For this final issue of MIT Sloan Management Review, Benjamin Laker and Maria Papacosta offer advice on ending things well. In that spirit, we want to reflect on the impact our editorially independent publication has had in its 67 years. Over the past few months, we’ve been buoyed by many messages and online comments validating our core mission: to identify and share new ideas that advance management practice. We’ve also heard about the value SMR has created as a curator and connector at a time when so much information of variable quality is available from so many sources. SMR has given scholars and practitioners a platform to develop their thinking on innovative approaches to business challenges, and leaders the opportunity to deeply engage with new insights and put them to work to solve problems and improve results.

Many people have contributed to SMR’s success; here are some thoughts from a few of them.

Paul Michelman

Editor in chief, BCG; former editor in chief, MIT SMR

“What SMR delivered for decades was innovative, well-expressed management thinking that can light a spark. It can provoke somebody to cast their eyes left or right when they’ve only been looking straight. Lighting a spark that leads to a solution is a highly practical thing. It begins with identifying ideas that matter, that can move the needle, that are fully additive to what’s already been published — but then shaping it in a way that delivers actionable inspiration. Another thing that set SMR apart is its place at the collision of technology, innovation, and industry — its MIT-ness. SMR was never afraid to be wonky, and I mean that in the best ways — that it’s OK to get underneath and a little bit nerdy.”

Deb Gallagher

Publisher, MIT SMR

“The most meaningful and rewarding part of the work we did was that people acted on our content — we helped change how people work, how organizations function. There’s a lot of great work being done in management education to make companies run better, but university scholarship can be very esoteric. Our mission was to create a bridge between academic research and organizational need. That remained steady while we evolved how we delivered it, adapting to new media formats and channels. We extended our reach by licensing our content to learning organizations and to publishers overseas that saw our unique value in speaking to the needs of leaders and workers in organizations and wanted to bring rigorous thinking to leaders in their markets.”

Martin Reeves

Founder, Martin K. Reeves Advisory; former chairman, BCG Henderson Institute

“BCG collaborated with SMR for almost 20 years on Big Ideas research projects on sustainability and on AI, in addition to contributing articles. Without a doubt, SMR has made a significant impact on the field of strategy, being one of the few publications bridging the rigor of academic research with practical relevance and digestible insights for practitioners. In particular, one of SMR’s superpowers has been to use its large and loyal subscriber base to both gather fresh field data and build an audience for consuming and discussing new insights. Thank you to all my friends in the editorial and marketing teams for your excellent work and impact on the field.”

Jeff Schwartz

Senior adviser, Gloat; former principal, Deloitte Consulting LLP

“For several decades, my job has been to look toward what’s coming next — and no publication shaped that view more than MIT Sloan Management Review. What we lose with its closing is a rare devotion to praxis: the living intersection of theory and practice, where technology, strategy, and operations meet — the very thing a place like MIT exists to think about. Our collaborations with SMR on digital transformation, talent, and strategy yielded early insights that continue to be relevant today. Later, we pivoted to a multiyear exploration that reframed the workforce itself — opportunity markets, workforce ecosystems. SMR seldom treated an idea as once-and-done. Themes were introduced, explored, and followed through. It seated academics and practitioners side by side, built communities of practice, and quietly orchestrated ambitious ideas.”

Robert W. Holland Jr.

Former managing director and publisher, MIT SMR

“SMR was losing money when I joined in 2010. We recognized that the academic journal model was no longer viable and made a strategic pivot to reposition SMR to bridge academic content and practitioner needs. The Big Ideas program generated significant revenue and extended SMR’s reach. We became self-sustaining and added “MIT” to our name to become MIT Sloan Management Review, which started to bring the ethos of MIT into what we were doing, with a greater focus on innovation and new ideas. That focus was unique to SMR compared with other journals.”

David Kiron

Editorial director, MIT SMR Big Ideas

“Big Ideas was never just a research program. It was a bet: that rigorous, independent research, produced with academics and with thought leaders at the world’s top consulting firms, could influence management practice and pay for itself. The bet paid off. Over 16 years, the program produced more than 40 reports and articles, garnering more than 3 million page views and 7,000 citations. Our thought leadership earned gold medals from the American Society of Business Publication Editors and Axiom. And program director Allison Ryder created Me, Myself, and AI, a respected podcast with nearly 2 million downloads. I am grateful to the researchers, editors, collaborators, and readers who made my role feel more like a privilege than a job.”

Martha Mangelsdorf

President, Mangelsdorf Communications; former editorial director, MIT SMR

“The heart of MIT SMR’s work is communicating rigorous and highly relevant academic research in a way that makes the findings easily accessible to managers. I worked with an author team as they developed a 2017 MIT SMR article about the Oxford approach to scenario planning. That topic has become increasingly important as our world becomes more uncertain, and the article is currently one of the bestselling pieces in MIT SMR’s store. MIT SMR’s continuing impact on practice can also be found in a new report from the MIT Institute for Work and Employment Research about negotiations between Kaiser Permanente and its unions over AI; the labor and management participants read and discussed two MIT SMR articles on AI implementation. That’s SMR at its best: providing practical, research-based information that helps leaders and organizations address critical issues.”

Linda Hill

Wallace Brett Donham Professor of Business Administration, Harvard Business School; author and editorial advisory board member, MIT SMR

“SMR provided space for those of us who do research to share our ideas, even when they were evolving, and get feedback from readers to shape how we would do our research going forward. That link between theory and practice starts a dialogue between academics and practitioners that makes both the theory and the practice richer. SMR helped practitioners and academics to frame the questions we should be asking. There are very few people who know how to be those translators. SMR editors were the bridgers — people who can actually curate and translate and integrate across worlds. We need that so much today. Emerging technologies are transforming everything. SMR helped practitioners understand what the opportunities and challenges really are. It created a global ecosystem of thought leaders in both practice and the academy that helped push both forward.”

We’re proud of the work we did at MIT Sloan Management Review and honored to have convened and participated in a lively, productive, and exciting conversation on leadership, innovation, and how business can contribute to a better world.

Many companies today declare a corporate purpose and subsequently find that translating it into strategy, culture, and daily decisions is difficult. Underlying that challenge is a critical question: Who will ensure that a purpose commitment is operationalized?1

In response, companies seeking to institutionalize purpose are creating a new executive role. Cisco, Deloitte, Virgin Atlantic, and others have appointed chief purpose officers (CPOs), signaling that organizations are increasingly recognizing the need for dedicated leadership of these efforts. (See “What Every CPO Must Do.”) Our study of 56 of these emergent leaders reveals that individuals can take markedly different approaches to the role, depending on how purpose is conceived at their organization and the degree to which it has been operationalized.2

We identified four leadership archetypes that reveal distinct ways in which purpose becomes operational and examined the conditions that enable or constrain their impact. Different types of CPOs face distinct execution barriers that can cause purpose to stall. In this article, we’ll look at each leadership profile and explain how to determine which aligns best with current organizational needs.

Four Types of Purpose Leadership

Of the four distinct types of CPOs we encountered in our sample, the largest group (39%) was the visionaries, who aim to inspire shared meaning and emotional connection. About one-fifth were activists, who confront institutional contradictions and use purpose as a moral compass, while one-quarter were architects, who strive to institutionalize purpose through governance and systems. PR strategists were the smallest group (16%); they prioritize crafting coherent narratives that align brands and communication. We’ll look at each archetype in turn.

Visionary CPOs often emerge in smaller, family-owned, or founder-led organizations. In these settings, values are often deep in the company’s DNA but not explicitly articulated or translated into strategy.3 For these leaders, purpose is an ethical and social obligation that must be lived collectively but not necessarily communicated outward. They inspire others to embrace and embody the organization’s purpose, fostering collective commitment to executing a purpose-driven strategy. They translate the company’s core values into a shared sense of meaning for employees. Their leadership is deeply personal, and they identify closely with the organizational purpose.

Visionary CPOs prioritize helping employees connect to purpose on an emotional level. Their storytelling is unscripted and relational: It surfaces in conversation, in shared rituals, and in how employees describe their own work to one another. They may host reflection circles, where employees share how their work connects to the company’s purpose; run coaching sessions focused on aligning personal and organizational purpose; and hold offsite retreats that give teams space to explore why their work matters and how it contributes to fulfilling the company’s purpose. One Danish fashion company holds weekly meetings in a park to reflect on its social impact. Small rituals like these sustain collective reflection and strengthen what one leader called the “emotional glue” that binds people across the organization.

These types of CPOs tend to cultivate influence by building personal connections and informal networks rather than via formal authority. They participate in company sports, join lunch conversations, and use everyday interactions as opportunities to reinforce shared meaning. Many view such a relational approach as the optimal way to create soft power and trust. By weaving purpose into daily relationships, visionary CPOs allow it to diffuse organically to become part of how people experience belonging and identity at work.

These leaders also seek to empower others to lead, recognizing that making purpose a shared responsibility helps facilitate organizational change. Many create cross-functional teams that host purpose dialogues, pilot new practices, and surface dilemmas for senior leaders to address.

Visionary CPOs may find it challenging to be seen as a strategic peer in the C-suite, given that purpose work often does not yield immediately tangible outcomes. Several study participants noted that their work was initially dismissed as “nice to have” rather than essential to organizational performance. Over time, most visionary CPOs learn that their own credibility depends on demonstrating how purpose advances business results. To achieve that, they must balance the work of translating moral conviction into measurable impact with efforts to maintain the human connection that makes purpose stick among their organization’s internal audience.

Activist CPOs work on the front lines of corporate transformation. They are most often situated in large, complex organizations characterized by conflicting stakeholder demands and sensitive issues. When such organizations are confronted with commercial pressures, political tensions, or inertia, they often struggle to uphold their stated values. The goal of the activist CPO is to challenge the status quo and ignite new energy. They see their role as reminding people what the organization stands for, holding it accountable to its purpose, and confronting drift and ethical blind spots.

Their leadership style is bold and unapologetic as they ask uncomfortable questions to probe whether decisions are prioritizing profit over people and purpose. One CPO urged the board to withdraw from a lucrative but ethically questionable market; another paused a multimillion-dollar product launch because it conflicted with the company’s stated values.

Some activists push for change beyond organizational boundaries, seeing purpose as a vehicle for systemic change. As one said, “I want to help change entire systems. … If this is how capitalism works, then let’s rewire it.” These leaders may run awareness initiatives like climate science briefings and social justice workshops, aiming to focus attention on the organization’s role within broader society.

The greatest challenge for activist CPOs is maintaining legitimacy: They must balance moral conviction with strategic credibility in order to push boundaries but avoid being labeled as troublemakers. Their credibility may also be questioned if some stakeholders see them as showing inadequate commitment to particular issues. Too much advocacy risks alienating leadership, while too much caution may erode trust among employees.

Failures occur when conviction turns into confrontation. Activists who push too hard without building coalitions risk marginalization. Without organizational support, isolation and moral fatigue can set in, turning purpose advocacy into frustration rather than transformation.

Activist CPOs who achieve sustainable cultural transformation build alliances while gradually shifting organizational logic and norms. For example, one CPO described how they consistently asked in meetings how decisions were connected to the organization’s purpose. It took years, but their colleagues eventually began proactively applying purpose as a decision-making filter in investment and strategy discussions — without needing to be prompted. Another noted that consistent questioning, once dismissed as idealism, eventually became part of how strategy meetings were framed. Activist CPOs’ influence is slow but transformative. They move the discussion away from asking whether an action is necessary and toward considering whether it aligns with the organization’s identity and values, and ultimately, its purpose.

Architect CPOs typically operate at the upper levels of large multinationals. In such environments, a formal purpose statement usually exists but has yet to be fully understood and adopted across the organization. Architect CPOs primarily identify as institutional changemakers who build the structural foundations that enable purpose to become relevant, durable, and measurable. Rather than relying on bottom-up engagement, they strive to hardwire purpose into the systems that drive performance. In doing so, they demonstrate that a strategically embedded purpose can endure beyond the tenure of any individual leader.

Architect CPOs focus on alignment between purpose, strategy, structure, and performance systems. For instance, from 2017 to 2019, Lara Bezerra was the CPO at Roche Pharma India, while also leading the division. She linked incentives and performance metrics to patient outcomes and community impact, ensuring that purpose shaped how success was defined and rewarded.

Architect CPOs’ work often involves redesigning policies, restructuring the organization, and building decision frameworks that help maintain alignment between values and commercial priorities. Most such frameworks track both financial indicators and metrics that evaluate sustainability performance or social impact so that they can show evidence that values-based decisions are aligned with the health of the business.

Sustaining momentum can be a challenge for architect CPOs amid bureaucratic inertia, competing priorities, and limited resources. Progress can be invisible and results easily overshadowed by short-term pressures, so they must learn to strike a balance between conviction and pragmatism.

PR strategist CPOs excel at crafting a cohesive narrative around purpose. The role often takes this shape in large, dispersed, brand-driven organizations where the purpose must be communicated consistently across geographies and stakeholders. PR strategists construct a deliberate, repeatable story designed to align how the organization is understood by employees, customers, investors, and the public. They align internal commitments and external reputation around a shared pro-social story that’s intended to engage and influence key stakeholders, foster identification with the organization’s values, and motivate behavior consistent with its overarching purpose.

Most such CPOs come from backgrounds in communications, marketing, or branding and view themselves as the storytellers of purpose — responsible for how it is articulated, shared, and believed both within and outside the company. Their appointment often signals a formal public commitment to purpose.

PR strategist CPOs look for creative ways to make purpose visible and credible. Some amplify employee voice through storytelling campaigns that highlight personal purpose moments and are curated, produced, and shared at scale. Others integrate social commitments into product portfolios or employer branding. These CPOs succeed by making sure the purpose is known and valued in people’s daily work.

However, while PR strategist CPOs understand the power of purpose to inspire stakeholders, they often lack the authority to embed it structurally within their organizations. Their challenge is legitimacy and reach among internal stakeholders. They are expected to build belief but are not always empowered to reshape the systems that sustain it. These leaders risk becoming figureheads who are powerful in communication but peripheral to real change in settings where purpose work is not actually mature and purpose remains confined to messaging rather than embedded in decisions and visible in actions. In the worst cases, storytelling replaces structural action, leading to “purpose-washing.” This misalignment can give rise to moral and identity dilemmas as CPOs try to reconcile their professional role as guardians of corporate purpose with actions that contradict the values they are meant to uphold. Over time, such tensions may precipitate personal and organizational identity crises, eroding both ethical integrity and institutional coherence.

Which CPO Type Fits Your Organization?

In deciding what kind of executive is best suited to lead the purpose initiative within an organization, senior leaders should ask themselves, “Are we trying to renew culture, realign strategy, or rebuild trust?” Our Purpose Leadership Fit model is a simple and practical way for boards and CEOs to identify the type of purpose leader who can help them move forward. We’ve found that organizational types often correspond to a particular stage of maturity, certain purpose objectives, and a particular type of leader. The four scenarios below set out the fit; the challenges each leader will face are summarized in “CPO Archetypes Approaches and Challenges.”

1. At smaller, founder-led, or early-stage organizations beginning their purpose journeys, the job to be done is to build purpose awareness and drive cultural renewal. The focus here must be on forging an emotional connection and engaging employees.

Best fit: Visionary CPO

Success factors: To build credibility and influence beyond inspiration, the visionary CPO should prioritize developing a small set of credible metrics that demonstrate how purpose affects outcomes such as engagement, retention, or customer advocacy. They should invest in informal, trust-based mechanisms (such as rituals, narratives, and peer networks) that anchor purpose in daily behaviors and decisions.

2. Large or complex organizations undergoing change where leadership is open to confronting profit-principle tensions are developing maturity in operationalizing purpose. In these settings, a CPO must drive accountability and challenge established norms to align purpose with action.

Best fit: Activist CPO

Success factors: Activist CPOs must build coalitions that span functions and hierarchies. They need to establish structured yet psychologically safe spaces for employees to voice misalignments. Their most important role is to secure visible sponsorship from senior leaders and governance bodies to sustain their mandate.

3. Multinational corporations with more maturity in their practice are typically integrating purpose into strategy and governance. Here, the CPO needs to more deeply embed purpose into systems and measures of outcomes, such as performance metrics.

Best fit: Architect CPO

Success factors: To ensure that their frameworks take root, architect CPOs must emphasize simplicity, clarity, and relevance at every level of the organization. They should design dual-KPI systems and governance structures that institutionalize purpose without creating unnecessary bureaucracy and link incentives to stakeholder value, not just financial returns.

4. Large, brand-driven organizations that have a mature, authentic, and well operationalized purpose practice must still build credibility and stakeholder trust as part of their competitive positioning. CPOs have to skillfully communicate purpose to build trust and keep internal culture aligned with external reputation.

Best fit: PR strategist CPO

Success factors: Communication must align with substantive progress. PR strategist CPOs should collaborate with HR, operations, and strategy to ensure that the stories they share are grounded in action and include concrete outcomes to make abstract values tangible.

No single leadership type fits every situation. The choice depends on how mature the organization’s purpose already is and whether the immediate work is to shift culture, confront contradictions, build systems, or sustain credibility.

Most organizations define purpose statements but falter when it comes to embedding purpose in strategy and execution. Our research suggests that this is a leadership gap. The four archetypes — visionary, activist, architect, and PR strategist — offer different approaches to the challenge of how to make a purpose commitment durable inside an institution.

As organizations move to appoint a CPO, they must ask: Which kind of leader is needed now? What work must they do? When will the work change? What support will enable them to succeed? The leaders who get this right will understand that their job is not to embody purpose permanently but to institutionalize it so thoroughly that their own archetype, in time, becomes unnecessary.

What’s wrong with loyalty? Isn’t it a virtue to display steadfast allegiance to something or someone other than oneself? After all, loyalty has been celebrated as a virtue across many cultures for millennia.

As economic headwinds and a tough job market appear to be empowering more authoritarian styles of leadership, it’s a good time to look more closely at loyalty. We must ask whether good leaders can fairly demand it of their followers and whether it’s an idea that modern management should finally discard.

Two recent high-profile examples illustrate just how problematic an emphasis on loyalty can be.

Staffers loyal to former U.S. President Joe Biden allegedly concealed troubling signs of the aging leader’s cognitive decline before his poor performance in a June 2024 televised debate against then-candidate Donald Trump. The public reaction to the event led the incumbent to withdraw from the presidential race, long after it was feasible for his party to engage in an open process for nominating another candidate.1

That election was won by a man who has made loyalty a central organizing principle of his leadership, at times demanding personal allegiance from those in his administration and reportedly subjecting candidates for senior roles to “loyalty tests.”2 Any sign of disagreement with President Trump is recast as betrayal, and those perceived as insufficiently loyal are sidelined or removed, even when their job performance or policy alignment is not the primary issue. Conversely, notably problematic performance and actions themselves may go unchallenged, as long as unwavering allegiance is maintained.

Loyalty, enacted in this manner, is not about shared commitment to a broader purpose or principle rather than to an individual. In practice, loyalty is often a measure of personal fealty, with silence and compliance positive indicators of commitment. When organizational scandals are exposed, it’s not uncommon to find that coverups of abuse, financial irregularities, and other forms of corporate malfeasance were rationalized as acts of loyalty to the institution or allegiance to a leader.

The Problem With Loyalty

Many readers likely have personally experienced these less admirable kinds of loyalty. Whether it’s “looking the other way,” “omitting all the facts,” or a host of other forms of ignoring or distorting the (ethical) reality of a situation, people commonly use the noble language of loyalty rather than the condemning language of fear, greed, and other types of self-preservation or self-interest. Broader history, likewise, shows that the word loyalty is commonly used in descriptions of the most corrupt regimes and destructive ideologies — where people have committed themselves to, colluded with, or enabled some of the worst actors in history out of fear, nationalism, or blind allegiance.

Loyalty is inherently about partiality; that is, it prioritizes some purpose, person, or group’s interests above others.3 In the context of work, loyalty means promoting the well-being of or preventing harm to a particular group of people or an organizational objective, even if it involves disregard for or harm to other people or purposes. Consider the manager who proudly recruits and promotes people from their college alma mater without considering how this affects potentially stronger candidates from elsewhere, or a subordinate who is fiercely committed to their manager because it benefits them personally, not because they’re both putting the organization’s mission first.

Loyalty is thus a conditional virtue: It might serve the pursuit of universal principles like justice or compassion, or it might serve much less noble things. Actions deemed loyal can, in short, be moral, amoral, or downright immoral. Covering up an in-group member’s unethical behavior because it also benefits you to do so might be described as loyal, but it’s still unethical. Similarly, it’s hard to be impressed by the morality of salespeople’s loyalty to their company when they use sales tactics to maximize their own income at the expense of customers. So does the term really reflect something to be proud of, or is it a nice-sounding way to deflect attention away from the harm being done or the self-interest of the actor? It’s certainly easier, after all, to say we’re “being loyal to the company” than to say that we’re “screwing customers,” or to say we’re “loyal to the boss” rather than that we’re “complicit in covering up abuse.”

Our goal, thus, is not to try to redeem the inherently problematic concept of loyalty. In work contexts, its usage has become so skewed toward the unadmirable that it’s not worth trying to salvage. Furthermore, decades of rightsizing and reorgs have made it abundantly clear that workplace loyalty is too often a one-way street. Research discussed in Sloan Management Review nearly 30 years ago pointed to the breakdown of a social contract under which employees could expect that their loyalty to the company would be reciprocated.4 In a recent survey, fewer than one-quarter of U.S. employees reported strong trust in their leadership or the belief that their organization cares about their well-being.5 The cynicism (or realism) of the majority was validated quite explicitly in August 2025, when AT&T CEO John Stankey fired off a frank memo in response to an employment engagement survey.6 In the memo, he addressed those who may have expected an “ ‘employment deal’ rooted in loyalty,” writing that the company had “consciously shifted away from some of these elements.”

Postures such as Stankey’s are strengthened by a difficult job market in which employees have few alternatives and authoritarian leadership styles are on the rise. The shift toward transactional work exchanges grounded in fear is well underway. Executives may label employees who challenge them in any way, or who try to maintain work-life boundaries, as “disloyal,” diminishing their career prospects. When loyalty means “having my back no matter what,” “sticking to the party line,” “not contradicting me in public,” and “making me look good,” workers may comply, but with an inevitable cynicism that looks nothing like the healthy employee engagement that organizations need in order to thrive. Similarly, when workers recognize that they are in a culture where it’s not what you know or contribute but who you know and what you do for them that gets you ahead, they may default to fawning and sycophantic behavior. Under those circumstances, it’s hard to imagine the organization itself is optimizing performance.

You may be thinking that the problems we’ve described are merely distortions of loyalty, not a problem with the notion of loyalty itself. You can probably readily think of workplace situations where the label of loyalty is applied to what seems like admirable behavior; situations where people remain committed to a mission and set of core values amid external challenges that would make drift toward personal comfort, safety, or short-term profit maximization easy to rationalize. That’s why we often associate the word loyalty with extraordinary effort, sacrifice, and resilience and claim it as a force behind necessary revolutions, solidified alliances, and the holding together of movements and nations. But all of the true virtues we attach to the term loyalty can stand on their own: One can be a person who demonstrates commitment and integrity, and keeps their word, without necessarily being or needing to be called loyal.

The good news is that we don’t need to use the word loyalty at all in the context of organizational life. Decades of work in the social sciences have given us a better understanding of what qualifies as healthy social exchange at work: where an ongoing interdependence is rooted in some degree of socioemotional support, reliability, and commitment to both contractual and more diffuse goals and obligations. Management research, meanwhile, has told us a lot about what happens when workplace relationships between leaders and employees are healthy: better performance, more timely truth-telling, more adaptability, less burnout, more engagement, and a sense of belonging. Given all this, and our observations of the mostly unhealthy ways that loyalty manifests at work, we suggest that when leaders explicitly demand loyalty, healthy social exchange is not on the agenda and people should be on the lookout for dysfunction. Loyalty tests or demands aren’t necessary or healthy in friendships or family relationships, and they aren’t in work relationships, either.

Not All Social Exchanges Are Created Equal

Once we stop valorizing loyalty, we can take a more clear-eyed look at the social exchanges that shape our allegiances at work. Many work relationships are characterized by what we call imbalanced allegiances. They can be coercive, where one side has the power to demand adherence to their agenda. In cultures shaped by coerced allegiance, “do it or else” is the spoken or unspoken mandate. In a survey we conducted, people in business units dominated by coerced allegiance were statistically more likely to respond that “agreeing with powerful others is the best alternative” and that it’s “safer to agree with managers than to say what you really think.”

This has a chilling effect on employee voice: People stay quiet and comply due to the feared consequences of being labeled disloyal. Leaders relying on coerced allegiance often mistake compliance with agreement and authentic commitment, but without real trust, innovation and agility silently die. Unsurprisingly, adaptive performance — measured as the ability to deal effectively with unpredictable situations and to adjust quickly when changes occur — was lowest among people working under coerced allegiance.

One-sided allegiance needn’t be coerced, however: Some cultures breed sycophantic allegiance. People desperate to be accepted, to be “in the room,” or just to survive when their own objective performance wouldn’t be enough will often willingly do whatever they think their boss wants. When a conflict arises between the organization’s mission or other stakeholders’ needs and their sponsor’s agenda, the sycophant’s loyalty is clearly to the latter.

Sycophantic allegiance and coerced allegiance often go hand in hand. Once people see the effects of coerced allegiance — disagreement is quashed, and those who speak up get pushed to the fringe — some employees start to fawn. That’s why survey respondents’ reported confidence in their ability to speak up effectively in relationships marred by imbalanced allegiance was by far the lowest across the types of social exchange we identified. Sycophantic allegiance likewise causes drift from the organization’s core purpose and leads to the well-known ills of a culture in which it’s not what you know or contribute that gets you ahead but who you know and what you do for them. But since sycophantic loyalty is not reciprocated, the sycophants, too, will be thrown under the bus when it becomes necessary.

In social exchanges based in mutual personal advantage, both parties value trust, and they care for and protect each other. These relationships can create cohesion, psychological safety, and a sense of belonging. But when one respondent, a consultant in the public sector, said, “I am not alone in saying I am there for the people … not the client or the work,” it pointed to the problem with these kinds of relationships: The ultimate focus is the relationship itself, not a broader shared purpose. These types of exchanges do fine in steady-state operating environments, but when turbulence hits, self-interest usually replaces any sense of mutuality.

A relationship-first mentality might be appropriate in other spheres of life, but few formal organizations have “advancing the interests of our employees above all else” as their primary mission. The mission of UVA Health, part of the university where one of us (Jim) works, is “transforming health and inspiring hope for all Virginians and beyond,” not “enriching our doctors” or “making work as easy as possible for all employees.” And when the organization needs to evolve to adapt to unforeseen headwinds or an employee’s poor fit, this form of commitment becomes even more problematic. Taken too far, “we protect each other” and “we don’t rock the boat if someone might fall off” ultimately undermine the organization’s fundamental reason for being and can allow unethical behavior to flourish.

However, even a culture that encourages a laserlike focus on the mission, what we call idealized devotion, can lead to dysfunction when it becomes an expectation enforced by others or driven from within. In moderation, devotion to a cause can energize people and build camaraderie toward the pursuit of noble objectives. But when it’s understood that “we do whatever it takes” and “the cause always comes first,” self-sacrifice is valorized and leaves burned-out and disillusioned workers as collateral damage. Our survey respondents in units of this type reported feeling a significantly weaker sense of belonging and less emotional attachment to their units compared with people in other units. They were also more likely to report that some people are treated as impersonal objects and that people have become more callous toward each other over time.

The healthiest type of social exchanges we identified reflected what we call shared principled commitment. Here, people are committed to the organization’s mission and will prioritize it when hard choices are necessary, but not without two-way regard for the humans involved. When organizations have a high degree of shared principled commitment, they’re more likely to benefit from people telling hard truths and holding one another and their leaders accountable to the mission and its underlying shared values.

What’s the benefit of this approach? Survey respondents who reported shared principled commitment as the most common form of social exchange in their unit also reported significantly higher overall performance than respondents in units dominated by all other types of exchange. That superior performance included the quantity, quality, and efficiency of work done. It also included the generation of innovative solutions to improve the unit or organization, which is what likely makes shared principled commitment more durable than other types of social exchanges. When turbulence hits, when values are tested, and when decisions are unpopular, shared purpose and mutual respect combined help people persevere and remain committed.

Shared principled commitment isn’t reserved for organizations with a lofty mission, like those in health care or environmental conservation. It can be found in any organization that has created a compelling reason for people to come and stay together in the service of something no one could achieve individually. At WD-40, former CEO Garry Ridge established a set of company values: The first was, “we value doing the right thing,” and the second was, “we value creating positive, lasting memories in all of our relationships.” That ethos went far beyond selling lubricant. It was about how employees treated one another, customers, and partners every day. Ridge often said that the real product wasn’t the can on the shelf but rather the trust, connection, and sense of belonging people experienced through their work. By elevating “making memories” as a shared human purpose, he gave employees a reason to stick together that was bigger than profits or personal gain.

Taking Stock of Your Organization

What types of relationships define your unit or organization? Presumably you’d prefer not to be seen as coercing allegiance or rewarding sycophants. But how can you know?

Your first step needs to be an honest assessment of what the people around you see as the primary currency of social exchanges. Below are some hard but important questions to ask. And, given that the answers may well be unflattering and thus risky for those whose truth you most need to hear, you need to be sure to ask everyone for their input and do so in a way that guarantees they will suffer no consequences for their honesty throughout the process.

• What kind of behavior is most likely to get you ahead around here? Unwavering commitment to specific people, sycophancy, or silence? Clear commitment to the mission and the truth, even if it upsets people sometimes?
• Does our organization have “loyalty tests”? Loyalty to whom or what?
• What’s the implicit definition of disloyal held by those in power? What happens to those deemed disloyal?
• Do people with power show as much concern for and commitment to the individuals they lead as they expect from them?
• Are most people willing to make sacrifices for the greater good, or are they ultimately transacting in a self-interested, self-promotional way?

You’ll also want to ask people whether the pattern of social exchanges seems to be changing or has changed in ways that are undesirable. Such shifts can happen in reaction to perceived signals about what behaviors are rewarded or punished. You might learn, for example, that what was once primarily principled commitment has drifted into less-healthy idealized devotion. Or what began as strong mutual trust may have shifted toward imbalanced allegiance because leaders have centralized authority and now routinely punish dissent.

We suspect that in most organizations, the honest answers to these questions will reveal gaps between what actually seems virtuous and suitable in organizational life and what’s playing out around you.

Building Culture Based on Shared Commitment

It’s easy to say that shared principled commitment is the most desirable kind of relationship we’ve described. But doing the work to get there and maintain it is hard; it’s work that will likely demand personal sacrifice and new behaviors as you commit to putting mutual regard in pursuit of the organization’s purpose over personal agendas. If you’re serious about earning the upsides of shared principled commitment — discretionary effort, trust, high morale, sustained commitment — here are some ways to begin.

1. Live the organization’s values in everything you do. If purpose is a driver, it must be more than a slogan. When people see leaders aligning actions with mission-based values, especially when those actions cost them something, it signals true commitment to a shared purpose, not just to personalities or performance outcomes that exploit the organization for individual benefit. This consistent modeling reinforces the mutual trust that earns long-term commitment of the right kind.

Where your work on organizational purpose has been cursory or grown stale, identify (or clarify) three to five core values that should underlie purpose-driven decision-making. Then audit recent decisions (such as ones related to hiring, budget, or conflict resolution) and consider whether they reflect those values. A leader who names fairness and merit as core values and follows through by ensuring consistent criteria and processes for determining promotions across the organization is one example.

Earn buy-in — don’t simply demand it — by communicating not only what has been decided but why. Tie the rationale for every significant decision explicitly back to values and purpose. And monitor yourself: Create mechanisms for others to hold you accountable when your actions drift from declared values.

2. Welcome challenge and dissent. People who speak up to share different ideas or views are invested, not disloyal. Creating space for dissent signals that the mission is more important than avoiding conflict. When challenge is welcomed, people feel safer acting in service of what really matters, even when it’s uncomfortable.

Explicitly frame dissent as taking responsibility: “This organization’s purpose is ours; everyone must speak up to pursue and defend it.” Then publicly thank individuals who raise tough issues, even when those issues have yet to be worked through. If, for example, a direct report raises concerns about a flawed product strategy championed by a powerful team, thank them and make space to reevaluate the plan.

During meetings, ask, “What’s a viewpoint we haven’t considered yet?” and “Why might this thinking be wrong?” Look for the people most likely to challenge you when your decisions seem at odds with the organization’s core values or mission, and appoint them to your team or put them in other key roles.

3. Remove the sycophants. People who do whatever you say, or even try to anticipate what you want and do it without question, aren’t invaluable — they’re dangerous. They enable and amplify whatever blind spots and biases you have rather than helping you do better. If you feel completely comfortable going into every meeting with your team, you’re almost certainly lacking the “team of rivals” President Abraham Lincoln successfully surrounded himself with to pressure-test his thinking and decision-making on behalf of his core purpose of keeping the United States of America unified.

Track (dis)agreement and pushback expressed by each member of your team. Remove those who never dare to differ, take their cues from you for every decision, or don’t ask hard questions.

4. Demonstrate relational reciprocity. Healthy, sustainable relationships require ongoing demonstrations of fairness, advocacy, and care. They require that you stand up for people even when it costs you something to do so — when you’re reciprocating based on their commitment and contributions to the organization’s values and purpose, not to reward their fealty.

Advocate for people in the rooms they are and aren’t in (for example, in talent reviews or budget debates). Fight to protect people who are serving the mission when decisions to enrich or empower others are being considered. That might mean, for example, speaking out against proposed layoffs of talented, committed people that may undermine the organization’s long-term well-being.

Nurture relationships by, for example, normalizing checking in on people’s well-being at work and in life outside of work — not as surveillance but as an act of caring. Where people are falling short, develop your capacity to tell them hard truths in ways that show respect and care.

5. Put limits on your leadership. It’s hard for employees to fully separate an organization’s stated mission and values from the behavior of people at the top. It’s why people do very good or very bad things for charismatic leaders; it’s why people quit organizations with terrible leaders despite loving the mission. So it’s not enough to say that “the mission is bigger than me” and that “your commitment should be to the mission, not me.” You have to show it by routinely illustrating the distinction.

One way to do this is to regularly ask yourself and your truth-tellers, “Am I the best person to lead this decision, or should I step back?” You can also share credit widely and name contributions you couldn’t have made alone or don’t deserve credit for.

And signal that the mission matters more than your personal control of it by involving others in shaping direction, not just executing it. A founder might step back from day-to-day decision-making and instead trust a new team to lead an important growth phase, for example.

Leaders often point proudly to the “loyalty” they command when, in reality, the term reflects relationships based in self-serving, short-sighted motives. They conflate obedience, indebtedness, or favoritism with commitment to the organization’s purpose and values. And they fail to recognize that mission-first, care-based relationships aren’t something that can be demanded or assumed. They must be earned, reciprocated, and grounded in consistently aligned behavior. It’s hard, sometimes painful work — work that is likely clouded and undermined by relying on the word loyalty where it isn’t helpful or needed.

We’re not suggesting that the effort it takes to create shared principled commitment is worth it just because it feels more virtuous than social exchanges based on problematic forms of allegiance. As our data and others’ suggest, doing so may also make your organization more resilient in the long run. When people stay merely because they feel stuck, or when they don’t believe that their leader cares about them or demonstrates reciprocal concern, exploitation increases and performance and commitment decline. Conversely, when they feel strong emotional attachment to and identification with their organization and have trust in their leader, performance and prosocial behaviors increase, and absences and turnover decline.7

If that’s not enough to interest you in shared principled commitment, we suggest at least being honest about the deal you’re offering. Using “loyalty” as a euphemism for ignoble ways of transacting fools only the fools, and the rest will just trust you even less. And using the complicated, oft-abused term to describe healthier forms of relating isn’t necessary or helpful either. It’s time to lay off loyalty and employ something better.

“I feel set up,” said the CEO of a professional services firm, breaking the long silence after his presentation. The transformation he had envisioned — and his executive team had championed and worked on for three years — had failed to deliver returns, and the hope that his leadership had once inspired had soured into blame. We witnessed that scene toward the end of a four-year study of the leadership team that strategized and implemented that firmwide transformation. It was the anticlimax of a drama that we have heard many executives recount.

The script of that drama goes like this: The future is uncertain. Trends in the world, the market, or the industry call for radical change. Executives craft a vision of transformation centered on adopting new technology, new structures, or new ways of working. Plans and people are drafted; projects are put in motion. Those initiatives generate enthusiasm. Over time, however, tension and exhaustion ensue. Then resentment emerges, leading to disengagement and, in the worst cases — as in the one we studied — to the organization’s demise.

Analyses of such cases often highlight that transformations fail because people lack the awareness that change is needed or the motivation to pursue it. They may point to the absence of strong leadership, sound strategy, committed employees, or the right process. Our work, in this study and beyond, has made us wary of such explanations. Transformations fail even when awareness and motivation are as abundant as leadership, strategy, and commitment. An urgent issue, or a crisis, provides a burning platform that leaders use as a bonfire to gather people around. But the fire starts to spread, and eventually people flee or burn out.

Our research uncovered what fuels that fire. It is anxiety, an emotion that can be mobilizing at first but becomes paralyzing if it is neglected and uncontained by leaders. As we saw in the case we studied, it may then be organized in ways that keep everyone busy while the status quo endures.

Change may be the only constant, but anxiety is only human. When leaders lose the capacity to name, share, and manage anxiety, their efforts to organize change, and the work of people in the organization, can get co-opted to keep anxiety at bay. This covert function sabotages the overt aim of transformation. We call that process defensive organizing. Below, we describe its unfolding and outline how leaders can avoid it by learning to work with anxiety. We begin by discussing the roots of defensive organizing because people must recognize how easy it is, as leaders and followers, to be taken down a defensive route. Only then can they stop, get to work, and set a different path.

Fear of the Future

Two conditions lay the foundation for defensive organizing. First, one or more executives identify with their organization. Second, the workforce trusts those executives’ competence and benevolence. Those conditions make leaders feel responsible and followers feel reliant.

Leaders who are responsible and trusted are vigilant and agentic, attentive to turbulence in their environments and eager to steer their organizations through it. To do that well, they must make sense of what is happening around them; understand their own reactions, fears, and impulses; and act according to their sense of what must be kept running and what needs to change. However, that capacity often eludes them just when they need it most: When uncertainty rises, they care, and anxiety follows.

It can happen to leaders, we observed, when their organization’s performance flounders. When leaders care, those dips challenge more than their reputation. They threaten their sense of self. Performance anxiety morphs into existential anxiety (a concern that one might not be the leader that one aspires to be) and social anxiety (a concern that one might not be the leader others believe and expect them to be).

In the professional services firm mentioned at the beginning of this article — we’ll use the pseudonym “Recco” — we observed that as its growth slowed, executives acknowledged those worries privately but would not share them with one another. As one confided, “I don’t want to look like a chump because everyone can deliver … and I can’t.” We have heard similar concerns in our executive development work. People in leadership roles often feel uncomfortable or, more precisely, embarrassed to admit anxiety — even to themselves.

Sometimes the reason is personal: A work predicament may trigger anxiety tied to particular sensitivities. More often, the reason is cultural. In many organizations, admitting anxiety is disqualifying for leaders; a common reason executives seek coaching, we have observed, is to discuss their anxieties in private. Such cultures estrange anxiety from leadership, turning a normal, common, and often useful human experience into a problem.

Leaders with a modicum of humanity are bound to worry on occasion that they might fall short or let others down. If simply experiencing that worry feels like a failing, leaders may lose the capacity to modulate and make sense of anxiety, and it can become overwhelming. If leaders believe that their job is to ooze confidence and align people behind a vision of change, that becomes their coping strategy. Anxiety then shapes the vision and the choices that flow from it, and defensive organizing is on its way.

Tracing the Path of Defensive Organizing

Defensive organizing follows a route that begins with senior leaders and spreads across the organization as more people are drafted into the effort to keep anxiety at bay. Our research allowed us to draw a road map of sorts that will help you recognize the signs that you are being led down that route and find the exits at different points.

The comforting vision. Defending against anxiety often takes the form of adopting a reassuring narrative. The first step of defensive organizing therefore requires leaders to convince themselves that performance is ailing for reasons unrelated to them and that they can do something to reverse course. In our study, Recco’s CEO concluded that the firm’s hub-and-spoke structure was too rigid for a fast-changing market. At a strategy retreat, he proposed breaking down silos, operating as a shared leadership team, and championing collaboration across the organization. We watched the word “collaboration” generate a ripple of excitement that seemed to make the palpable worry about the firm’s performance vanish.

The unquestioning enthusiasm of a group coalescing around an idea that affirms their leadership is the first sign of defensive organizing. It is a sign of idealization, a defense mechanism that protects us from the conscious experience of anxiety by crediting something — a principle, a process, a product — with having the potential to restore a bright future. In companies we have worked with, we have seen ideas like digital transformation or customer centricity serve as that rallying point. These mantras then spread through the organization, mentioned in every town hall, strategy document, and performance review.

Not every bold idea is a form of idealization. The sign that idealization might be taking hold is that people avoid asking for evidence that the idea will work. At Recco’s strategic retreat, no one challenged the CEO’s diagnosis or asked what sharing leadership would mean in practice. This lack of inquiry reveals the real purpose of many a strategic vision: to give leaders a story about the future that makes the present feel manageable.

At Recco, the idea of collaboration was plausible and the leader reasonable. The company’s rigid structure indeed limited its agility, and the CEO had a record of being a thoughtful and responsible boss, which made it harder to question his thinking. That silent participation is a form of unconscious collusion with the emerging defense. It is a scenario we have witnessed many times. The future is uncertain, a leader feels pressured to come up with a vision, and their team becomes relieved when they do. It is a plot twist everybody expects — the leader showing that “we can do something,” whether it is introducing a new focus on customer centricity, reorganizing to flatten a hierarchy, or making a large investment in artificial intelligence.

Idealization short-circuits sensemaking by locating the organization’s problems not in the competence of its leaders but in the structures that confine them. A fervor for planning without real action is another symptom. For months, Recco’s leadership team met to strategize collaboration initiatives while making very few changes to the way they ran their business.

The convenient foe. In the second stage of defensive organizing, leaders must find a reason for their vision’s failure to transform the organization’s fate — a reason that is, again, beyond them. That reason is usually found in other people’s lack of competence and care. Others, usually with less power, are drawn into the defensive drama to become targets of blame.

Growing uneasy about the lack of substantive progress, the Recco leadership team identified a group of “future leaders” among their direct reports to form project teams. On the surface, it looked like a textbook change process, thorough and participative. The teams were assigned transformation initiatives and tasked with presenting analyses and recommendations. But the more they took up their work, the easier it seemed for the leadership team to critique them.

The reason was that the data kept pointing back to the leaders’ work. While they championed collaboration, each executive held on to control. When one project team uncovered fierce competition between divisions for new recruits — with vice presidents cherry-picking the best candidates before others could see them — the leadership team attacked the messenger, with one even calling for a senior manager to be fired on the spot. The mirror the project teams were asked to hold up to the leaders presented an unflattering image, and the leaders declared it faulty.

This rise of intergroup tension is the second stage of defensive organizing. The mechanism underpinning it is projective identification, a defense in which one group, usually more powerful, unconsciously attributes to another group qualities they wish to disown in themselves. In Recco’s case, leaders anxious about their competence came to see others as incompetent and then acted in ways that undercut those others’ competence by putting obstacles in their paths, stoking frustration, or setting them up to fail.

We have seen this pattern repeatedly, and the result is resentment for the wasted effort and lack of appreciation. In this case, the leaders’ pushback was overt and intense, and the project teams soon reciprocated their acrimony. “I’m not prepared to put my head on the block for this,” said one team leader. “We are running around doing all this work and getting pushback. It’s not acceptable.” As the pattern continued, the teams stopped caring.

The empty ritual. As intergroup friction keeps everyone from learning, people get tired of conflict and tune out. They let the embers of embattlement smolder under the ashes of disillusionment and go through the motions of initiatives they barely believe in.

At Recco, the recognition that momentum was stalling, and the ongoing unconscious effort to deny having caused it, led the firm’s leaders to launch an internal communication campaign. All managers were asked to gather their teams for regular Monday meetings meant to reengage employees by reaffirming the strategy of collaboration and soliciting ideas to speed up its implementation. In another company going through a transformation lull, teams were cajoled to document small wins in short videos shared internally. Both initiatives had the appearance of change management best practice. In informal spaces, however, people were cynical and complained about the extra work.

This tension between public commitment and private disillusionment is often described as “change fatigue,” but it would be more accurate to call it “change initiative fatigue,” since what drains people’s energy is not actual change but dissonance. The rift between what people say in the engagement meetings and record in the videos, and what they feel and say in the corridor, is costly: for individuals, for their relationships, and for their organizations.

The mechanism underlying this stage of defensive organizing is the production of a secondary problem, where the very principle, process, or practice once idealized comes to be regarded as an obstacle. At Recco, this was collaboration, which soon began to draw blame for the company’s inertia. Leaders complained that involving more people slowed them down. Senior managers resented the extra demands it placed on them. Worrying about the downsides of collaboration became a substitute for confronting what truly ailed the firm.

This is the moment when the purpose of defensive organizing is revealed. It does not produce any substantive change. It produces a shared problem of the organization’s own making. Anxiety stays covered, positions stay intact, and everyone feels justified in looking inward.

The reckoning. Failure to make substantive changes — and the depletion of both the energy and the meaning that keeping the status quo entails — eventually endangers the organization, eroding its ability to deliver. When that occurs, we found, people may resort to a defense of last resort: scapegoating. The leader whose vision had ignited hope becomes the lightning rod for everyone’s blame.

At Recco, as performance continued to falter and disillusioned employees put little effort into the transformation projects, pressure mounted among executives. The CEO set up a change oversight group to prioritize initiatives. It made no headway, however, and the leadership team gave the work back to the CEO, asking him to prioritize the initiatives alone. At their next meeting, he presented a spreadsheet scoring 16 initiatives against seven criteria. The team stared in bewilderment. Thirteen were identified as high priority. “So we are prioritizing everything,” one vice president said flatly. “How does that work?”

A long silence followed. The CEO broke it with emotional honesty, at last: “I feel set up.” He had been, and he had been an active if unconscious participant in that setup. The effort that had begun three years earlier with a room full of hope ended with one man holding a spreadsheet that his team openly mocked. Shortly after that, the firm was acquired and the leaders lost their jobs. The tragedy they quietly feared eventually came to pass.

The Recco story is not unique. Nokia’s mobile phone business, for example, followed a similar trajectory between 2005 and 2010. As Timo Vuori and Quy Huy have documented, Nokia’s senior leaders became caught up by worries about competitors’ smartphones while middle managers worried about the leaders’ opinions. Those unacknowledged worries led the executives to abandon a promising internal platform and bet on the Windows Phone operating system, with little pushback from their skeptical direct reports. Windows Phone never took off, and Nokia’s mobile phones are mostly a fond memory for those who owned one before the iPhone launched.

Defensive organizing distorts leaders’ sensemaking — that is, the effort to interpret what is happening and act accordingly. Sound sensemaking requires paying attention to strong signals and weak ones. Strong signals are visible and urgent: the performance gap, the warring teams, the stalling initiative. Weak signals are subtler and easily ignored: the worries that no one shares, the barely disguised cynicism, the feelings of going through meaningless motions, the sheer exhaustion. Defensive organizing makes leaders focus on strong signals and ignore weak ones. People are more likely to fall prey to it when they cannot admit to, share, examine, and process the anxiety that comes with living and that leading only amplifies.

The Anxiety of Leading

Our research was a study of failure. We documented a leadership team defeated by anxiety. However, much work in clinical psychology and organizational behavior offers useful pointers for countering defensive organizing. Dealing with anxiety productively, that work suggests, does not require strategic or structural changes. It requires changes in mindset and relations. Leaders must stop estranging anxiety and begin to befriend it instead, turning anxiety into an inevitable and even informative experience that is best managed by sharing it with others.

Working well with anxiety requires three capacities that leaders must cultivate within themselves and demonstrate around them: the courage to recognize anxiety, the curiosity to inquire about it with others, and the care to see and soothe distress. Each of those capacities counters one of the stages of defensive organizing we described above, whereas a capacity’s absence helps that stage take hold. What follows is a brief guide for cultivating and demonstrating those capacities.

Brave the worry. The first exit from defensive organizing involves challenging the heroic stereotype of leadership. In our study, executives on the leadership team acknowledged their anxiety in private, but they did not share it. Like many leaders we have worked with, they saw their worries as shameful weaknesses that a strong leader should keep under control.
As neuroscientist Joseph LeDoux discovered, however, human brains are not designed to control anxiety. Our worries are too useful for survival. It is anxiety that can hijack our brains. When anxious brains belong to leaders, we found, anxiety hijacks organizations, too.

While anxiety can hardly be controlled, it can be managed. Doing so requires a different kind of courage than the one featured in heroic leadership portraits — the kind for which scholars have borrowed John Keats’s poetic term “negative capability.” That is the courage to be with one’s experience: to recognize without judgment or reaction when and how one gets anxious. Perhaps it manifests as difficulty sleeping, drinking more than usual, or feeling restless and ill at ease.

Reframing anxiety from a personal failing or a dire prediction to a simple signal that requires attention and interpretation, like all other data, makes it easier to demonstrate courage too. You might say, for example, “Before I share my vision for what we might do, I want to say that I find our situation worrying, and I wonder if others do too.” Then invite a conversation about why we worry and what that signal says about the team and its circumstances.

This is not about being vulnerable; rather, it’s about showing the bravery to normalize acknowledgements of anxiety. Doing so can be framed as a sign of strength. “I do not want us to be the kind of leaders who jump into reckless action because we are too scared to think,” you might say. “I am confident that we can figure out what we are concerned about, share it with each other, and take care of it.”

The courage to be the first to say what everyone else is thinking is the antidote to idealization. It signals that no one needs to perform confidence they do not feel by feigning certainty they do not have. If covert anxiety hijacks sensemaking, emotional honesty frees it up again.

Find your truth tellers. In our study, the leadership team defensively dismissed feedback from senior managers who knew them well and were close to operations. But those people are a leader’s best allies to stress-test a vision.

Cultivating curiosity begins right after you find the courage to acknowledge your anxiety. Once you have a good understanding of what worries you, as discussed above, find allies you trust to share your anxieties with and to challenge you, your team, and the viability of your vision. Tell them that you care about their views and that you hope they care enough to tell the truth.

You are not asking for criticism or appointing a devil’s advocate; those only trigger more anxiety. You are asking for robust support. You are imparting the same emotional honesty that, ideally, you have cultivated and demonstrated with your team. You might seek feedback in two areas: What would have to change in the way you work, and the way you work with others, for the vision to succeed? And what would have to change in the way the business works? Then listen carefully. Pay attention to insights that you find unfounded or unpleasant. If the feedback seems inaccurate, ask yourself what data you could share to make it more precise. If it seems inconsiderate — that is, fails to recognize your own efforts and the difficulties you are facing — ask yourself whether you have shown how much you care.

Curiosity, like courage, requires one to suspend judgment. When you notice contempt for another group’s competence or commitment, ask not “How do we fix them?” but “What is this judgment protecting me from?” That question will help you avoid projecting onto others what you fear in yourself and will foster collaborations that modulate anxiety and unlock change.

Care visibly. The third capacity that helps leaders work with anxiety is care: giving it and receiving it. Cultivating care toward oneself, through mindful practices and supportive relationships, is essential to demonstrating care toward others. Leaders cannot care if they do not have energy, and they are seldom capable of showing care in ways that they have not experienced themselves.

Leaders who can acknowledge and investigate their own anxiety are likely to realize that others feel it too. One senior executive we know was aware that his anxiety sometimes made him initiate change, and that change became a source of anxiety for others while old ways were being dismantled and new ones were still works in progress. “If others help reassure me by going along, I need to reassure them by being there,” he noted. That comment captures what it means to work with anxiety — to care rather than defend against it.

During a long change process, a gap will often open up between public commitment and private concerns, between what is said in meetings and what is grumbled about in the corridors. You might be tempted to try to close that gap by pushing people to speak up about the actions they will take to accelerate change. What we found, however, is that those pushes ring hollow because they let people avoid a conversation about what they are feeling.

If you take time to genuinely care what people are feeling, you will find out what is working, what is not, and who needs what. The organization we studied failed to do this. It only needed its leaders to acknowledge that they had lost the plot and that they cared about their employees’ exhaustion, not just about their productivity. Instead, the leaders invested their care in devising a performative ritual to reengage the crew.

Better care can take two forms. One is interpersonal. It begins with asking, “Whose efforts are being ignored?” and showing them gratitude, or, “Who is feeling forced to pretend that they are OK when they are not?” and letting them know that they can stop. Another is institutional. It begins with asking, “What is working, and what is not?” and directing resources toward the former while pruning the latter. Both forms of care show that leaders are committed to supporting an ongoing change process, not imposing a change initiative.

Taking care to develop a culture where people can share their worries and what might be causing them is the best antidote against anxiety festering into a collective feeling that nothing matters, no one can be real, and the safest choice is to hide and wait. A culture that estranges anxiety constrains not only leaders but everyone from working productively with anxiety.

The Only Constant

When we say that anxiety is only human, we mean that it is inevitable in the life of an organism that can forge attachments and imagine the future. As people do both, they come to worry about loss — of face, of people, of activities that give them meaning, and of life itself. This is especially so when events remind them that loss is possible. Anxiety in the face of uncertainty, then, is not a leadership failure. It is a leadership condition. Our research and our work with leaders show that when they acknowledge uncertainty but deny anxiety, the latter becomes overwhelming and distorts how they deal with the former. It pushes people to make and to fall for empty promises, to blame others for their own failures, and to eventually withdraw.

When we look around in the workplace and beyond, we see those defensive patterns more and more. They draw energy from the shame of acknowledging anxiety. This article is an attempt to make anxiety less unthinkable or unspeakable, and less shameful. Anxiety can be informative and humanizing. If you meet it with courage, curiosity, and care when it emerges within you, you will be able to show the same capacities to people around you. You will grow into a more connected leader, and your organization will grow more adaptive for it.

John Pearson has been CEO of DHL Express and on the DHL Group board of management since 2019. He joined the global logistics and courier company in 1986 and has held senior management positions in its divisions in the Middle East, the Asia-Pacific region, the U.S., and Europe. MIT Sloan Management Review spoke with Pearson about DHL’s commitment to a corporate culture rooted in its purpose: “Connecting people, improving lives.” This interview has been edited for clarity and length.

What do you see as the role of corporate purpose in supporting alignment and performance at DHL Express?

John Pearson: It starts with our traditional four pillars: hiring motivated people, driving service quality, creating customer loyalty, and delivering a profitable network. I simplified that to three letters: P plus Q equals G — people plus quality equals growth. In both cases, we start with people — getting people engaged and being a great place to work.

People are at the center of everything we do, and what people relate to most is purpose: knowing what we turn up to work for, what we do every day.

There are two sides to purpose at DHL. One side is what we do every day: connecting people and connecting businesses. We know that society and countries are better off when they are more open and connected. The other side amplifies that through our Go programs that support employee volunteering and social impact: GoTeach, GoHelp, GoTrade, DHL’s Got Heart. The more I speak with young people coming into our organization, the more I realize that while they’re proud to say they connect countries and businesses, they get particularly motivated by that second side, and they ask for more volunteering time. Being a global company in 219 countries, which is the most global company of any, there is a lot of opportunity to volunteer in small and rather remote communities all around the world.

Does support for corporate purpose factor into managerial evaluations?

Pearson: We ask our leaders to take self-assessments where they rank themselves low, medium, or high on six leadership attributes. One is how much they manifest and develop the purpose of the company. Leaders have to give themselves one low mark and one high mark. It used to be that purpose was the default option for the low mark among about 60% of our executives; they might think, “Maybe I’ll put that as the low [attribute] because we haven’t talked about it much.” I raised this and talked about purpose with members of my team. The following year, that 60% dropped to 40%, and then it became around 20%.

Do you have advice for leaders who are considering how their organizations can express their purpose through action?

Pearson: You should use the assets that you have to define how your purpose is implemented. Our assets are aircraft, trucks, vans, and having people in just about every locale in the world. We are located on just about every runway on the planet. In partnership with the U.N., we’ve developed the GARD program — Get Airports Ready for Disaster. The program helps airports around the world be better prepared to manage incoming humanitarian aid and personnel when disaster response situations arise. This is a great example of using what you have in your toolkit to live your purpose.

Human beings are not good at separating fact from fiction. But surely hard-nosed B2B professionals are different? Our research suggests otherwise.

In two experiments with 622 B2B sales professionals, we gave participants a negotiation scenario in which a buyer either lied or told the truth. We then gave half the participants a story from the buyer — a brief anecdote about their company helping a farming family in need. The other half received no story.

Salespeople who read the story were 17% more willing to make concessions and developed 10% more trust in the buyer’s integrity. Surprisingly, the story worked just as well even when the buyer had lied.

Why? Psychologists call it narrative transportation: When people become immersed in a story, they momentarily lose sight of their context and their skepticism. Think of crying at a movie you know is fictional.

This vulnerability may become harder to guard against as AI tools proliferate. In a separate unpublished experiment, only 17% of 308 participants recognized that their negotiation counterpart was a bot. The inability to distinguish human from AI, coupled with our vulnerability to narrative persuasion, points to a risk that warrants attention as AI becomes more capable.

Together, these findings suggest that in negotiations, we are vulnerable to deceptive storytellers — human or not. Here are three things managers should do differently.

1. Don’t decide in the shadow of a story. Stories are most persuasive the moment they’re told, because that’s when human judgment is most impaired. Introduce a simple rule: No pricing or concession decisions during or immediately after listening to a storyteller.

2. Separate persuasion from verification. Assign someone on your team the explicit role of checking facts in real time, not participating in the negotiation. Their job is to surface inconsistencies while the main negotiator is in the flow.

3. Verify your counterpart, not just their claims. Don’t assume that you’re interacting with a human. For high-stakes negotiations, move to a video call or incorporate checks that require human judgment.

In a world where machines can negotiate on someone’s behalf, knowing when you’re being told a story — and by whom — has never mattered more.

Companies are betting that AI chatbots will deliver faster and cheaper customer service. But if you’ve ever tried to circumvent a chatbot and get to a human, you’re not alone. Here’s what three recent studies discovered about when customers will and won’t let AI do a human’s job.

1. Customers avoid chatbots for two compounding reasons. In a study simulating a customer service scenario, participants repeatedly chose between two unlabeled options: One required waiting in line before their request was resolved with certainty; the other skipped the line, but occasionally failed, routing the customer into the line for the first option. Researchers designed the choices so that a person optimizing for time saved should have picked each option about equally often. Instead, participants chose the no-queue option just 28% of the time — a reluctance researchers call gatekeeper aversion, driven by its uncertainty and multistage structure, regardless of who or what runs it. When that same no-queue option was presented as a chatbot rather than a person, adoption fell by another 10 to 20 percentage points — a separate effect called algorithm aversion. Follow-up experiments hint at possible remedies: Offering transparency about what the chatbot can and can’t do, and showing customers the expected wait time for each option, appears to increase chatbot uptake.

2. AI is a better messenger for bad news; humans, for good news. Across several experiments, customers who received a worse-than-expected offer (say, a low resale price) were more likely to accept it from an AI than from a human agent. In one study, 78.6% accepted an AI’s offer, versus 60.4% for a human’s offer. But in another study, when the offer was better than expected, the human agent’s offer was accepted 89% of the time, versus 76% for the AI. The reason: People don’t ascribe human intentions to AI, so don’t regard it as “selfish” when it lowballs them, nor as “generous” when it overdelivers. The effect is strongest when the AI is presented as machinelike; a humanlike persona erodes the advantage for delivering bad news.

3. A simple two-question test can predict whether customers will embrace or reject an AI. A meta-analysis of 163 studies involving over 82,000 participants found that customer preference for AI over humans comes down to two factors: whether the AI is seen as more capable at the task than a person, and whether the task is seen as requiring personalization. When AI is seen as more capable and personalization is seen as unnecessary, such as when forecasting sales or playing chess, people prefer it. In every other combination, people favor humans out of a desire for individualized treatment. Before automating a customer-facing role, leaders should weigh AI’s capability against customers’ expectations of personalized service.

Getting manager buy-in for a good idea is harder than it should be. Our research points to a communication style that reliably breaks through this wall — and that is disproportionately exhibited by white-collar employees with working-class origins.

We studied how employees present ideas to their managers by conducting a field survey of employees and their managers in a shared-services division of a multinational engineering company in India (218 employees, 32 managers); running three follow-up behavioral experiments involving more than 1,000 working adults and university students; and interviewing dozens of professionals. In each study, we found that employees from working-class backgrounds who had moved into white-collar work, whom we call upward transitioners, were more likely to acknowledge that they might be missing something, invite the manager’s perspective, and appear open to revising their position.

In interviews, 77% of upward transitioner employees described communicating this way, compared with just 15% of colleagues who had always occupied higher-class positions.

We then tested whether this difference in tone and attitude affects how managers respond to an idea.

In one experiment, some participants were asked to take on the role of either a low- or high-power manager. Each listened to a participant cast as an employee deliver an idea either as a firm recommendation, in a highly confident manner, or as a more open-ended recommendation, in a humble tone that allowed for input. Each manager was asked to rate how willing they were to act on a recommendation delivered in one of the two styles. Among managers cast as having little power, the two pitches landed nearly the same way. Among managers cast as powerful individuals — people with the most resources to act and the most to lose by publicly reversing course — the open-ended pitch landed far better.

Why? Communicating in an open manner lowers a manager’s defensiveness, shifting the conversation from who’s right to what could work. Inviting a manager to help shape the idea prompts them to develop a sense of ownership over it. Both dynamics are important when pitching ideas to powerful managers, who can be resistant to ideas that are not their own.

Across industries, companies are investing unprecedented sums in reskilling programs to prepare employees to use emerging technologies.1 The programs are typically built around forecasts of which skills, such as data literacy, digital fluency, systems thinking, and adaptability, will matter most. Each year, when the forecasts are updated, training courses — and their related costs — proliferate.

Yet in the three years we spent studying 10 European manufacturers navigating exactly this kind of technological change, we found that the most relevant new skills workers developed were almost never the ones that had been forecast. They emerged organically, as workers and managers figured out together how to make new tools fit the existing work — or realized that they could not. In most cases, these skills were recognizable as important capabilities only to the few managers who understood where and how to look for them.

At a well-known Italian furniture manufacturer in our study, the head of the varnishing department wanted to identify and support new-skill development. Walking the floor was part of his routine, but what made him unusual was how he responded to what he saw. When workers raised concerns about equipment used on the job or devised their own ways of handling an awkward step, he carried those observations to a newly appointed head of production and negotiated changes to the workflow. He treated the production floor as the site of ongoing capability development, and his job as the connective tissue between the people doing the figuring out and the people with the authority to act on it. What he was watching grow was concrete.

At this company, the rollout of new production equipment, including computer numerical control (CNC) machinery, was steadily turning manual artisans into machine operators and digital production monitors. That meant that craftspeople’s judgment about quality and finish now had to be expressed through digital settings and on-screen interfaces. The varnishing head’s own emerging skill was a form of shop-floor diagnosis and reengineering: spotting where a new machine hindered or disrupted the work and devising a fix for it. It was a bricoleur capability that sat between hands-on craft and process engineering — one that no job description had ever named.2 The workflow changes that he negotiated were the visible trace of that skill taking shape, within both him and the workers.

Our research found that the companies managing technological transitions most successfully were not the ones with the best skills forecasts. They were the ones whose managers had developed a particular habit of attention — one that let them see what was already emerging in the work and to harness it before an employee walked out the door with an emerging skill. We call this practice SPOT. Later, we’ll explain what it is, why it matters now, and how to start doing it within your own organization.

The Forecasting Trap

Walk into a large company today and chances are you will find someone building a future-skills matrix. The rationale is simple: If we can name the skills we will need, we can train for them in advance. But this logic does not survive contact with an actual shop floor.

From 2023 to 2025, as part of the Horizon Europe Up-Skill project, our team conducted ethnographic field work in 10 companies across Europe, from a large automotive manufacturer to small artisanal workshops. Each was adopting advanced manufacturing technologies, such as collaborative robots, mixed-reality training systems, and 3D printers.3 We watched these companies discover that the skills they needed became visible only after the new technology they had introduced collided with work on the ground.

At one company in Sweden, to help workers learn lock-assembly procedures, managers introduced a mixed-reality system — a headset-based class of tools that overlay digital guidance directly onto the physical workspace, blending elements of virtual and augmented reality.4 The visual aspects of the system could show workers what to do but could not convey the reasoning behind the steps. Workers and managers eventually developed workarounds together, and the company found that the tacit understanding of the process the system was supposed to capture was the very thing it could not. At another Swedish firm, a plan to automate a grinding operation fell apart because the automated line could not replicate the judgment of experienced human workers. Only when the automation failed did the depth of the human expertise become visible.

But the more revealing part of each story is what the workers built next. At the lock-assembly company, the gaps in the system became the catalyst for developing genuinely new skills: Workers learned to program, re-sequence, and troubleshoot the system themselves, and they worked out how to teach the unwritten “why” that the headset left out (for instance, why a particular part of the lock should or should not be greased) so that the reasoning could pass from one person to the next. At the firm whose grinding line resisted automation, a new digital system for tracking production had a parallel effect: As operators worked with it, they began to read how their own task fed the wider flow of the line — a kind of systemic awareness that the job had never previously demanded.

It is worth separating two things in these cases. What the machine could not do exposed a skill the workers already had; what the workers built around its limitations was the skill that was genuinely new: the programming, the teaching of the “why,” and the new perspective on the whole production line.

A third firm, a small Italian manufacturer of high-end accordions, watched a competitor adopt robots for a sensitive manual step. It decided not to follow suit because it suspected that the competitor was automating away something the robot could not replicate.

We observed a pattern: The skills that matter most during a technological transition are the ones that surface when the new tool meets the old workflow: when something breaks, when a worker improvises a fix, when a manager notices that the thing the machine cannot do is the thing the customer is actually paying for. You cannot forecast what has not yet emerged. So the question for leaders is not “Which skills will we need next?” It is “Which skills are already trying to grow inside our company, and are we paying enough attention to notice?”

The SPOT Framework: Seeing and Growing Emerging Skills

We developed SPOT — a mnemonic for see, partner, orchestrate, transform — as a framework for capturing the habits we observed among the managers who were best at identifying, stabilizing, and retaining emerging skills. Let’s explore each of the four elements.

See the invisible. Most managers walking a production line look for problems, but an emerging skill does not look like a problem. The head of the varnishing department we mentioned earlier was not scanning for failures. He was scanning for moments when someone was solving a problem the system had not anticipated.

The skills you are trying to see are ones the worker cannot yet fully articulate. If you ask, “What new skill are you developing?” you will get a shrug. The better questions are about the task: What is this machine doing today that it was not doing last week? What are you doing differently since the new line came in? The skill is hiding inside the answers.

Take the high-end furniture manufacturer. The useful question its managers learned to ask was not “Can you run the machine?” but “How did you decide on that setting?” The answer revealed a skill that had migrated rather than disappeared. An artisan reads the grain, the density, and the absorbency of a particular piece of wood — judgments that were once expressed through the hand and eye alone — and now translates them into the digital settings that drive a CNC machine, and the on-screen checks that monitor its work. That blend of material sense and interface fluency is itself the new skill, and it lives in the doing, not in any manual that could have been written in advance.

Managers who have been in the same department for years may have difficulty seeing what is emerging without making a deliberate effort to reframe their scanning approach. (See “Four Questions for Your Next Floor Walk.”) Rotating managers into unfamiliar settings, or deliberately hiring from outside the function, may provide a fresh set of eyes that are better able to see an emerging skill. The newly appointed head of production at the furniture company was effective in part because he was able to notice what had become overly familiar to others.

Partner with workers. Seeing signs of emerging skills is only the beginning. An emerging skill lives in the head and hands of the person developing it. The manager’s job is not to diagnose it from the outside. It is to sit alongside the worker and interpret it with them.

At a small manufacturer of brass musical instruments in the United Kingdom, management and workers were actively discussing whether collaborative robots (those designed to work safely alongside humans) could handle delicate components. Rather than making a decision in isolation and rolling out the technology, the company was treating the question as something to be worked out with the people whose work the cobots would affect. What the joint evaluation surfaced, though, was a skill the company had never named. To judge whether a cobot could be trusted with delicate, one-off components, workers had to articulate exactly what they themselves were doing: reading the small irregularities of a handmade piece and adjusting their handling by feel, one piece at a time. That judgment had always been treated as simply “how the work is done.” Putting the skill into words turned it into an explicit capability — one the company could then choose to protect, teach, and build on rather than lose through inattention.

Schedule a conversation whose only purpose is to understand how the work has changed since a new tool arrived. Ask the worker to describe what is different and then ask what they would teach someone who was about to take over the job. The answer to that second question is almost always the emerging skill. Pay attention to where the worker hesitates or gestures instead of describing. Those are the places where capability is forming. At the lock-assembly firm, the answer to “What would you teach your successor?” was not the sequence of steps that the system already displayed but the reasons behind them: the unwritten logic that tells an experienced worker when the standard procedure should not be followed. That is the skill the conversation is trying to surface.

Orchestrate learning in real workflows. Once a capability has been surfaced, the tendency of most organizations is to pull the emerging skill out of its context and turn it into a training course. That rarely works. Skills that emerge in the flow of work tend to die when they are lifted out of it, because they are closely tied to the specific problem they were solving.

At a smoking-pipe maker, an employee with engineering and 3D printing expertise took on a substantial project: developing an in-house solution for producing mouthpieces. He designed a prototype machine and partnered with another company to manufacture it, resulting in a custom lathe integrated with bespoke software. He did not develop this capability in a training room. He developed it while solving a concrete production problem. The project was the curriculum, and it led to a highly specific new capability. It was not simply “3D printing” or “machining” but the ability to combine engineering judgment, hands-on additive-manufacturing experience, and software integration well enough to specify, commission, and program a custom production machine from end to end. That composite skill existed nowhere in the firm before the employee assembled it on the job, and it is now part of what the company can do.

When a worker shows signs of developing a new skill, resist the temptation to let them practice in a sandbox. The skill will develop faster if it is applied to a real production problem with actual stakes. Your job is not to remove the risk but to make the environment around it supportive enough that the worker can learn from what happens. At a specialty print firm, workers migrating a manual engraving process onto a new machine were themselves contributing their tacit knowledge to the digital systems. When tacit knowledge must be translated into something a digital system can use, the person best placed to do the translating is the person whose knowledge is being translated. And the translation is itself the emerging skill, not a preliminary to it. Learning to turn a feel for the work into instructions a machine can follow happens only on the live system, against real material and real consequences, not in a classroom where it is rehearsed in the abstract. The ability to transfer knowledge to digital systems becomes a new skill that the company has at its disposal when new digital technologies come along.

Transform insights into lasting capability. New skills identified and developed in the three steps above can live in the head of the worker and the memory of the manager for a while but will eventually vanish if not transferred to a more robust medium.

Let’s return to the accordion maker. The firm’s refusal to follow its competitor into robotic production looked, from the outside, like a conservative choice. Viewed through SPOT, it was a transformative decision. By drawing a circle around a capability it had recognized but could not yet fully specify, the firm converted a tacit and fragile skill into a strategic commitment the organization could articulate and defend. The subtle feel and sound of handcrafted components was no longer something the firm happened to have.5 It was something the firm was now explicitly protecting as a matter of policy. That stance, not any specific training program, is what locked the capability into the organization’s future — what is sometimes referred to as a company’s DNA.

A parallel example came from the furniture manufacturer’s varnishing department. The newly appointed head of production was actively encouraging employee involvement in technology adoption across the company. What began as one department head’s way of working was being supported and, in the process, normalized by a shift in managerial culture at the top. The transform move was the institutional decision to frame bottom-up innovation as how the company worked rather than as an exception or aberration. What had started as the varnishing head’s hybrid skill — reading where a machine failed to support the work and adapting to that — started as one individual’s practice before it migrated to the rest of the firm. Treating support for bottom-up innovation as established practice turned it into an embedded competence that could deliver competitive advantage into the future.

SPOT is not a new training methodology. It is a reorientation of managerial attention, away from the forecasting of skills and toward noticing the ones emerging quietly right in front of you, if you care to look, as workers solve problems and get on with their day. This managerial work is slower and less visible than delivering training and certainly less dramatic than strategic restructuring. But it is also, based on our study of manufacturers adapting to technological change, what actually works. Start with one floor walk this week, incorporating the SPOT framework, and see what your dashboard never told you.

By the ordinary measures of any new technology, the current wave of generative AI has moved fast. By some estimates, about 2.4 billion people worldwide use generative AI platforms each month, and coding agents have changed how software is written.

Efforts to commercialize the technology have scaled just as fast. AI coding platform Cursor reportedly passed a $2 billion revenue run rate by early 2026, and, as of April 2026, Perplexity was reported to have more than 100 million monthly users across its products by challenging one of the internet’s most entrenched markets: search. And this growth is not confined to AI-native companies. Salesforce’s Agentforce has reached $1.2 billion in annual recurring revenue, Harvey has spread across large law firms, and Shopify has made AI use a baseline expectation across its operations. By many conventional markers, these developments increasingly resemble the early stages of a platform ecosystem.

Yet AI’s larger promise is not to become another successful technology platform. It is to become a true general-purpose technology — like electricity or the internal combustion engine — that reshapes organizations, industries, and, ultimately, the broader economy. Judged against that standard, progress remains shallow.1 The process of complementary innovation, organizational integration, and economywide transformation expected of a general-purpose technology remains in its infancy.

The obvious culprits — immature models, ordinary adoption friction — are real, but they’re not the constraint. Like earlier general-purpose technologies, AI will not become economically transformative simply because it is broadly applicable. It will realize that potential when a surrounding technological, industrial, and institutional architecture enables decentralized organizations to confidently build upon it — in other words, when the technology becomes platformed.

Here, I will explain what platforming entails (the technological, industrial, and institutional architectures a technology needs), why AI remains only partly platformed, and how organizations can innovate and invest effectively while that process is still unfolding.

Platforming a General-Purpose Technology

Scholars have long argued that general-purpose technologies are able to transform economies because they can be applied across many industries while stimulating successive waves of complementary innovation — as was the case with electricity, the steam engine, and the internet.2 By the same token, the potential of such technologies is unusually hard to realize. Broad transformation requires large numbers of independent organizations to make interdependent investments; redesign products, processes, and business models; develop new capabilities; and coordinate despite deep uncertainty about how the technology and its ecosystem will evolve. Therefore, the central challenge is creating the technological, industrial, and institutional conditions under which decentralized organizations can confidently build upon the technology — a process of platforming. Someone has to build that foundation: It is what makes decentralized downstream integration, complementary innovation, and co-invention possible at all.

Electrification illustrates this. Electricity was technologically proven and commercially viable by 1882, yet widespread electrification did not follow for nearly four decades. Technological architecture stabilized when the Niagara Falls hydroelectric power project (1895-1896) confirmed polyphase alternating current at commercial scale. Industrial architecture matured as a division of labor settled among utilities, equipment makers, and financiers, under the regulated utility model that took hold between 1898 and 1907. Institutional architecture followed, with the first comprehensive state public-utility commissions forming in 1907.

As these technological, industrial, and institutional architectures progressively aligned, organizations gained sufficient confidence to invest and experiment, and electrification accelerated. Platforming emerged through the combined efforts of inventors, manufacturers, utilities, financiers, standards bodies, and regulators. Other general-purpose technologies — notably, personal computing — illustrate alternative pathways: Platform leaders, such as Microsoft and Intel, more directly orchestrated these architectures to support ecosystem growth.3

A technology becomes platformed when a surrounding technological, industrial, and institutional architecture creates conditions stable enough for decentralized organizations to confidently build upon it. Platforming reduces uncertainty by stabilizing expectations about how the technology can be used and how it will evolve. It establishes clear lanes for complementary innovation — where to innovate, where to rely on others, and what can be treated as stable — and the governance, rules, and incentives that enable organizations to capture value from their investments while coordinating with others. Decentralized investment and experimentation can then scale from isolated successes into broad transformation through the alignment of the three architectures.

The Platforming of AI: Where Are We Now?

The platforming of AI remains incomplete, but recognizable technological, industrial, and institutional architectures are emerging. Understanding what has stabilized — and what has not — clarifies the opportunities and the frustrations of building on AI before it has been fully platformed. Let’s take a look at the current state of AI.

AI’s technological architecture is still emerging. Today’s dominant AI architecture rests on a relatively specific trajectory, especially among leading frontier developers: pretrained, predominantly language-based foundation models; specialized hardware; cloud-based training and inference; and API-mediated delivery.4 AI is taking shape as a layered stack — chips, cloud infrastructure, foundation models, and the applications built on them. (See “Key Elements of the AI Stack.”) The stack’s lower three layers are converging on a centralized foundation in which model development and most computation reside with a few cloud-hosted frontier models, with most organizations consuming intelligence remotely through APIs rather than owning it. This departs from the digital services economics that were once taken for granted: Rather than distributing software that runs locally at little additional cost, AI delivers intelligence through continual, cloud-hosted inference, performing heavy computation each time intelligence is used.5 Although the prevailing architecture continues to evolve, the likely alternatives — open-weight ecosystems and parallel stacks developed by Chinese companies — are variations on it rather than fundamentally different trajectories. It’s likely that to the extent it continues, much of the uncertainty around the lower layers will subside.

The application and deployment layer, where most organizations hope to build complementary products and services, remains fluid. As emerging orchestration, agent, and middleware layers compete to define how AI should be integrated into products, workflows, and enterprise systems, some companies build around chatbot interfaces and others directly on foundation model APIs. Meanwhile, frontier models keep absorbing capabilities that many people expected to reside elsewhere: Enterprise search, retrieval from knowledge bases, and persistent memory are increasingly being handled by the model rather than by separate software. It remains unclear which abstractions, interfaces, and patterns will become stable enough to support broad complementary innovation.

AI’s industrial architecture is still cohering. Technological architecture determines how intelligence is built — the trajectory and approach, and the division of the larger problem into components — whereas industrial architecture determines who builds what: how the ecosystem divides problem-solving and commercial activity across specialized organizations, and how value creation and capture are distributed among them.6 Whether that division of labor can be occupied at all is another matter — one that is dependent on the capabilities organizations build, the skills the labor market supplies, how organizations align with one another and with the wider economy, and the returns that sustain them. Where these misalign, complementary innovation falls far short.

A recognizable division of labor has begun to emerge around the lower layers of the stack: Nvidia in AI accelerators; Amazon Web Services, Microsoft, and Google Cloud in compute; OpenAI, Anthropic, Google DeepMind, and xAI in frontier models; and Meta (Llama), Mistral, DeepSeek, and Alibaba (Qwen) in the open-weight ecosystem. Much of the ecosystem’s measurable investment is concentrated in these foundational layers — chips, compute, power, data centers, cloud, and frontier models.

Some businesses have clearly emerged as complementors on top of the foundation models. But AI has produced nothing like the governed marketplaces and stable interfaces earlier developers could build on — the app stores of iOS and Android, or the backward-compatible APIs of Windows — with the clear categories and rules that once gave thousands of them the certainty to commit. OpenAI’s GPT Store has stayed thin; meanwhile, a different candidate layer is forming one level up, around AI app builders such as Replit and Lovable, on which nondevelopers can generate and ship software. Whether a durable application economy consolidates there, among the model providers, or within incumbent suites is the contest still unresolved.

For now, AI reaches users through a heterogeneous mixture of forms: the frontier providers’ own applications, applications built on frontier models, agents, vertical applications, AI embedded within incumbent software, and proprietary enterprise deployments. Even what counts as the application and deployment layer is unsettled: thin chatbots, AI features embedded in existing tools, orchestration and middleware, wrapper apps, or third-party agents.

Vertical integration further blurs these boundaries as frontier developers move upward into user-facing products while incumbents embed foundation models throughout their suites. Rather than competing within established categories, companies are competing to define them — advancing not merely different products but competing hypotheses about how foundation model capabilities should be organized, accessed, and converted into value.

Meanwhile, returns on complementary investment remain uncertain, most fundamentally with regard to the industry’s eventual division of labor, dominant application architecture, and sources of durable advantage. AI also runs into a familiar conundrum: The same foundation models that reduce the cost of innovation also reduce the cost of imitation, making differentiation harder.7

Investments in the lower layers also face uncertainty, but for different reasons. Frontier-model developers are pouring vast sums into future scale, ecosystem leadership, and pricing power — expectations that depend on an industrial structure that has not yet emerged. Switching costs remain modest, customers frequently use multiple models, and the mechanisms that historically produced durable platform leadership have yet to develop. Meanwhile, training and inference costs weigh on profitability, and capable open-weight models keep pressuring proprietary ones. Ultimately, these investments are a bet that AI will become platformed enough to generate value sufficient to justify today’s spending.

AI’s institutional architecture is emerging slowly. Institutional architecture governs how decentralized organizations coordinate, invest, and build on a common foundation. Successful platform ecosystems require more than technology and market forces; they require institutions and governance that let organizations invest independently while remaining collectively coordinated. AI’s institutional architecture is emerging through standards bodies, consortia, technology providers, and governments developing protocols, but it remains far less developed than the technological and industrial architectures.

Institutional coordination can arise through several mechanisms. Governments can establish legal frameworks and public standards; industry alliances, standards bodies, and multiparty initiatives can coordinate interoperability and shared conventions. In AI, we might particularly expect the emergence of platform leadership, in which a central company organizes the ecosystem by establishing stable interfaces, governing participation, signaling architectural direction, committing to what it will not absorb, and creating credible incentives for complementary innovation. Such leadership is itself an investment: It takes a company with enough platform power to set and enforce the terms and enough incentives to bear the cost. Microsoft and Intel exemplify that in personal computing, Apple in the iPhone, and Google in Android.8 Today we can see that while many governments have taken a light-touch approach, industry consortia, evaluation frameworks, and private protocols are emerging.

A shared protocol, such as Anthropic’s Model Context Protocol, is a useful standard, but platform leadership runs far deeper. It is the active coordination and orchestration of an ecosystem that extends well beyond the platform owner’s own boundaries — governing participation, aligning incentives, signaling architectural direction, committing to what it will not absorb, and giving large numbers of independent companies the confidence to build. In that deeper work, today’s frontier companies are investing comparatively little. The ecosystem lacks not leadership in technology but leadership in coordination.9 Someone has always supplied that coordination — such as a lead company in personal computing, and public authorities and engineering bodies in electricity. In AI, organizations are integrating vertically instead, which is not the same thing.

In the absence of mature platform leadership, many leading companies are instead solving coordination problems through vertical integration, combining frontier models, cloud infrastructure, developer tools, enterprise software, consumer applications, and distribution within integrated ecosystems. But direct control by one company is not ecosystem governance: It forgoes the diversity and decentralized effort of large numbers of independent complementors — the engine of broad transformation — and so may postpone the mature platform ecosystem that it appears to be substituting for.

Building and Innovating Before AI Is Fully Platformed

Most companies will not compete by building frontier models. They will compete by building on them — integrating them into products and operations, and creating the complementary goods and services around them. That places them, awkwardly, where the architecture is least settled: Although the lower layers are converging, the application and integration layer, where most of this building happens, is the part still in flux.

It can seem natural to wait for the costs, risks, and uncertainty of building on AI to fall as the technology becomes platformed. But the real challenge is to keep them in mind and act anyway — for a company to invest in the ways that best strengthen its position while managing the problems of building on a general-purpose technology that has not yet been fully platformed. That means making investments that will pay off however the architecture settles.

Learn faster than you commit. When the architecture is unsettled, what a company learns is worth more than what it locks in — and the cost of learning is unusually low right now. Much of the experimentation can run on open-weight models — such as Llama, Mistral, or DeepSeek — on a company’s own hardware, where the marginal cost of a query is near zero and proprietary data never leaves the building. Frontier models can be reserved for the work that genuinely needs them. The most valuable thing a company can record — where its experts overrode the model and why — is also at its richest now, while the models are still making enough mistakes to generate corrections. Those corrections also map the jagged frontier of what AI does reliably on a company’s own tasks. But that margin is closing: Once a model reliably beats a company’s experts on a task, the corrections, and the signals they provide, disappear.

Build assets that will survive architectural change. No one yet knows which model, stack, or way of organizing intelligence will win, so businesses should be wary of making bets specific to any one of them. The danger is subtler than it looks. Most companies ask only the technical question “If the model were swapped tomorrow, would the system still run?” and stop there. But even a perfectly swappable model can be locked in commercially, through the way a vendor bills and what its contract permits. Salesforce’s Agentforce meters agent work in “agentic work units” it alone defines, and SAP’s 2026 policy restricts what outside AI is allowed to do with the data inside its software.10 A company should keep its options open at the applications level, too. Holding back on bets wired to an architecture that is still unsettled is a strategy in itself — not a failure to act.

Invest in complements, not intelligence. Nearly every company will buy rather than build its AI, as will its competitors — who will often purchase the very same models. Intelligence that everyone can rent cannot be anyone’s advantage: The same models lift the floor for a business and its rivals alike, and the work they do well converges toward a common mean.11 Advantage has to come from what the shared model cannot reach, such as proprietary data, domain expertise, trusted customer relationships, distribution, brands, and specialized workflows. This is not a new idea: Rents accrue to the holders of co-specialized complements, not to the freely available input itself.12

Where to invest turns on the seam — the interface between layers where value can be captured. A defensible seam rests on something that the layers above cannot easily reproduce: a regulator’s standing trust, an audit trail, an embedded billing relationship, a proprietary data corpus. This is why many hospitals’ clinical AI runs through Epic: A smarter interface still has to clear the trust bar, which Epic has already done. An illusory seam is a thin wrapper around someone else’s model — useful this quarter but enveloped the next, when the provider folds the same capability into its own product at no extra charge. Before investing in a seam, an organization should ask who will try to take it — the model provider reaching up or the platform incumbent reaching down — and whether what anchors it can be reproduced.

The logic holds even in the case that seems to overturn it: a model capable enough to do the integration, the judgment, and the work itself, leaving little apparent need for a wider ecosystem. Even then, the assets that endure are the ones a model cannot internalize, such as the regulator’s trust, the audit trail, the embedded contract, and the proprietary data. Betting on intelligence pays off only for whoever wins the frontier race; betting on complements pays off regardless of who wins.

The question, then, is not how to own the AI. It is how to own the assets that become more valuable as AI becomes abundant.

Build organizational capability. The most valuable use of AI in this period is the least obvious one: not to do today’s work faster but to become an organization that understands its own workings — where its knowledge sits, how its decisions get made, where its bottlenecks are, and how work moves across its teams. These are the assets the earlier moves depend on. Vendors already have a name for this work — the “forward-deployed engineer” they send to sit inside a customer — and adopters will need the role in-house. The capability a company builds now, such as the memory, the routines, and the judgment about where AI helps and where it does not, is itself among the complements a shared model cannot reach, and it is what keeps producing new ones as the architecture shifts. Learning feeds capability; capability yields the durable complements.

The deep gains from a general-purpose technology always arrive late, and they require two things: The surrounding architecture must settle, and organizations must rebuild themselves around the technology. A company cannot hurry the former. It can begin the latter now, and history suggests that those that do are the ones that pull ahead. Manufacturers that learned to redesign factories around electric power saw gains long before electrification was universal; companies that learned to reorganize around information technology saw benefits long before personal computing matured. The companies that come out ahead in AI may look less like today’s adopters and more like knowledge factories — organizations whose advantage is not throughput but the rate at which they turn their own operations into validated understanding.

Building before AI is platformed is necessarily about investing in what the technology cannot supply — learning, complements, and organizational capability that compound while the architecture is still in motion. It is not about adopting the most tools or predicting the final form first.

After receiving some challenging feedback about my management style, I’ve done a lot of introspection, taken responsibility, worked with a coach, and I believe I’ve made real changes in how I work with others. But I don’t sense that others’ perceptions have changed accordingly. What else do I need to do?

It can be frustrating to do the hard work of growth but then still feel defined by who you used to be. You’ve changed how you engage and react, but your team still seems to see you as the same leader. So, what does it take to be recognized as having changed?

In a situation like this, a few things could be at play. What may feel like a major change to you may be perceived as a minor adjustment by others or go unnoticed. Some people may be reserving judgment, especially if they’ve had difficult interactions with you in the past and want to see sustained evidence of change before they buy it as genuine.

Although we may feel transformed by growth, and that we are beginning a new story, people may interpret our new behaviors through old narratives. For instance, someone who was inconsiderate of others by being consistently late may have tackled this issue and is now regularly on time. However, any slip — just one instance of tardiness — will stand out, reinforcing the original perception.

The persistence of your old image is not a failure of effort but rather a lack of adequately signaling your efforts. It might be difficult to share how you are trying to effect positive change if you feel shame about old behaviors. But one of the most effective ways to signal change is to talk about it openly.

When you are working on a specific behavioral goal, don’t just keep it to yourself. Share it with your team! Let them know what you are trying to improve, and then create a rhythm of accountability. Regularly ask, “How am I doing? What could I do better?” Ideally, ask not in the abstract but with specifics. For example, say, “I’m trying to run meetings more inclusively. At our team check-in yesterday, do you think that everyone who wanted to contribute was able to?”

This technique, which I learned from my mentor Marshall Goldsmith, does two things. It reinforces that you are a leader committed to continuous growth and trains others to notice your changes. When people are explicitly invited into the process, they become more attuned to your progress rather than defaulting to old assumptions.

Equally important is how you reinforce that progress in everyday conversations — without boasting or seeming to fish for affirmation. Simple comments like “I’m trying something new with the weekly meeting agenda so that everyone has a chance to weigh in on what’s working and what’s not” could subtly shift perception. These small signals accumulate over time and help others update their mental models of you.

People can only respond to what they can see. Make your internal evolution visible externally. Signaling change is not about self-promotion. It’s about making your actions visible so that others can clearly see the growth that has already taken place. And those changes in perception will amp up your future-forward leadership.

Leaders may prioritize executing strategy and driving organizational performance, but they are also responsible for developing the next generation of leaders. Traditionally, leadership development has focused heavily on performance — an understandable emphasis, given that leaders themselves are evaluated and rewarded based on execution and results.

However, as organizational strategies increasingly rely on cross-functional teams and external partnerships, employees’ success increasingly depends not only on technical capabilities and performance but on the ability to navigate relationships, influence stakeholders, access information across boundaries, and build credibility beyond their immediate roles. Yet many organizations still develop high-potential talent primarily through experiences that advance performance — that is, the ability to execute tasks. They overlook the importance of providing development experiences that help individuals develop a critical asset for leadership effectiveness and advancement: social capital.

In organizational settings, social capital reflects the trust, goodwill, and credibility that shape how individuals are perceived, influence how contributions are interpreted, and enable the support and advocacy required for advancement. It determines whether performance stays where it is produced or travels across teams, functions, and decision makers, translating into broader visibility, advocacy, and opportunity.

Yet, despite social capital’s influence, few professionals are explicitly taught how to build it, convert it into opportunity, or use it to sustain their career mobility. Rarely explicitly discussed and seldom incorporated into development, it often operates as an invisible system governing how individuals advance within organizations.

Social Capital in Practice

Social capital determines whether performance stays where it is produced or travels to where opportunity exists. To better understand professionals’ perspectives on how it shapes career mobility, I interviewed 21 people representing a range of functions and career stages, from just a few years of experience to more than four decades in the workforce.

While we might expect networking activity to be at the heart of developing social capital, participants in my research rarely discussed it solely in those terms. Instead, they saw it operating in the practical realities of organizational life: who is trusted, who is sought out for advice, whose judgment carries weight, who receives advocacy, and who is afforded grace when mistakes occur. As one research participant put it, “Social capital is the amount of blind faith that someone will give you based on how they feel about you and their prior experience with you.” While formal organizational structures define reporting relationships and responsibilities, participants consistently emphasized that many opportunities, decisions, and career movements are enabled by less visible relational dynamics.

Several interviewees described social capital as influencing how performance itself is interpreted. Strong performance may establish credibility, but credibility alone does not guarantee visibility, advocacy, or advancement. Rather, social capital shapes whether accomplishments remain confined to an immediate team or become recognized across functions, business units, and decision-making circles.

This dynamic becomes particularly important when individuals pursue movement across organizational boundaries. These shifts across teams, functions, or business units are what the study refers to as zigzag moves, in contrast with progress up a traditional, defined hierarchy. Unlike advancing within a familiar environment, making zigzag moves often requires individuals to establish credibility with stakeholders who have little direct knowledge of their previous work. In these situations, social capital — reputation, relationships, and advocacy — provide signals that reduce uncertainty and increase trust. In this way, social capital serves as a mechanism through which performance becomes portable.

As high-potential individuals ascend the ranks of leadership, their social capital becomes increasingly important to how they are evaluated. Their work is unlikely to be directly observed by those considering candidates for advancement, and decisions about opportunities (especially emerging opportunities), promotions, and leadership potential are increasingly informed by reputation, trust, relationships, and the perspectives of others. In these situations, social capital often determines whose capabilities are recognized, whose ideas gain traction, and whose name surfaces when opportunities emerge.

Interviewees suggested that social capital is commonly accumulated through everyday work interactions: moments where credibility is reinforced, trust is established, and relationships are strengthened. It is accumulated in meetings through thoughtful contributions, collaborative problem-solving, and consistent follow-through — small but meaningful social capital deposits that compound over time.

Social capital is also seen as fragile, reflecting the old adage that reputation can take a lifetime to build but be lost in a minute. Trust, credibility, and goodwill can be diminished quickly through poor judgment, broken commitments, lack of reciprocity, or misaligned relationships.

To that point, social events such as company “happy hours” that might be seen as prime opportunities to build one’s network were viewed with caution by participants. One noted, “Social events are always weird. There’s always weird power dynamics. There’s already cliquishness. That social capital has already been established. It’s easy to lose social capital at a social event.”

For Black professionals, these dynamics carry additional significance. Across interviews, participants observed that social capital comes with different risks, scrutiny, and rewards for Black employees. While all professionals are expected to build trust and credibility, participants noted that the same behaviors are not always interpreted or rewarded equally. Black professionals were seen operating with a smaller margin for error, where trust often has to be earned repeatedly and is more easily challenged.

How Social Capital Is Built

Based on the experiences people shared in the interviews, I identified four underlying mechanisms through which social capital is built and converted into mobility: performance, reputation, influence, and network.

Performance reflects an individual’s ability to execute, deliver results, and establish credibility through consistent output. This is the foundation of reputation: one’s subject-matter expertise, ability to solve problems and handle complexity, and reliability. As noted earlier, performance is not sufficient to enable career mobility, but it is necessary for credibility.

Reputation captures how an individual is perceived across the organization based on their behavior, judgment, and consistency over time. It is shaped not only by outcomes but by how individuals engage others across levels, upward, laterally, and downward.

Influence reflects the ability to shape decisions, align stakeholders, and mobilize action without relying solely on formal authority. As roles expand, success increasingly depends on the ability to cooperate with others rather than through direct execution.

Network represents the relational infrastructure through which information, opportunity, and advocacy flow. More than proximity or access, it reflects the strength, relevance, and credibility of relationships, particularly those that extend beyond immediate teams or functions. It also serves as a reputational signal that is reflected in who associates with an individual and whether those relationships amplify credibility and opportunity.

Together, these four components form the architecture through which social capital is built. While each contributes independently to professional effectiveness, participants saw their value as emerging in concert. Performance establishes credibility, credibility develops into reputation, and reputation enables influence and networks. Together, influence and network amplify reputation beyond an individual’s immediate environment, creating the conditions for career mobility.

As individuals advance into management and leadership roles, performance becomes assumed rather than continuously evaluated. Mobility increasingly relies on the interactions between reputation, influence, and network. Each advancement strengthens reputation, expands network reach, and increases influence, which in turn creates additional opportunities for mobility.

Why Building Social Capital Can Be Difficult

If social capital governs mobility, then understanding why some otherwise capable professionals struggle to build it is important in identifying how to help high-potential individuals prepare for leadership roles. The following patterns emerged in my research.

1. The productivity trap. High-performing individuals whose social capital is built almost exclusively through performance may be operationally indispensable but strategically invisible. These professionals consistently deliver results, solve problems, and become highly valued within their immediate teams. Yet their reputation, influence, and network remain largely confined to the environment in which their work is performed, especially if they work on projects that don’t create touch points outside the immediate team.

As a result, they become indispensable to current operations but increasingly invisible to those selecting potential leaders for growth opportunities. Their contributions are recognized locally but rarely translated into visibility, advocacy, or mobility elsewhere in the organization. Over time, performance ceases to generate mobility and instead becomes the mechanism through which they remain anchored to their existing role.

2. Reputation without advocacy. Some individuals develop strong credibility and positive reputations but lack the relationships or influence necessary to convert that reputation into opportunity. These professionals are respected, viewed favorably by colleagues, and often considered capable of larger responsibilities. However, when advancement opportunities emerge, they lack relationships with people who could actively advocate for them.

In these situations, reputation becomes passive capital. Trust exists, but it isn’t mobilized. As a result, advancement frequently lags behind capability.

3. Network without credibility. The third pattern reflects the opposite imbalance: Some individuals successfully develop broad networks and visibility but lack the credibility necessary to take advantage of the opportunities those relationships create. Access outpaces demonstrated capability, and the individual’s visibility exceeds their readiness to perform at the required level.

While networks can create opportunity, long-term mobility still depends on trust and performance. Without credibility, relationships alone rarely generate sustained advancement and may even create reputational risk, when expectations exceed demonstrated capability. One participant observed that an employee’s network could be seen as a distraction if their “ducks aren’t in a row.”

These patterns suggest that career mobility is rarely constrained by a complete absence of social capital. Understanding these obstacles can help point to where development efforts should be focused and what may be preventing otherwise capable professionals from moving forward.

How to Develop for Social Capital, Not Just Performance

Excellent performance is typically what makes potential leaders stand out, but often, development efforts remain focused on building expert skills, leaving the broader mechanisms of mobility largely unaddressed. Social capital should be developed as intentionally as any other professional capability. Managers and sponsors are uniquely positioned to do this, but it requires them to offer different kinds of support than they may be accustomed to providing.

What Managers Can Do: Develop More Than Performance

Managers often focus development conversations on execution, productivity, and technical capability. While these factors remain essential, performance alone rarely prepares individuals for the complexity of future leadership roles. Managers should therefore view development through a broader lens, helping employees build the components of social capital while creating opportunities to accumulate trust, credibility, and relationships that support mobility.

They can begin by creating opportunities for employees to establish credibility beyond their immediate responsibilities. Cross-functional projects, enterprise initiatives, stakeholder-facing work, and opportunities to present ideas to broader audiences all help expand visibility and reputation. Managers can also help employees strengthen their influence by involving them in decision-making discussions, encouraging them to lead through collaboration, and providing them with opportunities to navigate competing stakeholder interests.

Equally important, managers should recognize when high performers are becoming trapped within the boundaries of their current role. Development should focus on helping performance travel beyond the environment in which it originated.

Perhaps most important, however, is helping employees recognize that these experiences create opportunities to accumulate social capital through repeated deposits. Completing a stretch assignment may create initial visibility, but sustained mobility often depends on what happens afterward. Maintaining relationships with stakeholders, staying connected with peers across functions, and continuing to engage with leaders from temporary assignments help reinforce trust, credibility, and familiarity over time. These deposits keep an individual’s name present in conversations long after a project has concluded and increase the likelihood that opportunities will emerge from relationships built through prior experiences rather than current performance alone.

Where Sponsors Can Help: Extend and Educate

Sponsors occupy a unique position because they can extend their own social capital on behalf of another individual. Like managers, sponsors can help individuals build social capital by creating opportunities for visibility, exposure, and relationship development. However, sponsors contribute something additional: Through advocacy, introductions, endorsements, and visibility, they lend their own social capital to create opportunities that may not have been accessible through performance alone, in a dynamic described by scholar Ronald S. Burt in his writing on social capital. However, with any loan, there is an expectation of repayment: The sponsor expects to benefit in some way that will enhance their reputation and influence. This reciprocity is key to a successful relationship between a sponsor and the individual they’re sponsoring.

But effective sponsorship extends beyond opening doors. Sponsors routinely make decisions based on trust, reciprocity, credibility, judgment, relationships, and reputation because they understand how mobility operates within organizations. However, these dynamics are often left implicit. Individuals may benefit from sponsorship without fully understanding why opportunities emerged, what reputational considerations were involved, or how advocacy was earned.

Effective sponsors make these mechanisms visible. They help individuals understand how social capital is accumulated through consistent deposits and built through the four components I outlined, and how it ultimately translates into opportunity. In doing so, sponsors move beyond providing access and begin teaching others how to build, accumulate, and benefit from social capital throughout their careers.

Performance remains essential for career success, but it is rarely sufficient on its own. As careers progress, advancement increasingly depends on whether performance is recognized, trusted, advocated for, and carried beyond the environment in which it is produced. Social capital provides the mechanism through which that occurs.

For professionals, the implication is clear: Social capital must be built, accumulated, and intentionally developed alongside performance. For managers and sponsors, the responsibility extends further. Developing talent requires more than improving execution; it requires helping individuals cultivate the trust, credibility, influence, and relationships that enable mobility. Ultimately, social capital determines whether performance stays where it originated or travels to where opportunity exists.

Algorithmic tools promise to democratize access to knowledge and thus spark creativity and innovation, but research we conducted revealed a hidden risk: Those tools may be silently narrowing organizations’ creative potential by suppressing the value of expertise. The fault lies not with the experts but with the hidden architecture of the tools they use. The invisible design choices embedded in algorithmic tools fundamentally shape creative output.

Standard algorithms behind search, discovery, recommendations, or large language models (LLMs) are designed around exploitation logic: They prioritize popular, relevant results, which reinforces what users already know instead of challenging them to explore. When such tools are designed for efficiency rather than exploration, they channel users toward conventional information, creating what we call ideation bubbles: clusters of similar ideas that represent a dangerous homogeneity of thought.

But our findings also revealed a solution: When we modified exploitation-based algorithms to surface diverse, uncommon information, experts who used them generated solutions that were significantly more creative, and they were able to break free from the convergent thinking patterns that can trap entire organizations.

Most digital tools we use today are designed to prioritize efficient access to popular answers. They also draw on a user’s existing knowledge frameworks (such as their search or chat histories) when presenting information and rarely challenge them to explore new territory. While this approach excels at delivering quick, useful results, our research found that this kind of bias is detrimental to creativity and innovation.

At the individual level, creators gravitate toward familiar solutions. The problem compounds at the organizational level. When multiple people use the same exploitation-based tools to brainstorm solutions to the same challenge, they are independently channeled toward the same information, and they independently generate similar ideas. As is the case with news bubbles, ideation bubbles are imperceptible to those inside them: Individuals believe that they are generating diverse ideas because they are working independently, but the shared algorithmic infrastructure steers everyone toward the same solution space. This convergence risk is especially dangerous for strategic challenges requiring breakthrough thinking.

Surfacing a Wider Spectrum of Ideas

To test our thinking about exploration versus exploitation, we designed an algorithmic modification we called XYZ that uses natural language processing. Built on top of Google Search, it surfaced results from semantically distinct clusters of ideas rather than the most popular or relevant matches — prioritizing exploration over exploitation. We then conducted two complementary studies: a controlled laboratory experiment with 104 participants, who were asked to generate creative ideas for reducing resource overconsumption; and a global field experiment, in which 245 participants, ranging from sustainability novices to seasoned sustainability experts, participated in an ideation challenge to reduce food waste in households. In both cases, we compared their creative output when using either Google Search or XYZ. The ideas were evaluated by independent expert judges blind to the experimental conditions.

In the laboratory study, ideas developed with XYZ were rated 14% more creative than those developed with standard Google Search, evidence that exploration can lift creativity even without deep domain knowledge. The more striking finding came in the field study: When using standard Google Search, domain experts showed no statistically significant advantage over novices at generating creative solutions. When using exploration-based algorithms, experts significantly outperformed novices, and ideas were rated 11% more creative on average. The algorithmic design, operating imperceptibly beneath the surface, shaped whether experts could make their strongest contributions.

What explains the gap between experts and novices? The key lies in what we call recombinant innovation: synthesizing diverse information elements into novel combinations. Exploitation-based algorithms surface familiar information that echoes existing mental models, but exploration-based algorithms expose users to insights from fields that may be unfamiliar to them, such as, in our study, behavioral economics or supply chain optimization. Experts have the knowledge foundation to harness this diversity effectively. A novice encountering the same diverse information lacks that scaffolding.

Experts can recognize which unfamiliar ideas are relevant, which are dead ends, and how to integrate them into a workable solution. For instance, one expert in our study combined insights about community food-sharing platforms with smart-home technology and behavior change techniques to propose a neighborhood-based “food rescue network” with automated inventory matching. That recombination required a deep understanding of food waste patterns.

Helping Domain Experts Break Out of Ideation Bubbles

While the effects of exploration-based algorithms are profound for individuals, the organizational implications may be even more significant. Using natural language processing to semantically cluster all ideas from our field study, we found two distinct effects and a critical interaction between them. First, exploration-based algorithms increased idea diversity for all participants. Novices using Google Search (exploitation) produced ideas that fell into just one semantic cluster; novices using XYZ (exploration) produced ideas spanning two clusters.

Second, expertise alone had a similarly modest effect: Experts using Google Search generated ideas across two clusters compared with novices’ one. But the interaction between expertise and exploration-based algorithms was dramatic: Experts who used XYZ generated ideas across five distinct clusters, compared with one or two from every other group. These experts did not just contribute more ideas within existing solution spaces; they generated entirely new ones, breaking dominant ideation bubbles and creating unconventional clusters of thinking.

These findings have immediate implications for how organizations structure their innovation processes.

First, they should treat algorithm type as a design input, not a default. Most tools that organizations use are optimized for efficiency, not exploration, which can suppress the value of what experts can contribute. When tackling strategic challenges that require breakthrough ideas, companies should consider approaches that surface diverse and uncommon information instead of the most popular or obviously relevant results.

Second, organizations should match the algorithm to the task. Exploitation-based algorithms remain valuable for accuracy and efficiency — identifying best practices or answering well-defined questions. Exploration-based approaches are best reserved for early-stage ideation, when divergent thinking is most valuable. This applies directly to how organizations use LLMs: Prompting for exploration rather than exploitation — for instance, asking for approaches that draw from unrelated industries or explicitly challenging dominant assumptions — can dramatically affect creative output. When reviewing AI-generated material, experts should particularly attend to unexpected or unfamiliar elements and avoid gravitating toward expected, recognizable patterns. Novel results are where the raw material for recombinant innovation often lies.

Third, organizations should audit their idea portfolios for ideation bubbles. If a team’s ideas cluster around a narrow set of solutions, the problem may be the tools. Semantic clustering can help identify bubble formation; deploying domain experts with exploration-based tools can be the best way to break them.

Fourth, they should invest in expertise. Our findings show that domain knowledge remains essential for innovation, particularly when paired with the right tools. The organizations best positioned to benefit from AI in innovation are those that develop expertise and configure their algorithmic tools to unlock it.

Looking Forward: The Expert Advantage

Proponents of AI argue that it reduces the cost of accessing expertise, but our research suggests that AI also does something more interesting: It transforms how expertise creates value. When algorithms democratize access to information, the premium shifts to those who can synthesize, recombine, and innovate with that information. Our research suggests that, far from being diminished by AI, expertise is transformed by it.

True breakthroughs depend on people having the ability to make unexpected connections — a skill at which domain experts excel, provided that they have the right tools. The question for leaders is not whether experts are needed but whether their organization’s tools are designed to let experts do what only they can do.

Many companies around the world have made significant progress in adding a higher proportion of board directors from traditionally underrepresented groups, such as women and people of color. But despite the increased diversity, many boards lag on inclusivity — that is, ensuring that diverse voices are actually heard and that all board members are able to contribute to decision-making.

In our interviews with more than 25 board and committee chairs and 20 nonchair board members, a contradiction emerged: Chairs overwhelmingly believed that they led inclusive boards, but many board members disagreed.

This pattern mirrors findings from a recent Egon Zehnder global survey of board directors.1 While most chairs reported that they create inclusive spaces, only half of board members said their boards were actually inclusive. Men were significantly more likely than women to agree that they “can bring their full selves to the boardroom.”

This disconnect between what chairs and directors perceive matters. A board can be diverse — which means that people with a variety of different perspectives, backgrounds, and demographic profiles are seated around the table — without being inclusive. Without explicit inclusive leadership from the chair, a board that is diverse can actually see its effectiveness reduced rather than enhanced. Research on groups and teams shows that when a wider range of perspectives enters a discussion without being accompanied by enabling conditions, communication breaks down, factions form, and decision-making slows.2

Boards today face complex strategic challenges, including geopolitical uncertainty, climate risk, disruptive technologies, stakeholder activism, and rapid shifts in consumer value systems. They need the cognitive diversity brought by different lived experiences, disciplines, and backgrounds. But to benefit from that diversity, chairs must create environments where all directors feel able to speak up — even when their perspectives challenge prevailing views.

Defining the Inclusive Board

In our interviews, we asked the question “What does an inclusive board mean and look like to you?” The answers we received can be distilled into the following definition: An inclusive board is one where every board member feels respected, valued, and empowered to contribute, regardless of background, identity, or expertise. An inclusive board fosters a culture in which diverse perspectives are actively solicited and thoughtfully considered, ensuring that all voices are heard. Importantly, an inclusive board does not mean that everyone’s input has equal weight in decision-making; rather, it ensures that decisions are enriched by the breadth of input and made transparently, with clarity around how and why certain viewpoints shape certain outcomes.

However, we discovered that many chairs unintentionally overlook certain behaviors that meaningfully shape their boardrooms. The unintentional part is really important: Every chairperson we talked to expressed their intention to create inclusive boardrooms. And the nonchair directors with whom we spoke emphasized that even in the least inclusive boardrooms in which they sat, there was never the feeling that the chair wanted to exclude certain people or voices. The chairs just weren’t aware of the implications of their actions or failure to act.

Based on our interviews, we identified five overlooked behaviors that any chair, regardless of personality, leadership style, or cultural context, can implement to foster a more inclusive board. These aren’t the obvious behaviors — such as “ensuring that everyone speaks once” or “asking good questions.” They are less visible, more structural, and more consequential behaviors that the board members we interviewed consistently identified as shaping inclusion. And these behaviors span before, during, and after meetings. When chairs use the behaviors intentionally, they shift boards from simply looking diverse to harnessing their diversity in the discussions they have and decisions they make.

Previous research has suggested that the inclusivity of boards is dependent on the chair’s traits.3 But we have witnessed inclusive (and non-inclusive) boards run by chairs who are introverted, extroverted, analytical, relational, hierarchical, consensus-seeking, and everything in between. Inclusion is not about who the chair is. It’s about what the chair does.

Five Ways Board Chairs Can Act More Inclusively

The board members we interviewed repeatedly highlighted a consistent set of behaviors that shaped their feelings of inclusion, but the board chairs overlooked, underestimated, or misunderstood those behaviors. Here are the key ways chairs can ensure more inclusive board meetings:

1. Use pre-meeting calls to understand perspectives, not to control them. What happens before the meeting itself matters. Pre-meeting calls with individual directors ahead of board meetings are de rigueur for many chairs. But the purpose and format of those calls vary widely — and matter greatly. Some chairs and board members cited pre-calls as detracting from inclusivity, while others cited them as contributing to inclusivity. Pre-calls during which the chair tries to align with certain members ahead of time, expresses their views on specific positions that they want the board to take, or discourages dissent on certain issues all detract from inclusivity. As one seasoned board member told us, “I am terrified to learn that board members voice their views in pre-calls versus in the boardroom and that they do not dare speak up [in the meeting].”

Pre-calls can better promote inclusivity when they are used to review the upcoming agenda to ensure that, based on the directors’ perspectives, nothing is missing. They may also be helpful in mapping the landscape of opinions and, in particular, to surface “quiet voices” and then encourage the more reticent to speak up in the actual meeting. Finally, they are also helpful in building relational trust, especially with newer members who might need to socialize certain ideas before putting them in front of their peers.

For such pre-calls to enhance inclusivity, board members need to be confident that the calls occur with all members, not a select few. If chairs contact only some directors and not others, they’ve created a second board within the board.

2. Frame agenda items for discussion, not presentation. How chairs introduce agenda items shapes who speaks, what gets surfaced, and whose voices carry weight. Board members described two typical approaches — one oriented to presentations, and the other to discussion.

In the former scenario, the chair walks through a topic, outlines key considerations, and only then asks for input. This structure tends to bias the group toward the presenter’s framing, limit dissenting voices, reinforce hierarchical dynamics, and reward directors already aligned with the dominant perspective.

A chair who is oriented toward discussion will introduce the topic, state why it matters, and open the floor to the others before offering a personal perspective. This invites more diverse viewpoints, encourages questioning and constructive disagreement, allows lower-status members (those with a non-CEO background) to speak without contradicting the chair directly, and signals that the decisions are not predetermined. While our interviewees noted that discussion-style meetings can feel messier, they consistently reported that they lead to better decisions and greater inclusion.

One former chair and current board member who sits on the boards of multiple publicly listed companies told us that when he chaired, ahead of the meeting, he circulated a one-page summary on each issue describing why it mattered to the company, and the CEO’s and/or executive team’s perspective on the topic. He then indicated whether it was a topic that was merely up for discussion (and, if so, why it was on the agenda) or whether the board would be making a formal motion on it. He also indicated the time that would be allotted for the agenda item so that directors would know how much discussion would be possible.

Chairs should open the conversational space before narrowing it, and delay their own viewpoints until the board’s perspectives have surfaced. They can also separate discussion from decision-making.

3. Demonstrate hearing, not just listening. All of the chairs we spoke with said that they not only invite everyone to speak up but also signal active listening through nodding, making eye contact, and so on. In fact, active listening was the most common inclusive behavior mentioned by chairs. Nonchair directors told us that these cues are insufficient. Instead, they look for deeper indicators that their contributions were heard, valued, and integrated.

One important way of doing this is to call on directors based on their expertise. Several directors shared stories about meetings in which their domain expertise (such as sustainability, cybersecurity, or AI) was directly relevant to the agenda — and yet the chair did not invite their perspective. At the same time, chairs should avoid confining directors to their perceived expertise. Some chairs believe that they are being efficient by calling on directors only when a topic aligns with their past functional role or domain of expertise, but board members said that this practice reinforces silos, discourages cross-disciplinary thinking, restrains growth and contribution, and sends subtle signals about who “belongs” where.

Likewise, chairs should signal that everyone has their full attention. The most exclusivity-building behavior that directors cited was chairs appearing to favor certain board members by giving them more time to speak, asking them more follow-up questions, making more eye contact with them or smiling at them, and making more frequent references to them (“As Jackson said …”).

4. Ensure that seating arrangements reflect equal distribution of power. Physical space, especially in a boardroom, communicates status and relationships. When it came to creating inclusive board environments, the chairs often had one thought about what led to inclusive seating while the directors had another. More than three-fourths of the chairs we talked to said that they explicitly think about where they sit relative to the other directors. Specifically, they said that they are careful not to sit at the head of the table — a position that denotes power and authority. But when we talked with board members, they mentioned that it was where others sat (relative to the chair) that made the difference.

Most board members preferred assigned seating over free seating. In the latter case, they told us, the higher-status members gravitate toward the chair, while newer members or those from underrepresented groups end up in more peripheral seats, which results in limited side conversations and static influence channels. Directors advocated for assigned seating that is regularly rotated. This disrupts power clusters by preventing the same individuals from forming an inner circle, distributes proximity to the chair, and enables varied informal interactions. New seating leads to new side conversations during breaks. One board member sat on a board where, at every break, the seats were rotated. She said that this provided a new dynamic that allowed for different conversations to emerge within the different seating constellations. However, if there are members of the board with hearing impairments, the chair should ensure that they are regularly seated in a central location so that they don’t miss out on the conversation.

5. Ensure that all members have access to the same information and people. Inclusivity extends beyond meetings. The board members we spoke with also highlighted the importance of equal access to internal stakeholders and information.

Some chairs grant members unfettered access to company executives and employees, whereas others require directors to route all interactions through them. Directors did not express a preference for one model over the other. What mattered most was consistency. Boards became less inclusive when some directors had informal access and others did not; certain members were quietly gatekept.

Directors also emphasized the importance of equitable access to information — particularly committee-level documents. Most chairs said they shared these documents broadly, beyond just the committees for which they were intended. But many nonchair directors reported that only half the boards on which they were serving actually consistently received committee documents. Without question, committee chairs need deeper knowledge and awareness of certain issues on their desks. And when chairs share the documents exclusively within committees, it creates information asymmetry, signals whose perspectives matter, reduces some members’ confidence to speak during discussions, and introduces a hierarchy of insiders versus outsiders. We also heard that inclusivity is enhanced when committee chairs are required to give a comprehensive update during the broader board meeting on what they discussed in the committee. The rule of 6:1 is a good one: Each 60-minute span spent on an item in committee should have a corresponding 10-minute update in the broader board meeting.

Two Additional Practices That Boost Inclusion

Beyond the five core behaviors mentioned above, chairs and directors described two additional practices — neither of which occur around meetings — that meaningfully strengthen inclusive boardrooms: explicitly measuring board inclusiveness and providing board members with opportunities to receive training to fill knowledge gaps.

There are a couple of ways that we have seen chairs effectively measure board inclusiveness. The first is to do it during annual conversations with directors, and the second is to have an outside party measure it through a survey or interviews.

Chairs typically hold yearly one-on-one conversations with directors. Few explicitly ask about inclusion, but those who do generate richer insights and deeper trust. Effective questions can include, but are not limited to, the following:

• Over the past year, when did you feel most included or heard in board discussions? Why?
• Were there times when you felt that your voice was overlooked or undervalued? What was it that made you feel like your voice wasn’t heard or valued?
• Have you ever hesitated to speak up on the board? When was that, and what kept you from speaking up?
• To what extent do you feel that your background and experience are recognized on the board? What is it about the board that leads you to feel this way?
• Do you possess experiences, skills, or perspectives that you feel the board has not tapped into?
• How connected do you feel with your fellow board members outside of formal board interactions? What might make you feel more connected?
• Are there any formal or informal board practices that you think unintentionally exclude some members?
• What support or changes from me, as the chair, would help you feel more engaged or valued on our board?

Such questions normalize conversations about inclusion without making them political or personal. They can also surface board members who might feel less included. And they signal that the chair explicitly cares about the topic.

Another option is to have an outsider, such as a board assessment consultant or academic expert, conduct surveys and interviews. For one board, we surveyed the directors about specific inclusion behaviors, including balanced participation and evidence of listening. At the start of each subsequent meeting, the chair reviewed the aggregated results with the full board and invited discussion of any patterns that required attention. Over time, this consistent feedback loop helped members adjust their own behaviors, reduced dominance effects, encouraged quieter voices to contribute, and fostered a shared sense of accountability for how the group listened. As these listening norms improved, the board became more forward-looking and able to engage in strategic initiatives that strengthened the organization’s prospects.

Helping members close expertise gaps is essential as boards increasingly face complex topics — AI, sustainability, cyber risk, geopolitics — on which only a few directors might feel confident contributing. Some boards even provide their directors with an education budget. One director described staying silent during AI discussions because she felt unqualified and uninformed. Recognizing this, her chair sponsored optional AI training for all directors. Afterward, more directors felt empowered to contribute to these discussions. By investing in learning, the chair democratized influence. This expertise gap can also apply to knowledge about the company. Providing newer board members with onboarding briefings about the organization, its history, and its current strategy can bring them up to speed and allow them to feel more confident to speak during meetings.

Traditional assumptions about strategy, governance, technological readiness, and stakeholder expectations are shifting rapidly. The boards most likely to navigate these shifts successfully will be those that fully capitalize on the diversity of thought sitting around their tables. But that cannot happen without intentional inclusive leadership. We repeatedly heard that inclusion is not an outcome of good intentions; universally, the chairs we spoke to and heard about had good intentions. Inclusion is the output of consistent, conscious behaviors. The encouraging news from our research is that the intention is there and all behaviors that we cited above are accessible, learnable, and controllable. And incorporating those behaviors is unmistakably the chair’s job.

A recent battle for AI talent illustrates how difficult it is for companies to retain key employees. In 2025, OpenAI’s stock-based compensation averaged roughly $1.5 million per employee — unprecedented for a pre-IPO company — yet OpenAI still experienced high-profile defections. Rivals such as Meta were reportedly extending offers in the hundreds of millions of dollars for top AI talent, prompting OpenAI to issue multimillion-dollar one-time retention bonuses and to relax equity vesting requirements twice in a single year. The result was an escalating bidding war with no clear ceiling and no guarantee of success.

That dynamic has not ceased. Rather, the competition for talent has intensified and expanded, with AI companies recruiting not only elite researchers but also senior executives and business leaders from across the technology sector.

The AI talent wars are an extreme case, but they illustrate a broader problem across industries: The most valuable employees are also the most mobile. In sectors such as artificial intelligence and consulting, intense competition reflects rapid technological change that makes certain capabilities suddenly scarce. In health care and skilled trades, retention pressures stem from long-standing workforce shortages. Construction companies are facing a growing labor gap driven in part by retirements, even as demand is rising with the expansion of large-scale infrastructure and data center projects. Across all of these contexts, when an employee leaves, their employer loses embedded knowledge, client trust, and innovation capacity along with them.

Yet most companies respond to this challenge in a tactical, episodic fashion: a counteroffer here, a culture initiative there, or a noncompete added to a contract. These one-off measures address symptoms rather than causes, and, as OpenAI’s experience shows, pay-based competition alone can trap companies in a race they cannot win.

Traditional approaches to understanding the problem focus on factors that make employees move, but what is missing is a systematic framework for understanding why they stay.1 Our recent article in the Journal of Management investigates the factors that limit employee mobility away from an employer — what we call employee mobility barriers.2 Drawing on research from multiple disciplines, we have synthesized these barriers into a practical architecture that managers can use to retain talent more deliberately — and compete for new hires more effectively — in today’s talent wars.

Understanding Employee Mobility Barriers

Whenever an employee considers leaving, they navigate a set of frictions that make moving harder and staying more attractive. These barriers range from the concrete — noncompete agreements, unvested stock, pension accrual — to the more intangible, such as the satisfaction of meaningful work, the promise of career advancement, and social ties that make teamwork productive and fulfilling. Together, such forces shape whether a move feels possible and worthwhile or not worth the disruption.

Barriers differ along two dimensions: level and control. Some barriers operate at the individual level, such as an employee’s career stage, personal preferences, or priorities. Others stem from organizational systems and practices, including social ties, employment contracts, or career development systems. Still others are rooted in broader societal factors, such as visa restrictions or labor market conditions.

Control is the degree to which employers can influence the mobility barrier and employee experiences. Employers can directly shape factors such as compensation, job design, and employment contracts, but they have less influence over others. For instance, employee attributes such as age or personality can be “shaped” only at the point of selection, while social ties and networks emerge over time and are difficult for a company to directly engineer. Other barriers, such as location preferences, regional labor-market conditions, or visa and licensing requirements, are largely outside an employer’s control.

Perceptions also play an important role in how mobility barriers function. How employees interpret such barriers they face and the opportunities available to them elsewhere can influence their decision to stay or leave. For example, an employee who believes that they have better career development opportunities with their current employer than with a competitor may choose to stay, regardless of whether that belief is accurate.

Employee Mobility Barriers in Practice

In practice, employee mobility barriers and companies’ effectiveness at retaining talent can vary across industries, companies, roles, and individuals. What keeps an AI engineer from moving may differ from what retains an industrial salesperson or a veteran nurse. Stock options are a powerful retention mechanism in the tech industry but largely irrelevant in the public sector or education. An entry-level employee building their initial professional network and reputation may experience very different barriers than a more senior employee with strong ties to the organization. Managers need to understand which barriers matter most in their specific contexts and for their employees so that their retention interventions are effective.

This context dependence also informs a crucial strategic principle: The best mobility barriers are those that are difficult for competitors to replicate. Compensation alone is a weak barrier in this regard because it can trap companies into bidding wars that better-resourced rivals can win, as OpenAI’s experience demonstrates. In contrast, Anthropic, founded by former OpenAI engineers, has emphasized employee autonomy and AI safety and has achieved higher retention rates than its rivals as a result. Similarly, Periodic Labs lured more than 20 engineers away from Meta, OpenAI, and Google DeepMind, despite offering lower pay, by promising a distinctive work environment focused on scientific discovery — closer in spirit to traditional research labs than the typical Silicon Valley firm is. Culture, mission, and working conditions can be more durable barriers precisely because they are embedded in how people work and interact — and thus far harder for competitors to imitate or re-create.

Understanding employee mobility barriers is also valuable for attracting new hires. An effective job offer should consider the barriers a candidate faces in leaving their current role. Sometimes that means offering what a candidate would be giving up. That could mean convincing a candidate that future colleagues and work conditions will be as good as what they are leaving behind, for example. Other times, it means offering something that the current employer cannot match, such as a path for career growth or a unique project. Thinking about hiring through the lens of mobility barriers turns recruiting from a pitch into a more intentional, targeted conversation.

An Architecture for Managing Employee Mobility Barriers

The varying levels and degrees of managerial control of employee mobility barriers make developing a coherent, companywide talent management strategy challenging.

To help leaders navigate this complexity, we developed an architecture of strategic modes that characterizes how companies manage employee mobility barriers, based on the two dimensions introduced above: the level of the barrier (individual, organizational, or societal) and the degree of control the employer has over it (high or low). (See “Architecture of Strategic Modes for Managing Employee Mobility Barriers.”) While firms may rely on or encounter a portfolio of different mobility barriers, the architecture of strategic modes focuses on how those barriers are managed.

Together, these dimensions define four strategic modes: For any given mobility barrier, its level and the degree of employer control determine how companies should manage it — either by directly shaping the barrier (proactive) or by adjusting other organizational practices (reactive); and either leaving it to individual managers to design and implement (delegated) or rolling out organizationwide policies (centralized).

When mobility barriers operate at the individual level and are outside the employer’s control —such as personal ties to colleagues or family circumstances — the appropriate response will be reactive and delegated (upper-left quadrant). The right move here is to empower managers to recognize these situations and respond with accommodations where appropriate, such as making schedule adjustments or redefining a role to better fit an employee’s circumstances or needs.

When mobility barriers operate at the individual level but are within the company’s control, the employer can take a proactive and delegated approach (upper-right quadrant). In these cases, managers actively design roles and work experiences — through job design, meaningful project assignments, or customized development opportunities, for example. In this quadrant, managers have the discretion to shape these elements based on the needs of their teams.

In contrast, when mobility barriers operate at the organizational or societal level and are outside the company’s control, the employer must respond in a reactive and centralized manner (lower-left quadrant). In these situations, businesses cannot directly control the focal barrier. Instead, they need to act indirectly by reconfiguring how work is organized or by introducing complementary practices that allow them to operate effectively under the constraints. Accommodating existing mobility constraints may involve reallocating work across teams, restructuring collaboration patterns, or redesigning how tasks are distributed.

Finally, when mobility barriers operate at the organizational level and are within the company’s control, the employer can take a proactive and centralized approach (lower-right quadrant). This mode captures the types of organizationwide programs and policies that companies typically associate with retention efforts. However, although these system-level interventions are important, they represent only one part of a broader set of strategies that organizations must use to manage mobility barriers effectively. No company manages in only one quadrant. Effective management layers responses across all four modes, combining tailored individual-specific interventions with companywide systems and applying different approaches to different types of mobility barriers.

Managing Employee Mobility: A Toolkit for Managers

Understanding the architecture of employee mobility barriers can inform how employers manage them. To develop a systematic approach, managers should take these three steps:

1. Inventory employee mobility barriers. Effectively managing employee mobility is not about implementing a single policy, program, or perk. Rather, it requires a systematic approach that combines multiple mobility barriers and management strategies. As a starting point, managers should develop an inventory of the barriers that matter for current employees and for the talent that the company wants to attract. This inventory must be customized to different functional areas, the needs of managers responsible for specific teams, and, in some cases, strategically important employees. It should also be reviewed and updated over time as business conditions, labor markets, and employee priorities change.

2. Evaluate for importance and impact. Once an inventory has been done, managers should assess it along two lines. First, they should determine which mobility barriers matter most in their specific context; not all barriers carry the same weight in shaping employee retention or attraction. For example, stock-based incentives may be highly effective for senior technical talent, whereas embedded social networks may matter more for midlevel managers.

Second, managers should identify gaps in the current mix of barriers and consider how different barriers reinforce or substitute for one another. By understanding where the most significant barriers lie, companies can focus their resources more effectively, avoiding both over- and underinvestment while identifying opportunities to strengthen complementary barriers. Like the inventory, this assessment should be treated as an ongoing exercise that evolves with business conditions, labor markets, and employee priorities — not as a one-time audit.

3. Match management strategies to mobility barriers. With a clear understanding of which barriers matter most, managers can determine how best to respond to them. The appropriate approach depends on which aspects of mobility the company can directly shape and which it cannot.

When companies can control mobility barriers, they can often deploy them strategically through organizationwide systems and policies. For example, companies often design structured career pathways, job-rotation programs, or stock-based incentives to retain employees. Salesforce’s AI-powered internal mobility platform, Career Connect, illustrates this approach: By providing personalized skills recommendations, highlighting internal career pathways, and suggesting upskilling opportunities, it helps employees find growth opportunities within the organization so they will not seek them elsewhere. Such a proactive, companywide initiative can help organizations retain employees by supporting internal career advancement.

Organizations can also proactively deploy mobility barriers in response to changing conditions. When the Federal Aviation Administration faced critical air traffic controller shortages in 2025, driven by retirements and its inability to hire and train replacements quickly enough, it responded with systemwide retention bonuses aimed at keeping experienced controllers from leaving. Although prompted by external pressures, that response reflects the deliberate use of centrally managed barriers organizations can proactively deploy.

In other cases, mobility barriers arise from conditions an organization cannot directly influence, particularly at the individual level. Here, effective management depends on managerial discretion and flexibility. A delegated or decentralized approach works in such cases because the managers closest to employees have the best understanding of their needs and can design effective retention strategies if empowered to do so. Such delegated actions may involve designing customized job roles, offering flexible schedules, or providing targeted professional development opportunities. For example, when Goldman Sachs’s international vice chairman, Richard Gnodde, decided to relocate from the U.K. to Milan to take advantage of Italy’s tax policy, his role was adapted so he could continue working for the company from a different location. This type of intervention illustrates how organizations can respond to mobility barriers outside their control by adapting the employment relationship at the individual level.

Taken together, these examples highlight a broader pattern: Centralized approaches are most effective when companies can shape organizationwide mobility barriers directly, while delegated approaches are better suited to barriers rooted in individual circumstances. In practice, barriers often operate at both the individual and organizational levels simultaneously, so an integrated approach to managing them is called for. For example, businesses may combine formal, organizationwide initiatives with locally driven efforts that managers and employees help shape. At financial services provider Synchrony, for instance, employees have launched many of the firm’s volunteer and community programs, illustrating how decentralized initiatives can strengthen social ties and organizational identification. Such efforts reinforce employees’ connection to the employer while they operate alongside more structured organizational practices, enhancing mobility barriers that cannot be fully designed from the top down.

Effective management in these cases requires active coordination between centralized programs and local managerial discretion. That may require significant training and communication, along with a shared understanding among managers at different levels of hierarchy and across the organization. Realistically, organizations may need to be highly selective about the set of barriers for which such intensive coordination is used. By combining central control with delegated discretion, organizations can programmatically deploy mobility barriers while remaining flexible to individual employee needs.

The intensifying AI talent wars highlight how challenging it has become for companies to retain and attract mobile employees. As competition for highly skilled workers grows, technological change is making valuable skills more portable, while remote and hybrid work are weakening some of the geographic and social barriers that once kept employees in place. At the same time, advances in AI are reshaping which forms of knowledge will remain company-specific: AI agents are beginning to take on tasks previously performed by humans, altering sources of retention advantage in ways that are still unfolding.

Against this backdrop, the underlying logic of employee mobility barriers becomes even more important. Rather than relying on any single retention tool, organizations must understand the full range of barriers shaping employee decisions to stay and match their strategies accordingly. What varies across organizations is not whether mobility barriers matter but which ones matter most and how effectively they are managed. This holds true regardless of how the specific barriers evolve. The managers best positioned to navigate these shifts will treat talent strategy as a dynamic system that requires them to continuously identify relevant barriers, evaluate their impact, and adjust responses over time. Organizations that adopt this approach will be better equipped not only to retain critical talent but also to adapt their retention strategies as conditions evolve.

Every organization, and every hiring manager, wants to know how to hire the best people. And every company wants to feel like it has an edge over its competitors. But let me ask you a personal question: Can you think back to a moment when someone in a decision-making position took a chance on you? Perhaps you were young and inexperienced, and maybe you didn’t have a flawless resume or a spotless track record, but they saw something worth investing in. You almost certainly rose to the occasion. You probably gave more to that person or organization than your experience or credentials alone would have suggested.

Why are so few leaders making that kind of bet today? And what are organizations losing as a result?

I have spent more than two decades studying how organizations recruit, promote, reward, and retain employees. In doing so, I have discovered that a company’s willingness to grant opportunity shapes employment interest and recruitment, job engagement, effort, and retention — far more than compensation or job design do. Opportunity can unlock human motivation and potential, which, I would argue, are among the most underpriced assets on your balance sheet.

Most leaders, however, reserve their biggest bets and opportunities for people they already perceive as “the best.” They want the candidate who has the most impressive credentials and checks every box on the job description. This instinct feels prudent, as if leaders are making a safe bet. In practice, it is one of the most expensive talent-management mistakes any organization can make.

Consider two problems with the instinct to hire or promote only the best. First, contrary to popular wisdom, nobody truly knows who the best talent is. My research shows that people rarely agree on what qualities or achievements matter for success in a given role. Two hiring managers in the same company, evaluating the same candidate, can reach wildly different conclusions about their fit, potential, and leadership qualities. Every company likes to believe it has a rigorous, objective talent-selection process. But in practice, decisions frequently rely more on subjective judgments and gut feelings. Many managers also default to conventional, easily defensible choices. After all, no one ever got fired for hiring the guy with a prestigious university credential.

The second problem is a market dynamic. The fixation on the very best pushes organizations into what I call an only the best mindset, which is the belief that a scarce, identifiable elite is out there and that success depends on finding them, hiring them, and paying them extraordinary salaries.

There have been dramatic examples of this recently, such as when Meta CEO Mark Zuckerberg led an aggressive drive to recruit top AI talent by offering compensation packages of up to $300 million over four years. But the "only the best" mindset produces a predictable set of outcomes: It often takes companies longer to fill positions, compensation bands become eye-watering, and talent pipelines narrow to the same small set of schools and employers.

This mindset also creates talent risks you may be underestimating. Star performers brought in from other organizations often perform worse once they move, as Boris Groysberg, Ashish Nanda, and Nitin Nohria famously highlighted. You might be paying a premium for a signal that fails to travel.

Another Way: Opportunity Mindset

There is an alternative to this chimerical pursuit of the best. I call it the opportunity mindset, and I want to be precise about what that means. It does not mean lowering the bar and settling for second-rate talent. It’s recognizing that identifying candidates who look great on paper is often entirely different from identifying those who would be good fits for the role. Recruiters and hiring managers who recognize this fact have a huge competitive advantage over those who do not.

Leaders with an opportunity mindset hire candidates who are good enough to succeed and then build the structural conditions — through excellent onboarding, mentorship, development, and evaluation processes — that help good people become great. These leaders evaluate potential by what a person actually does on the job, not by schools attended, past employers, or previous job titles.

In practice, here’s what the shift to an opportunity mindset looks like. Say your engineering team has an open entry-level role. The traditional process would typically narrow the pipeline down to candidates with a four-year computer science degree from a highly ranked university. A company with an opportunity mindset instead asks what key skills the job actually requires and then opens the candidate pool to include nontraditional applicants, such as community college graduates, career switchers, parents returning to work, or apprentices. Some applicants might not have a college degree, but if they can demonstrate the necessary skills, they are considered for the role.

Alternatively, imagine that a more senior person is needed. Rather than conducting an expensive external search, your organization posts the promotion opportunity internally and invites applications from anyone whose skills match the requirements. The role might not go to the employee whose face is most familiar to senior leadership but to the employee whose current work trajectory best predicts success.

Each of these choices is small. But, over time, they compound, deepening the talent pool and creating a more versatile and diverse workforce.

This tension between the "only the best" mindset and the opportunity mindset is at the heart of my book The Meritocracy Paradox. In it, I show how talent management systems that claim to reward top talent and hard work — the so-called meritocratic systems — can unintentionally reinforce bias and inefficiency. When hiring managers believe that their decisions are driven by purely objective metrics or that they operate in meritocratic organizations, they may fall prey to subconscious biases, such as perceiving the best as people who are demographically or culturally similar to themselves. In contrast, when organizations embrace an opportunity mindset, a key feature of truly meritocratic organizations, they shift from chasing after talent perfection to investing in potential.

Some organizations have already moved in this direction. For example, years ago, under now former CEO Ginni Rometty, IBM concluded that the traditional four-year-degree requirement was screening out capable workers and driving up the cost of talent without materially improving performance. IBM rebuilt its hiring practices to emphasize skills, launched apprenticeship programs to recruit from high schools and community colleges, and increased professional development opportunities for existing employees.

Bank of America implemented a skills-first approach and reported in 2022 that internal promotions accounted for nearly 45% of filled roles. This practice ensures that institutional knowledge stays put. Google, meanwhile, treats its own professional certificates as equivalent to four-year degrees for many entry-level roles. Both organizations have done the math and concluded that the only the best pipeline is too narrow, slow, and expensive to defend. Researchers have found similar gains at companies like Costco, Trader Joe’s, Mercadona (in Spain), and QuikTrip, where investment in employee skills development positively correlates with lower turnover, higher productivity, and stronger financial performance.

What About the Hiring Recession?

I anticipate an objection at this point, and it is a reasonable one. Early examples of skills-first hiring emerged from a different labor market era, in which workers were scarce and employers were desperate to widen the hiring funnel. The market has since changed. The white-collar job market has softened for job hunters, wage growth has slowed, and employees are holding on to their jobs instead of searching. College graduates, even those with desirable degrees from prestigious schools, are reporting lengthy and challenging job searches. In other words, we’re in the midst of a hiring recession, and the leverage has shifted to employers.

If you’re a hiring executive, you might think that you can now afford to be selective in a way you couldn’t have been just a few years ago. So why, you might reasonably ask, should you adopt an opportunity mindset now, when the case for holding out for the perfect resume has never been stronger?

I would respond that the case for an opportunity mindset has never been about scarcity. It has always been about return on investment. First, a hiring recession gives hiring managers more applicants, but not better judgment about which ones will succeed. If anything, a flood of AI-polished resumes makes credentialed signals noisier: Robert Half, for instance, reported that two-thirds of HR leaders now say that AI-generated applications are slowing their hiring, and a majority of hiring managers find AI-enhanced resumes harder to verify and trust. Grade inflation has added to the confusion, frustrating those employers that want to use GPA as a distinguishing factor for potential hires. When the paper trail can no longer be trusted, structured skills assessments and paid work trials become more valuable, not less.

Second, a soft hiring market is exactly when bets on unconventional candidates are cheapest. You are not bidding against 10 competitors for the same star employee; you are choosing from a deeper pool at lower cost. The capable career-switcher who would have been snapped up elsewhere in 2022 is now in your funnel.

Third, and most importantly, the current hiring recession will eventually come to an end. Demographic changes, such as an aging workforce and lower immigration, will shrink the labor force regardless of the current economic cycle. The organizations that can take a long-term approach, anticipate their future talent needs, and begin building that capacity from within, while they can, will benefit enormously when the market turns. Those that do not will have to find that capacity under pressure, at premium cost.

Let me close with a practical challenge for you and your organization. When making your next hiring decision, try one small experiment. Identify a small set of necessary (and specifically job-relevant) skills that predict success on the job. Evaluate every finalist using the same consistent assessment, giving everyone, from recent graduates to Generation Xers, the same shot. Then, once someone has been hired, commit to a specific development plan (if necessary) for their first year. Carefully track what happens to their performance, their retention, their growth, and the performance of the team around them. If the experiment fails, you have likely lost little. If it works, you will have found something your competitors are still paying a premium to chase: a way to turn good candidates into great employees.

This is the essence of the opportunity mindset: building a more resilient, more inclusive, and future-oriented organization. Remember the bet that someone once made on you. Recall how the returns compounded. The leaders who learn to make similar bets on others, systematically and thoughtfully, especially now, will be the ones who build the enduring organizations of the future.

We live on a blue planet where water appears to be plentiful. But there are strong warning signs indicating that our reliance on the fresh water we perceive to be abundant will need to change. Although water covers 70% of the Earth’s surface, only 0.5% of it is effectively usable. Further, global water cycles are increasingly becoming unbalanced in ways that threaten humans, plants, and other life forms that depend on fresh water.

Rapidly growing demands for water due to population growth and industrial uses, as well as issues such as misuse, pollution, and the interconnected effects of climate change, are contributing to a situation in which companies and communities can no longer take the availability of water for granted. As competition for water intensifies, the consequences could be severe for health and well-being, economic development, societal stability, and security — that is, unless we significantly rethink how all of us, including companies, manage it.

Growing Water Risks and Challenges

Many companies treat water as a utility controlled by operations or plant managers. They calculate their demand for it based on their business models, process designs, and product characteristics but rarely manage it as a key factor shaping economic competitiveness and strategic decision-making. In many cases, water is seen as a hygiene factor: essential, but not critical enough to register as a risk. Accordingly, procurement teams seek to ensure the security of the water supply, and costs are kept low, while public affairs teams handle any regulatory questions. But this widely established approach is now coming under significant pressure amid consensus that water demand and related challenges are intensifying globally.

In a 2024 report, the Global Commission on the Economics of Water warned of a growing water disaster: “We can no longer count on freshwater availability for our collective future.” By the end of 2025, 35.8% of the U.S. was experiencing a drought, with conditions touching nearly every region. Impacts included catastrophic wildfires in Southern California, historic low water levels for the Mississippi River, and record‑low streamflow in the Northeast. Worldwide, roughly 30% of global land area experienced drought conditions in 2025, driven by near‑record warming.

Drought is one of many current problems, such as poor water quality due to pollution from farm runoff, pharmaceutical residues, and the proliferation of nanoparticles in water. All of those issues create new operational and regulatory demands and challenges. For example, France’s ban on PFAS, which came into force in June 2026, requires that drinking water be tested for those synthetic “forever chemicals,” which are estimated to affect the water supplies of some 12.5 million people across Europe. The law also penalizes companies for releasing such chemicals into the environment.

All of those situations raise the stakes for how companies manage their use of water, because the exposure to water risks is widespread. Consider fast-growing companies and sectors that significantly depend on water for their operational processes. These notably include AI data centers, which require substantial amounts of water for cooling purposes; growing public opposition to these operations is driven in part by their threat to the sustainability of local water supplies. The emerging green hydrogen sector, which uses water rather than fossil fuels as the primary raw material for hydrogen production, faces pushback, given the process’s high demand for water. The technology industry’s increasing demand for critical minerals, such as those used in rechargeable batteries, also puts pressure on water, which is used in mining operations worldwide. Traditional sectors, such as the food and beverage industry, account for 20% of water withdrawals. Textile production continues to consume — and contaminate — large amounts of water: It is responsible for 20% of industrial water pollution worldwide.

However, it’s not just large industries that face water risk. It also plays a nontrivial role in many small businesses, including car wash facilities, breweries, and restaurants; shortages can also affect the construction, health care, sports, and tourism sectors. What this means is that executives can no longer ignore water issues.

The Complexities of Managing Water

Executives tuned in to these issues have begun to appreciate the complex relationships of water flows and cycles and are taking greater interest in and ownership of the resources across their production and supply chain processes. Managing water and making its use more efficient requires the creation of multiple new KPIs everywhere the company uses water. Managers must monitor multiple indicators for water stewardship, such as total water withdrawal, the water recycle/reuse rate, wastewater treatment improvements, and watershed restoration impact, to name a few. They must account for both the inflows and outflows.

However, many companies have a worrisome blind spot: They typically view water as an organizational concern, potentially ignoring extensive dependencies in their supply chains. For example, companies dependent on agricultural inputs can’t ignore the fact that this sector is responsible for almost 70% of global withdrawals of fresh water. As one manager in the agri-food sector we interviewed put it, “If you are not monitoring water in your supply chain, what are you doing? In my sector, droughts and water contamination both happen in my supply chain.”

Another factor adding to the complexity is that water supplies are essentially local, held in basins and aquifers, and so water scarcity creates localized (rather than general) risks. For companies with large and extended supply chains, this means that one-size-fits-all approaches and policies are unlikely to be sufficient or effective. For instance, a company with three sites in the U.S. Northeast could be affected by sewage contamination while its two sites in the Southwest face severe drought conditions. Uniform corporate policies and standards provide insufficient guidance here. Instead, companies need to empower all site managers to complete a thorough water risk analysis, identify local stakeholders they can work with, and develop action plans to tackle such localized risks.

Strategic Water Management Responses

So how should companies respond to the changing realities of this critical resource? As with so many similar issues, understanding the risks and dependencies is key. There are freely available resources that can provide businesses with detailed guidance and support, including the International Water Stewardship Standard, the Science Based Targets Network’s Step Up for Nature initiative, the U.N. Global Compact’s CEO Water Mandate, the World Resources Institute’s Aqueduct tools, and the WWF Water Risk Filter.

Companies must shift from treating water as a utility to elevating it to a strategic priority. Doing this well may require that they establish cross-functional teams to obtain all relevant managerial perspectives and tap into diverse expertise. Crucially, they must engage with key stakeholders, including suppliers, regulatory authorities, nonprofits, and academia, in order to get the full and evolving picture.

Executives should familiarize themselves with the key issues and the specialist language used to discuss matters such as water abstraction and withdrawal, water basins, water stress levels, and so on. At a minimum, they need to assess the direct and indirect water footprints involved in running their businesses before gauging the extent to which this dependence is threatened by emerging water trends.

Operationally, reducing water consumption through greater efficiencies, recycling, and business model transformation is essential but requires investment and technological know-how. Dedicated supplier engagement and/or diversification will be critical if risks cannot be mitigated. Where possible, replenishing or balancing water at source levels must be considered as an option for increased legitimacy and resilience. This requires companies to actively engage or invest in efforts to help restore aquifers or other water reservoirs to effectively balance out the amount of freshwater withdrawn.

In short, there are no simple answers, but ignorance is likely to create growing business risks. Managers must therefore approach water as a valuable resource to be preserved, not drained.

Most leadership advice is written for beginnings — how to launch initiatives, build momentum, scale ideas, and manage new projects. Far less attention is given to the work of ending something well.

Yet endings are everywhere in organizational life. A team is disbanded. A project is closed. A product is retired. A partnership concludes. A strategy is abandoned. A publication, department, or institution reaches its final chapter.

These moments are often treated as administrative events: Announce the decision, set the dates, manage the handover, move on. But endings are cultural tests as well as operational transitions. People remember how something ended because endings reveal what leaders truly value. Did they tell the truth? Did they honor the work? Did they protect people’s dignity? Did they preserve what mattered? Or did they rush to the next chapter so quickly that the previous one felt erased?

Ending well does not mean making everyone feel good about the decision. Some endings are painful or contested. But leaders can still shape how people experience them. The following practices can help.

Say Clearly What Is Ending

First, use truthful language. In difficult moments, leaders typically soften changes with words and phrases like “transition,” “realignment,” “sunsetting,” or “moving in a new direction.” Sometimes these words are accurate, but more often they are ways to avoid saying the harder sentence: “This is ending.”

People cannot process an ending if leaders refuse to name it. Say plainly what will stop, when it will stop, and what will change as a result. For example: “This team will close at the end of the quarter.” “This product will no longer be supported after June.” “This partnership will not continue next year.”

Plain language cannot remove emotion, but it reduces confusion. It also signals respect. Adults can usually handle difficult news. What damages trust is the feeling that leaders are trying to manage employees’ reactions rather than telling them the truth.

Explain Without Overexplaining

People need to understand why an ending is happening, but they do not need a defensive essay. Overexplaining can make leaders sound as if they are trying to win an argument that has already been decided.

A better structure: Lay out the decision, reason, constraint, and consequence. “We are closing this program. Participation has fallen for three years, the cost base is no longer sustainable, and continuing it would require cutting investments from higher-impact work. That means the final cohort will finish in September, and we will support affected staff through the transition.”

This kind of explanation does not ask people to like the decision. It gives them enough context to understand the logic and sufficient specifics to understand what happens next.

Honor What Was, Before Discussing What’s Next

Leaders often move too quickly from closure to future plans. They want to reassure people that there is a path forward — an understandable instinct. But premature future-talk can feel like erasure.

Before asking people to move on, name what the work made possible. What did the team build? Who benefited? What standards did it set? What relationships did it create? What should people be proud of?

The answers do not have to be sentimental, but they should be specific. “This team reduced response times by 40%.” “This project changed how we serve customers.” “This publication gave managers a place to think seriously about practice.” Specific recognition tells people that the work mattered, even if it will not continue.

Explain What Is and Isn’t Known

During an ending, ambiguity becomes emotional labor. People fill gaps with speculation: What happens to my role? Who owns the remaining work? What should I tell clients? What happens to the archive? Who makes final decisions?

Leaders should answer practical questions as early as possible, even if some answers are incomplete. A useful checklist includes: timeline, responsibilities, decision rights, stakeholder communication, support available, work to be paused, work to be completed, and what will happen after closure.

Where certainty is not yet possible, say so clearly: “We do not yet know X. We expect to know by Y date. Until then, Z remains the operating assumption.” Stating what is and is not known is better than silence because it offers people a sense of certainty.

Let People Have Mixed Feelings

Endings rarely produce one emotion, even within one individual. People may feel proud, angry, relieved, disappointed, loyal, anxious, and exhausted all at the same time. Leaders should resist the urge to impose a single emotional narrative.

Do not demand positivity. Do not tell people to be excited about the next chapter before they have absorbed the end of this one. Do not interpret sadness as resistance or questions as disloyalty.

A better message is: “People will feel differently about this, and those reactions may change over time. We can acknowledge that honestly while still doing the work needed to close this down well.” This type of statement preserves a necessary sense of responsibility for people’s work while giving them permission to be human and feel emotion.

Preserve What Should Last

When something ends, its impact can vanish quickly. Files are archived, routines stop, institutional memory fades as people leave. Leaders should decide deliberately what needs to be preserved. This might include a final report, a lessons-learned document, a customer handover, a public archive, a closing note, a celebration of contributors, or a record of practices that should continue elsewhere.

The point is not to preserve everything. It is to protect what has lasting value. Ask three questions: What did we learn? What should others inherit? What should not have to be rediscovered later? These questions transform closure into stewardship.

End With Dignity, Not Just Efficiency

A dignified ending gives people enough truth to understand, enough structure to act, and enough recognition to feel that their work counted for something. An efficient ending is not necessarily dignified.

The final meeting, message, or milestone matters. It should not be treated as a formality. Leaders should use the opportunity to specifically thank people, to mark the transition clearly, and to communicate what will be carried forward. People do not need a grand ritual, but they do need a meaningful one. The goal is not to make the ending painless — which would be impossible — but to make it conclusive.

Endings ask whether leaders can be honest without being cold, compassionate without being vague, and forward-looking without erasing the past. When leaders end something well, they give their people clarity and closure: evidence that the work mattered, that the truth can be spoken, and that dignity does not have to disappear at the end.

Innovation leaders often know how to manage success well. A fairly good idea is recognized, a strong contributor is rewarded, and the lessons from the win are carried into the next team project. In many cases, that is exactly the right organizational response.
 
But, sooner or later, most organizations have a different kind of success altogether — a next-level idea and success tied to one person: the engineer behind a product that changes the trajectory of a business unit; the scientist whose patent is embedded across product lines; the designer whose solution unlocks a long-stalled project. After a breakthrough like this, CEOs and managers often respond in the same way they would to any success, but with the volume turned up: more reward, more recognition, more autonomy, and a push toward the next big idea.
 
It’s the obvious response to exceptional success. Yet, the next big idea rarely comes. Our research suggests that this is because extreme success is not simply a bigger version of ordinary success. It can fundamentally change how innovators see themselves and how they respond to the people around them.

Drawing on both archival and experimental data, our research shows that innovators who deliver one extraordinary hit are strikingly unlikely to repeat it, and the way organizations manage them after the win is part of the reason why.1

The Stress of Exceptional Success

Why is extreme success particularly counterproductive for innovators? Our hypothesis was that the strengths that help someone produce a breakthrough are not always the same as the strengths necessary to sustain innovation over time. Why?

First, exceptional innovators may be highly skilled at solving hard problems, but they are not necessarily leaders who instinctively build teams or recognize and draw on the strengths of others. After an extreme success, the exceptional innovator may come to attribute the achievement mainly to their own talent and hard work and assume that they can do it again on their own. As a result, they may undervalue what a team can contribute and overestimate the costs of coordination. Consequently, even though their earlier success often attracts collaborators, they may see less need to work with others on the next idea.

Second, extreme success can inflate how innovators see their own status in the organization. This can shift their attention away from developing the next big idea and toward preserving the standing that came with the last one. Such a shift can also make them less open to feedback that could improve the next idea but might threaten their newly acquired standing.

To test our hypothesis, we looked first at four years of data from a global automotive company’s pool of 1,145 ideas submitted by 236 serial inventors. This company runs an internal ideation platform where employees can submit ideas. When an idea is selected and implemented, the inventor receives a monetary reward whose amount is set after the fact. There’s no cap on the reward: Most ideas earn modest amounts, but a few have generated significant payouts. It’s not a bonus or commission structure; the variable cash prizes are tied directly to the value of each idea.

The monetary rewards for implemented ideas in this full data set ranged from zero to 11,313 euros (about $13,112), with a mean of 290.83 euros and a standard deviation of 760.28 euros. We defined “extreme success” as an exceptionally large reward relative to the system’s normal reward pattern — specifically, one that exceeded 1,051 euros (one standard deviation above the mean).

In comparing innovators who had experienced extreme success with those who had not, we found that prior extreme success significantly reduced the likelihood of subsequent idea implementation by approximately 42%, on average.

This analysis supported our hypothesis about the rate of future success. Extreme success triggers two psychological shifts that explain its negative effect on future innovations. First, the success inflates inventors’ self-perceived social status within the organization. Second, it reduces their willingness to engage in team-based idea development: Team collaboration decreased by 16%, on average, after an extreme success, our research found. Notably, this pattern is not limited to purely individual wins; it can also emerge when the extreme-success idea is developed in a team.

Validating the Findings With a Randomized Experiment

To understand more about the cause of this dynamic, we set up an online experiment with 300 professionals in the U.K.

Participants were instructed to imagine themselves as employees at a confectionery company that was organizing an innovation contest to develop a new kid-friendly packaging design. They read a detailed brief about the contest requirements, which included submitting a 500-word proposal that included details related to visuals and interactive elements of the new packaging design. To make the task feel more realistic and engaging to the test subjects, participants were asked to submit a short written overview outlining their proposed design.

Following the submission of their ideas, each participant received a randomly assigned evaluation from a supposed expert panel. Contestants rated as extremely successful were told that their proposal surpassed all expectations and scored 95 out of 100, while those in the ordinary success condition were told that their proposal met all expectations, with a score of 70, and those in the failure condition were told that their proposal fell short of the expectations and scored only 40.

To measure self-perceived social status, participants were then presented with an image of a ladder with rungs representing employees’ standing within the company based on their role in driving innovation. Participants were informed that people at the top of the ladder were the most influential in driving innovation — that they received the most recognition for their innovative ideas, earned the highest respect for their contributions, and were seen as leaders in innovation. In contrast, at the bottom were people with the least influence on innovation, who were least recognized for their ideas, received little respect for their contributions as innovators, and were not seen as leaders in innovation. Participants were then asked to consider how their colleagues would rate them and to select the number (on a 0-10 scale) that best reflected their position on the ladder.

After their valuation and self-evaluation, participants were invited to enter another contest, to develop a new chocolate flavor, and were presented with the following question: “As you prepare for the contest, you have a chance to join a team of your colleagues who have invited you to join their team. How would you prefer to participate in this contest?”

As we found in our automotive company’s ideation platform data, participants who had been rated as extraordinary successes saw themselves as better innovators and were less interested in joining a team.

Preventing One-Hit Wonders: Four Best Practices

Extremely successful innovators and contributors are disproportionately important for organizations. They signal exceptional talent, shape outcomes, and raise the bar for everyone around them. Yet, the very success that marks them as stars can undermine what they do next. We are not suggesting that extreme success makes future success impossible; in rare cases, innovators do deliver repeated breakthroughs. But our evidence suggests that they do so despite the headwinds that extreme success creates, not because of any momentum it provides.

Many organizations are good at celebrating exceptional innovators after a major win, but when that success is not managed carefully, celebration can become counterproductive. More status, latitude, and deference may encourage the very behaviors that make future success harder to repeat: less openness to input, less collaboration, and more reliance on individual judgment.

The person first celebrated as a “rock star” can gradually come to operate like a “cowboy,” work like a “lone wolf,” and be experienced by colleagues as a “brilliant jerk.” The labels vary, but they point to the same managerial problem: A past success has been allowed to distort the behaviors needed for future success. To reduce this risk, managers should rethink how they handle innovators after extreme success and try to contain status inflation and make collaboration the most attractive path to the next win. Consider these four practices for managers.

• Make the innovator’s support network visible, and reward it. Extreme success rarely happens in a vacuum. Even if an idea is formally attributed to one person, it was often shaped by a broader network, both visible and invisible. Borrowing from scientific publishing, ask innovators to complete a short contribution statement that names direct collaborators and “invisible” supporters, such as colleagues who offered feedback, resources, or encouragement. This could also be paired with a peer-bonus feature, similar to the peer-recognition systems used at Google: After a big success, the winner receives a small amount of additional bonus money that can be allocated only to people who contributed to the success but didn’t appear on the formal submission.

  Together, these practices dampen “it was all me” attributions, make collaboration tangible, and signal that the organization rewards the ecosystem around an idea, not just the individual who submitted it.

• Frame extreme success as a milestone, and follow it with honest feedback. To prevent innovators from feeling like they have “made it,” managers should avoid excessive hype and rock-star language. Instead, frame extreme wins as major milestones that mark progress, while making it clear that the real goal is to figure out what comes next. Crucially, this framing should be paired with constructive, even disconfirming feedback — an approach that has proved effective in sustaining high performance in both trading and innovation.2

  Honest developmental input after a big win counters the status-related thinking that can set in when everyone around a successful innovator shifts to pure praise.

• Restructure rewards to sustain motivation and encourage teamwork. Rather than maximizing immediate cash payouts for breakthrough ideas, cap short-term rewards and shift the bulk of bonuses to deferred structures, such as equity grants, monthly stock units, or phased payments over time. Meanwhile, also broaden what gets rewarded: Recognize not only the idea generator but also the contributors needed to evaluate, develop, and implement ideas.

  Research shows that team-based reward structures stimulate collaborative behavior, and deferred incentives maintain motivation while fostering long-term alignment with the organization’s innovation goals.3 This follows the same logic that CEOs already apply to executive compensation through vesting schedules and long-term incentive plans.

• Invest in supportive leadership after a big win. The period after extreme success is when innovators are most vulnerable to status inflation and collaborative withdrawal. This is when managerial attention matters most. Research shows that supportive leadership — motivation, encouragement, and active engagement from a supervisor — can strengthen collaboration and counteract the isolating effects of a high-profile win.4

  Rather than leaving star innovators to coast on reputation, managers should actively promote teamwork, connect innovators with new collaborators, and reinforce that the next idea is a shared endeavor.

All in all, extreme success carries hidden innovation costs because it changes how high performers see themselves and work with others. The patterns are predictable and fixable, but few organizations manage them deliberately.

The biggest unlock for your innovation pipeline may therefore be the simplest: Help your best people win repeatedly, without losing the behaviors that got them there.

The most valuable thing a professional produces is not a faster analysis or a better summary. It is insight: a genuinely new way of seeing a problem, a connection that changes how they understand a situation, or a pattern that nobody has named. Think of a strategy consultant who identifies the real competitive challenge behind a client’s margin erosion, or a team member who notices a silence in the data that everyone else has learned to take for granted.

What makes this kind of discovery hard is that expertise, the very thing that makes professionals effective, also makes certain kinds of insight difficult to reach. The frame that lets them see a problem clearly also shapes what they look for — and what they stop looking for. But insights that change thinking are often found at the edges of that frame, such as in the friction between competing interpretations.

Conversational AI is already moving in this direction. A well-constructed prompt can surface competing interpretations, expose gaps, and challenge assumptions. But agentic AI — systems that are configured and directed rather than conversed with — can take users further still. Unlike a prompted conversation that is bounded by what a human supplies and thinks to ask, an agentic system holds more data and sustains analytical orientations across entire data sets without losing the thread. The discovery moves are the same; the depth is not.

Using agentic AI this way demands a professional skill different from both prompting and automation: knowing how to design systems for insight and how to make sense of what they reveal. We call it directing intelligence.

Two Ways of Working With AI

Most professionals encounter AI as a conversation: They type a question, evaluate the response, refine, then ask again. Their job is to supply the context and hold the analytical thread, and the exchange exists only as long as the window is open. The skill this demands is articulation, and specifically knowing what to ask, how to phrase it, and when to push back.

Agentic AI requires a different kind of interaction. Where prompting is reactive (a human asks, the model responds), an agent is proactive: The user configures it, and it operates. That configuration rests on three choices:

• Context (what the agent can access). Where a prompt depends on what is pasted in, an agent’s context is persistent: It’s connected to databases, documents, and records that the prompter chooses and that endure across interactions.
• Capabilities (what the agent can do). A prompted AI generates text, but an agent acts: running analyses, querying databases, comparing data sets, executing multistep analytical routines, and invoking specialized skills and tools without waiting for human input at each stage.
• Orientation (what the agent pays attention to). This moves past an instruction on how to produce a specific output and instead serves as an analytical directive by setting a purpose and trajectory that shape how the agent encounters whatever the data reveals.

The same context and capabilities, given different orientations, will surface different patterns. A single professional can direct multiple agents against the same data set and get genuinely different discoveries, not by asking different questions but by designing different systems. In practice, the professional designs not just individual agents but the system that connects them, often including an orchestration layer that compares and synthesizes across diverse AI agent outputs.

A single, well-configured agent can produce genuine insight. If an AI agent oriented toward customer behavior is given access to transaction and service records, it might discover that churn is concentrated among clients who are in their second year. The larger opportunity emerges when that agent becomes part of a system. Give the same data to specialist agents oriented toward sales conduct, onboarding experience, product usage, and service history, and each surfaces a different explanation for that second-year pattern. Add an orchestration agent configured to compare their outputs, identify where the explanations converge and diverge, and surface the contradictions that matter most, and the system produces something none of its parts could generate alone. The professional’s job is to design this system and evaluate what it reveals.

Four Approaches to Discovery With Agentic AI

Discovery rarely arrives through a single, well-aimed question. It tends to emerge from friction: from putting things in contact that are normally kept apart. Each of the four approaches that follow creates a specific kind of friction: between competing interpretations, between data and the conversations that surround it, between causes and the levels where they hide, and between categories and the reality they were meant to describe. The insight emerges from the friction itself. (See “Four Ways to Direct Intelligence.”)

Each move is also available through prompting. What agentic architecture adds, through what agents access, do, and pay attention to, is depth.

1. Use multiple lenses. The most reliable way to see past a single interpretation is to hold several at once — not sequentially, but simultaneously — in deliberate tension. When multiple well-reasoned frameworks are applied to the same situation and they contradict one another, the contradiction is itself informative. It points toward something none of the frameworks would surface alone.

Consider a strategy consultant engaged with a midsize manufacturer of engineered metal parts serving the aerospace, automotive, and energy markets. Margins have eroded for three consecutive years, and the CEO has blamed it on competitive pricing pressure across all three segments.

The consultant designs an agentic system. Four specialist agents are configured against the full engagement data set, which includes financials, competitive intelligence, customer contracts, interview transcripts, and internal strategy documents. Each agent is oriented toward a different strategic framework, with an orchestration agent configured to synthesize the specialist agents’ analyses.

A Michael Porter agent finds that aerospace is structurally attractive whereas automotive is structurally punishing, suggesting that the segments should not be treated equivalently.

A Jay Barney/VRIO (value, rarity, imitability, and organization) agent identifies a proprietary metallurgical process and aerospace relationships as genuinely rare and difficult to imitate but notes that capital expenditure is split roughly equally across all three segments. The resource advantage is real but the company is not organized to exploit it, because investment is spread evenly across all three segments rather than concentrated behind the aerospace business where the advantage actually lies.

A Richard Rumelt agent argues that the stated strategy is not a strategy at all: “Grow through diversification across end markets” is called out as a goal dressed up as a direction.

A Roger Martin agent maps the “where to play and how to win” decisions and finds them incoherent. The capabilities required to compete in aerospace directly contradict those required in automotive, and the company is attempting both.

No single framework produces a diagnosis. The friction between them surfaces a question nobody has been asking: whether the company should narrow its scope to the one segment where its genuine competitive advantage meets structurally attractive conditions.

Conducting this analysis by prompting a large language model to analyze a situation through Porter, then VRIO, then Rumelt, then Martin will generate real tensions worth exploring. The constraint with such prompting is that the consultant holds the thread. Each framework shift requires a new prompt, and the synthesis depends on the professional’s ability to carry all four interpretations forward simultaneously.

Agentic architecture changes what is possible. Each specialist agent works with the same evidence base at the same depth, so the contradictions are directly comparable. The orchestration agent reviews all four analyses and identifies where the frameworks agree, confirming what is robust (and where they clash) and pointing to what is unresolved. The professional’s role shifts from maintaining the analysis to evaluating what the system surfaces.

2. Surface silences. Some of what an organization knows about itself never makes it into a meeting, a report, or a strategic plan. Surfacing silences means systematically comparing what exists in the data with what appears in the discourse that surrounds it, and treating the gap as analytically meaningful. The insight lives in what is absent.

Consider a strategy consultant engaged with a specialized practice group within a larger professional services firm. The practice does premium, expertise-intensive work in a technical niche that demands deep domain knowledge and long-standing client relationships. The practice leader has never lost a competitive bid, personally oversees every significant engagement, and turns away more work than the practice accepts. The growth narrative is compelling: strong demand and few competitors at the same quality tier.

The consultant configures an agent with the full engagement data set, including transcripts from interviews with the practice leader and company executives, the practice’s strategic plans, market analyses, competitive intelligence, and financial records. The AI agent can systematically inventory themes across both the interview data and the formal strategic documents. Its orientation: Pay attention to what appears in one body of material but not the other.

The AI agent surfaces a silence: organizational dependency on a single individual. The practice leader, a senior expert with two decades of experience, appears as an implicit presence across virtually every topic in the interview transcripts. Growth capacity, client relationships, quality standards, pricing, talent development: Roughly a third of all substantive discussions reference his judgment, relationships, or standards, yet he is never explicitly named. Strategic plans discuss market opportunity, competitive positioning, and investment requirements, but not one of them names what happens when every advantage the practice holds is embodied in one person.

The agent also flags other potential gaps. An apparent absence of succession planning turns out to be addressed in firm-level documents outside the agent’s data set; silence around competitive risk reflects a data gap much more than an organizational one. The consultant investigates and sets both findings aside to focus on the one that holds up: the silence around talent dependency.

Everyone in the company knows that the practice leader is exceptional. But knowing it and naming it as the central strategic constraint are different things. The strategic conversation is organized around market opportunity, but the operational reality is organized around one person. This changes the question from “Can this practice grow?” to “Can it grow without first solving the problem nobody has put on the table?”

Prompting an AI to compare interview themes against strategic plan priorities will flag disconnects, and a skilled professional can push toward what seems absent. However, after weeks or years of seeing things in a particular way, the professional’s perspective becomes constrained in that way, which also shapes how they prompt AI.

The AI agent does not share that immersion in a particular context. It compares the full interview corpus against the full set of strategic documents, without having absorbed the team’s framing. And because the comparison is exhaustive rather than selective, the pattern’s pervasiveness is verifiable rather than impressionistic. The AI can more easily notice what familiarity has made invisible.

3. Bridge levels. When something goes wrong in an organization, the explanation is usually found at the same level where the symptom appeared. For instance, a portfolio problem gets a portfolio explanation, or an operational failure gets an operational diagnosis. But causes don’t always live where symptoms surface. Bridging levels means tracing causal chains across scales of analysis, connecting micro decisions to macro outcomes, and linking macro patterns to the specific behaviors that produce them. The insight comes from traversing the levels that are normally examined in isolation.

Consider a strategy consulting team that is working with a diversified industrial corporation that comprises three divisions and is seeing a declining return on invested capital (ROIC). The corporate narrative is a familiar one: The company is facing market headwinds and competitive pressure. The quarterly review focuses on portfolio-level metrics. Everyone nods.

The lead consultant configures an AI agent with data from three levels of the business (corporate, divisional, and operational) simultaneously, including corporate portfolio metrics and capital allocation records, divisional P&Ls and competitive positions, and operational data within each division. The AI can trace statistical relationships across all three levels. Its orientation: Look for places where a cause and its symptoms sit at different levels.

The agent traces the ROIC decline downward. It is concentrated in Division A, but not because Division A is underperforming within its market. A capital allocation formula implemented three years ago weights recent revenue growth when distributing investment. Division A is the fastest-growing division, but it is also growing in a commoditizing market with declining margins. Division B, slower-growing but carrying the strongest competitive position and the highest margins in the portfolio, is being systematically starved of capital. Its competitive edge is eroding quarter by quarter.

When the ROIC is traced upward, it becomes clear that the capital allocation formula was designed to invest behind growth. At the corporate level, this sounds entirely rational. But at the divisional level, it overinvests in deteriorating competitive dynamics and underinvests in defensible advantage. The ROIC decline the CEO has been attributing to the market is actually being produced by a corporate policy operating exactly as designed.

So here, the symptom lived at the portfolio level but the cause lived in a corporate policy. And the intervention — redesigning how capital is allocated — sat at a level that neither divisional management nor the CEO’s market narrative was examining.

Prompting an AI to explain how corporate capital allocation might be affecting divisional performance, or whether investment patterns align with competitive positioning, will surface plausible hypotheses worth investigating. The constraint is that the prompter has to suspect the connection before they can ask the question.

The agent doesn’t need that prior suspicion. With access to all three levels at once, it can follow connections the consultant didn’t know to look for, because it was configured to work across the levels rather than to test a hypothesis already in hand. The diagnosis was there in the data all along; it was simply invisible from any single level. And the finding can feed the next step: A stress-testing agent can check whether the formula’s measure of revenue growth actually tracks competitive strength in each division or whether that category is itself part of the problem.

4. Stress-test categories. Every organization runs on categories: the classification systems that sort customers, failures, costs, and behaviors into named buckets that route decisions to the right people. The problem is that categories are designed to reflect how an organization thinks, not necessarily how the world behaves. Stress-testing categories means comparing your classification system against the operational reality it was built to describe, and treating the divergence as the finding.

Consider a data analyst at a ready-mix concrete company investigating a persistently high rejected-load rate, where trucks arrive at job sites and are turned away. The company classifies rejections into five categories: wrong mix design, late delivery, quality failure, customer change, and over-order. Each category connects cleanly to a specific department. Wrong mix and over-order go to sales, late delivery goes to dispatch, quality failure goes to the plant, and customer change is marked uncontrollable.

The analyst configures an agent with the formal rejection classifications alongside the full operational record, including order tickets, dispatch logs, GPS tracking data, batch plant records, driver comments, weather data, and customer communications. Its explicit job is to run a pattern analysis across the entire rejection data set and attend to where the formal categories and the actual causal structure diverge.

The agent finds that the categories are both imprecise and actively misleading. Many late-delivery rejections cluster on specific days and weather conditions. Traced backward, they are downstream effects of a batch plant aggregate hopper that slows under adverse weather conditions, delaying loads that then arrive outside the pour window. Categorizing the rejection as late sends the investigation to dispatch, but the cause lives in the plant.

A cluster of customer-change rejections follows a different pattern. They’re concentrated in jobs where a general contractor placed the order but a subcontractor controlled the pour schedule. The label marks those rejections as uncontrollable when they are, in fact, a predictable coordination gap that the company could address. And then there is a pattern for which the formal system has no category at all. Driver comment fields across hundreds of loads record the same informal notation: Site not ready. The truck arrived, but the job site could not receive the pour. Drivers waited, returned, or were rerouted, and dispatchers coded the rejection into whatever category seemed fitting. Here, a recurring root cause existed only in the margins of the data because the classification system was never built to see it.

Pulling a sample of rejection records and asking an AI tool whether the categories are capturing real causes will surface plausible candidate causes and flag obvious mismatches. The constraint is sample size and selection: What the analyst pulls inevitably reflects what already seems worth investigating.

The agent, however, tests every category against the full operational record, across thousands of loads, with no prior assumption about which categories are accurate. It zeros in on weather patterns because it examined everything rather than only a representative slice. It surfaces the “site not ready” pattern because it has no deference to an existing taxonomy. Once those mismatches are surfaced, the broader agentic system can pursue them. A bridging agent could trace the “site not ready” pattern upward from driver comments to dispatch costs to fleet utilization to portfolio-level financial impact, quantifying a problem the formal system had rendered invisible.

How to Direct Intelligence Skillfully

Gaining the most useful results from directing AI agents to surface new insights requires practice, like any new skill. The following are some guidelines to keep in mind.

Configure for discovery, not answers. Start with how you set the system up. The instinct, especially for professionals trained to specify deliverables clearly, is to tell an agent what to find. Resist that instinct. An agent configured to confirm a competitive advantage will confirm it and, in doing so, won’t discover that the advantage is real but being systematically diluted by incoherent investment. A system of agents configured to hold four competing strategic frameworks in tension will produce something harder to digest and considerably more valuable: a contradiction that points at a question nobody was asking. This is problem-setting rather than problem-solving. Humans define what agents should pay attention to and not what they should conclude.

Treat the unexpected as signal, not error. What you do when agents produce something you didn’t expect matters as much as how you configured them. The natural response is to treat the unexpected as error. In efficiency mode, divergence from the anticipated output is waste, and the job is to correct that divergence. In discovery mode, that instinct runs exactly backward. An unexpected output is the most valuable signal the system can generate, because it tells you something about your own assumptions or about the actual structure of the problem you thought you understood. The “site not ready” pattern had existed in the driver comment field for years. The talent dependency permeated a third of the interviews conducted with employees, yet it appeared in none of the professional services firm’s strategic plans. In both cases, the discipline for anyone directing intelligence is to resist the urge to explain that kind of surprise away before investigating what it points to.

Evaluate proposals, not conclusions. The discipline to investigate rather than reject the unexpected carries directly into how you interpret what agents produce. Patterns surfaced by an agentic system are proposals, not findings. They open inquiry rather than close it. When the bridging agent traced ROIC decline to the capital allocation formula, the right response was to test that pattern against alternative explanations and assess whether the effect was large enough to matter. When the Porter and VRIO agents contradicted each other on the strategic situation, the productive move was not to decide which agent was right but to recognize that the contradiction was pointing at a question the company had never confronted. And when the “site not ready” pattern came through, it required verification against GPS time stamps and driver logs before it became actionable. Machine-generated patterns become knowledge through the professional’s judgment, not in spite of it.

Track what you advance, and what you reject. Over time, tracking which proposals you pursue and which you set aside builds something useful: a record of your own analytical instincts, which moves generate value in which contexts, and where your judgment tends to foreclose inquiry before it should. Your rejections reveal as much as your discoveries.

The agentic systems described in this article already involve orchestration: specialist agents whose outputs are synthesized, compared, and set in tension through an orchestration layer that the professional designs. As these systems mature, the orchestration will deepen. A silence-surfacing agent could autonomously trigger a lens-multiplying agent to investigate what it finds; a bridging agent could hand a pattern to a category-testing agent to probe for boundary cases, without the professional initiating each step. The professional’s role could shift from designing individual configurations to shaping the conditions under which agentic systems discover productively. Such moves will compound in ways that are difficult to fully anticipate now.

What won’t change is the underlying skill. Directing intelligence draws on capacities that have always distinguished exceptional professionals: the ability to frame a problem so that new things become visible, and to hold contradictions open long enough to learn from them. Agentic AI does not replace those capacities. It gives professionals a more powerful medium through which to exercise them. The professionals who thrive will not be the ones who automate the most. They will be the ones who understand that discovery is the highest-value thing a professional produces, and who have the judgment to direct intelligence toward it deliberately.