternatives. What once required substantial analytical resources could now be done quickly and at low cost. Updated valuations supported a full divestment. New operating models favored retaining larger parts of the business. Revised market scenarios strengthened the case for alternative restructuring paths. As analytical capability became more widely distributed across the organization, competing futures remained alive long after the formal decision had been made. Resources were allocated cautiously, execution slowed, and preparations continued for more than one future.
The issue was not insufficient analysis. It was that analysis kept multiple futures alive, making it difficult both to close the decision and to sustain commitment once a decision had been made.
Across industries, similar situations are becoming more common. As analytical capabilities become more powerful and more widely available, leaders can generate more options and explore more futures than ever before. 1 Yet they often find it harder to choose a direction and harder still to sustain commitment to it.
Integrated Strategy Processes Were Designed for a World Where Analysis Converged
Traditional strategy processes were typically organized as integrated processes that linked analysis, decision-making, and execution in a single sequence. The underlying assumption was that these activities could be tightly connected because analysis would gradually reduce uncertainty and narrow the range of viable options. As understanding improved, organizations could move smoothly from insight to choice and from choice to execution.
Whether through industry analysis, scenario planning, or strategic planning, the logic was similar: additional analysis would improve understanding, eliminate weaker alternatives, and support a clearer strategic choice. As uncertainty diminished, organizations could converge on a direction and align execution behind it.
This convergence logic shaped many established approaches to strategy. Different methods varied in their tools and assumptions, but they shared a common expectation: better analysis would ultimately produce greater clarity. The central challenge was generating sufficient insight to reduce uncertainty and support a well-informed choice. Under these conditions, integrating intelligence, choice, and execution into a single process was both efficient and effective.
Analytical Abundance Changes the Challenge of Strategy
Organizations have moved—unevenly but steadily—toward analytical abundance, a condition in which data is widely available, analytical tools are powerful and accessible, and analytical capability is no longer concentrated in strategy or analytics functions. Business units, ecosystem partners, customers, and other stakeholders can increasingly generate analyses, challenge assumptions, and develop strategic alternatives for themselves. Advances in digital technologies, analytics, and generative AI have expanded not only the amount of analysis available to leaders but also the number of actors capable of producing it.
As analytical capability expands, so does the range of credible futures. Different assumptions, models, and priorities can produce different conclusions about what the future may hold and what strategic path should be pursued. Rather than converging on a single view, analysis increasingly keeps multiple credible futures alive, each linked to distinct strategic options.
The persistence of multiple credible futures changes the nature of uncertainty. In the past, leaders often struggled because they lacked information. Today they increasingly struggle because analysis can support several credible futures at the same time. More analysis does not necessarily narrow the field. It can reinforce competing options, strengthen opposing viewpoints, and generate additional alternatives. The challenge is no longer simply understanding the future. It is deciding which future to commit to.
As uncertainty changes, so does the primary constraint within the strategy process. When information is scarce, the central challenge is generating sufficient insight. When multiple credible futures remain viable, the challenge shifts to creating closure and sustaining commitment. Prior research has shown that expanding alternatives increases decision difficulty and limits convergence. 2 Under analytical abundance, leaders must choose among options that remain analytically defensible and maintain commitment to that choice as new analyses continue to emerge.
In this context, strategy is best understood as choosing and committing under uncertainty. 3 It requires selecting one path among alternatives that are not easily comparable or dominated and maintaining that commitment despite continuing analytical updates. Strategy is therefore not about eliminating uncertainty through analysis but about acting when uncertainty persists.
The implication is straightforward. As analytical capability becomes more powerful and more widely distributed, organizations can explore more options and support more futures than ever before. Yet greater analytical capability does not necessarily produce greater strategic clarity. Leadership therefore becomes less about generating additional analysis and more about creating closure and sustaining commitment as new analyses continue to emerge.
When AI Expands Options Faster Than Leaders Can Choose
The CEO of a semiconductor manufacturer and her executive team faced a narrowing strategic window. (This second example is a composite drawn from multiple real-world situations.) Rising geopolitical tensions, government subsidies, export restrictions, and surging demand for AI infrastructure required a decision about the company’s next generation of manufacturing investments. While many strategic paths were possible, the leadership team initially focused on three broad directions: concentrating investment in the United States, deepening its presence in Asia, or building a geographically diversified footprint across multiple regions.
The challenge was not a lack of analysis. AI systems continuously generated forecasts integrating customer demand, trade restrictions, geopolitical developments, supply-chain risks, technological trajectories, and competitor moves. Every week brought new analyses and recommendations. Depending on the assumptions, each broad direction appeared attractive.
The leadership team became divided. The head of government affairs favored a stronger US presence to benefit from subsidies and political support. The chief operations officer emphasized Asia’s manufacturing ecosystem and execution advantages. The chief financial officer preferred a diversified footprint to reduce geopolitical risk. The chief strategy officer argued for flexibility until future industry standards became clearer. The chief commercial officer favored locating investments closer to major customers.
AI intensified rather than resolved these disagreements. It generated not only new evidence but also new strategic variants. What began as three broad directions gradually expanded into a growing portfolio of increasingly granular alternatives: diversification across allied countries only, phased geographic expansion, customer-specific regional footprints, staged investment strategies, and various combinations of concentration and redundancy. Each option was supported by competing forecasts, detailed business cases, and increasingly sophisticated analyses.
As the strategic window narrowed, leaders commissioned still more analysis in the hope of reaching greater certainty. Instead, the number of credible options continued to grow. New analyses strengthened competing viewpoints while generating additional alternatives. The leadership team found itself trapped in a cycle of evaluation and re-evaluation. The challenge was no longer understanding the future. It was choosing a future before the opportunity disappeared.
The challenge is not doing more strategy work. It is organizing strategy differently. While prior research has examined uncertainty, ambiguity, and decision-making under complexity, 4 the central shift here lies in the expansion of viable alternatives and the resulting relocation of the constraint from insight generation to choice and commitment.
Integrated Strategy Processes Break Down When Analysis No Longer Converges
When analysis supports multiple credible futures and the bottleneck shifts from insight to choice and commitment, traditional strategy processes begin to break down. These integrated processes were designed for convergence. Under analytical abundance, however, analysis evolves in two reinforcing ways. First, it sustains multiple credible futures rather than narrowing them. Second, analytical capability becomes more widely distributed throughout organizations and their business ecosystems. Business units, partners, and other stakeholders can increasingly generate their own analyses, explore alternative futures, and build support for competing strategic options.
The first dynamic makes it difficult to close choices. When different assumptions support different strategic paths, no option clearly dominates. Decisions are deferred in the expectation that additional analysis will resolve the uncertainty. Instead, further analysis often reinforces disagreement, as competing options re-enter the discussion with renewed analytical support. This challenge becomes particularly acute under conditions of high uncertainty, where multiple futures remain credible and difficult to eliminate through further analysis.
The second dynamic makes it more difficult to sustain commitment once a decision has been made. In decisions that depend on multiple interconnected stakeholders, 5 execution requires continued alignment among actors whose actions affect one another. As analytical capability becomes more accessible, these stakeholders can generate their own forecasts, business cases, and strategic alternatives. Rather than converging around a chosen direction, they may continue to support different futures and advocate competing priorities. New analyses therefore do not simply inform execution; they can also reopen debates, delay resource allocation, and weaken commitment to the chosen path.
These dynamics reinforce one another. High uncertainty keeps multiple futures alive, while distributed analytical capability allows stakeholders throughout the organization and ecosystem to continue generating support for competing alternatives. The result is a recurring pattern of decision deferral, option persistence, and weakened execution. The issue is not insufficient analytical capability. It is a mismatch between expanding analytical capability and an integrated process designed for convergence.
A Three-cycle Architecture Separates What Must No Longer Be Integrated
The breakdown of traditional strategy processes is not caused by a lack of analytical capability. It occurs because these integrated processes were designed for a world in which analysis converged on a smaller set of options over time. Under analytical abundance, analysis often does the opposite. It keeps multiple futures alive and enables stakeholders throughout the organization and ecosystem to generate support for competing alternatives.
When this happens, a single integrated process struggles to perform three different tasks simultaneously: exploring possibilities, choosing a direction, and executing that direction. Each task requires a different logic. Exploration benefits from openness. Choice requires closure. Execution depends on commitment and coordination. Combining these activities into a single process allows the tensions of one task to spill into the others.
To address this challenge, we separate strategy into three interdependent cycles—strategic intelligence, strategic choice, and strategic change—each with a distinct nature, as seen in Figure 1. Intelligence expands possibilities, choice authorizes one path, and change stabilizes and realizes that path. Together they form a system that moves from exploration to choice to execution. Each cycle serves a different purpose and therefore requires different leadership behaviors, governance mechanisms, and success criteria.
The three-cycle architecture of strategy.
The first cycle,
The second cycle,
The third cycle,
When AI Gives Every Stakeholder an Alternative Strategy
The CEO of a global automotive manufacturer committed the company to an aggressive transition toward electric vehicles. (This third example is also drawn from multiple real-world situations.) The decision followed years of analysis and reflected the executive team’s belief that long-term competitiveness required accelerating the move away from internal-combustion technologies.
The strategic choice, however, did not end the debate. Business-unit leaders responsible for combustion-engine platforms remained concerned about the pace of the transition. Executives at major suppliers worried about the impact on their existing businesses. Dealer organizations questioned customer readiness. Many stakeholders believed alternative approaches—such as a slower transition, greater emphasis on hybrid technologies, or a more regionally differentiated strategy—would create more value.
Historically, challenging a corporate strategy required significant analytical resources. Under conditions of analytical abundance, that changed. Business-unit leaders could use AI tools to generate market forecasts, customer analyses, investment scenarios, and competitive assessments supporting their preferred strategic directions. Major suppliers and ecosystem partners could do the same through their own strategy teams or consultants. Analytical capability was no longer concentrated at the corporate center. It had become widely distributed throughout the organization and ecosystem.
As a result, stakeholders increasingly produced sophisticated analyses supporting alternative courses of action. A supplier generated scenarios showing prolonged demand for combustion technologies. Dealer groups produced forecasts suggesting slower customer adoption of electric vehicles. Internal business leaders developed analyses supporting greater investment in hybrid platforms. Because these arguments appeared rigorous and data-driven, stakeholders felt increasingly justified in advocating alternative priorities and redirecting resources toward them.
The result was weakening commitment to the chosen strategy. Different stakeholders pursued different priorities while using analysis to justify their actions. Because execution depended on coordination across a broader ecosystem, these deviations reinforced one another. The official strategy remained unchanged, but commitment to it eroded as stakeholders continued to generate evidence supporting alternative paths.
The challenge was no longer choosing a strategy. It was sustaining collective commitment to a chosen strategy when analytical capability had become widely distributed and AI continuously supplied support for competing alternatives.
When and How to Move to a Three-cycle Architecture
A three-cycle architecture is not necessary in every situation. Traditional integrated strategy processes can still work when analysis narrows options, uncertainty is moderate, and execution depends on a limited number of stakeholders. The need for a different approach emerges when three conditions coincide. First, analytical abundance makes it easy for leaders and stakeholders to generate new analyses, options, and alternatives. Second, uncertainty remains high because multiple credible futures continue to coexist. Third, execution depends on many interconnected stakeholders whose actions influence one another. Under these conditions, the traditional expectation that more analysis will produce convergence becomes increasingly difficult to sustain.
Once these conditions arise, leaders should separate intelligence, choice, and change activities. Intelligence should remain open and inclusive, drawing on insights from business units, business partners, and other stakeholders. Choice should be conducted through dedicated decision forums focused on evaluating alternatives, creating closure, and legitimizing a chosen direction. Change should focus on mobilizing stakeholders, allocating resources, and sustaining commitment during execution. The cycles remain connected, but leaders must actively govern the transitions between them. This includes establishing clear decision rights, defining who has authority to close debates, creating formal commitment points, and setting deadlines that prevent continued analysis from delaying decisions or reopening settled choices.
The move to a three-cycle architecture itself is a leadership challenge. This architecture changes how influence is exercised and how accountability is assigned. Some actors may resist because they benefit from repeatedly reopening strategic discussions or because they perceive a loss of influence. This can include strategy and analytics functions that have traditionally owned much of the strategy process. Leaders should not marginalize these groups. Their role becomes more important, not less. Rather than acting primarily as producers of analysis, they become stewards of the intelligence cycle, responsible for expanding exploration, challenging assumptions, and ensuring that alternative futures are properly considered. Leaders should also build support among operating executives and ecosystem partners who benefit from greater clarity, stronger commitment, and more consistent execution. The transition succeeds when the organization recognizes that intelligence, choice, and change remain connected but should no longer be managed as a single integrated process.
Strategic Leadership in an Age of Analytical Abundance
Analytical abundance changes more than the strategy process. It changes the role of strategic leadership. As AI becomes more powerful and analytical capability becomes more widely distributed, organizations can generate more analyses, explore more futures, and develop more alternatives than ever before. Yet the ability to generate insight is no longer the primary constraint. The bottleneck has shifted to choice and commitment.
This shift makes leadership more important, not less. Under analytical scarcity, leaders could often rely on analysis to create convergence. Under analytical abundance, convergence becomes less likely. Multiple futures remain credible, stakeholders can generate analyses supporting competing priorities, and strategic debates can continue long after decisions are made. In this environment, leadership cannot be replaced by more analysis. It must compensate for the limits of analysis.
Figure 2 highlights the leadership challenge. Leaders must encourage exploration without allowing exploration to prevent closure. They must make choices under uncertainty without waiting for certainty. They must sustain commitment without suppressing learning and adaptation. These are not analytical tasks. They are fundamentally human responsibilities.
Leadership roles and responsibilities across the three cycles.
Analytical abundance therefore acts as a stress test for strategic leadership. Success depends increasingly on qualities that AI cannot provide: judgment when evidence remains inconclusive, intuition when alternatives are difficult to compare, courage when choices involve risk, persuasion when stakeholders disagree, and resolve when new analyses tempt organizations to reopen settled decisions.
The challenge for leaders is no longer to generate more insight. It is to transform insight into choice and choice into sustained collective action. The bottleneck has moved from insight to choice and commitment. AI can generate options. Only leaders can create closure and sustain commitment.
Footnotes
Notes
Author Biography
Marc G. Baaij, Associate Professor of Strategic Management, Department of Strategic Management & Entrepreneurship, Rotterdam School of Management (RSM), Erasmus University Rotterdam (