The New Agile: A Collective Delivery Demand Matrix for Agile, High-Performance, and Dynamic Teams

Abstract

This article proposes the Collective Delivery Demand Matrix (CDD Matrix), which repositions the New Agile as a correspondence property between collective delivery demands and teams’ operating regimes, aiming to form teams that are simultaneously agile, high-performing, and endowed with dynamic capabilities. It addresses a gap in the agility literature: the Agile Manifesto principles have been extended to many contexts, yet no theory explains when they should be maintained, adapted, or replaced. Developed as a theoretical-conceptual essay, the work synthesizes studies on naturalistic decision-making, team learning, neurostrategy, sociotechnical systems, and decision support. It proposes a taxonomy of seven demand typologies along five classificatory axes, and a correspondence matrix that articulates demand types, operating regimes, shared heuristics, sociotechnical configurations, and managerial errors. The matrix clarifies that agility generates value when it first favors effectiveness, delivering the right objective in the right regime at the appropriate risk level, and only then efficiency. The New Manifesto reformulates the Agile Manifesto as conditional correspondence rules sensitive to demand typology, and six testable propositions guide future empirical research. The main finding is that a team becomes simultaneously agile, high-performing, and dynamically capable only when its operating regime corresponds to the demand typology it faces, matching effectiveness before efficiency, calibrating risk, and building a renewable repertoire of shared heuristics. For managers, teams, and organizations, this implies diagnosing the delivery demand before selecting a method, composing and leading teams according to the demand’s dominant axes, and protecting the psychological safety that turns reversible error into collective learning. The contribution integrates the Agile Manifesto, decision theory, the sociotechnical tradition, and neurostrategy into a single theory of Agile, High-Performance, and Dynamic Teams (EAD²), oriented toward verifiable objectives, goals, and results.

1. Introduction

In 2001, seventeen software development professionals gathered at Snowbird to articulate principles for handling demands whose requirements became clearer over the course of the work (Beck et al., 2001). These principles spread to manufacturing, healthcare, education, and corporate innovation. When a set of principles conceived for a specific class of demands begins to guide multiple forms of collective action, it becomes necessary to explain under what conditions it remains valid, under what conditions it requires adaptation, and which mechanisms sustain its effectiveness.

While the agile movement has its roots in software engineering (Beck et al., 2001), these methodologies have since permeated various departments and entire organizations, driving a surge of contemporary reviews that evaluate the discipline’s current state (Perides & Vasconcellos, 2025; Steegh et al., 2025). Current literature on organizational agility increasingly frames it as a set of dynamic capabilities—specifically sensing, seizing, and reconfiguring. This conceptualization is deeply anchored in the dynamic-capabilities and resource-based perspectives, moving beyond a strictly contingency-based approach (Asghar et al., 2026; Nguyen et al., 2025). Analyzing agility at the team level reveals that adaptability does not widely guarantee enhanced performance. Instead, the advantages of a team’s adaptive processes are highly contingent upon the specific nature of the task—proving beneficial in scenarios like creative endeavors, yet potentially less effective in others, such as knowledge-integration activities (Georganta et al., 2023). Furthermore, collective competencies, including adaptive and absorptive capacities, are critical drivers of both innovation and overall performance (Singh & Jha, 2024). Rather than resolving the core inquiry of this theoretical essay, this emerging body of evidence intensifies it: establishing precisely when an agile operating regime ought to be preserved, modified, or entirely substituted based on the specific collective delivery demand confronting a team.

The contrast among four Collective Delivery Demands (CDD) illustrates the argument. A team engaged in digital prototyping or product R&D generally operates with still-open requirements, successive learning cycles, and failures that can be converted into useful information. A team that opens a store in thirty days faces a fixed deadline, interdependent fronts, and costly though recoverable failures. An emergency room or firefighting team may make decisions in seconds under acute stress and at risk of death. A sports team acts against an active opponent in real time, adjusting heuristics during the game. All four situations require high-level coordination but differ in time pressure, error cost, uncertainty, adversariality, and emotional load; they therefore require distinct operating regimes.

In day-to-day organizational life, CDD appear in commercial, industrial, and service operations; in professional offices such as accounting, law, engineering, and PMOs; and in field operations such as construction, transport, and technical assistance. These arenas map onto the CDD Matrix typologies and situate the model’s managerial applicability. Added to this scenario is the acceleration of digital transformations, encompassing artificial intelligence, next-generation telecommunications, cloud computing, wearables, and collaboration platforms, which continuously alter the tools, flows, and heuristic repertoires available to teams. A team that does not reconfigure itself in response to these innovations ages its operating mode even without changing typology. It is this imperative of permanent reconfiguration that justifies the second D in EAD²: not only high performance, but also Dynamic, the capacity to adapt to the regime the demand requires.

This essay proposes the CDD Matrix and maintains that agility operates as a correspondence property. A team is agile when its operating regime, understood as the set of technical, cognitive, affective, and sociotechnical processes, artifacts, and results, corresponds to the typology of the collective delivery demand it executes. The 2001 Manifesto describes a regime particularly suited to exploratory demands with emergent requirements, learning in cycles, and greater error reversibility. The New Manifesto preserves this legacy and extends it, making the principles sensitive to demand typology and the required neurocognitive architecture.

Three labels structure this contribution. The New Agile names the thesis. The CDD Matrix names the model that operationalizes it, organizing empirical theories on decision-making, biases, decision support, and sociotechnical design into microfoundations for leading Agile High-Performance and Dynamic Teams (EAD²) at work. The New Manifesto names the normative output of the model, with a common base for every team and differentiated layers by typology. The object on which the model operates is the Collective Delivery Demand (CDD), the unit the matrix classifies and to which it assigns a regime.

The theoretical gap of the article lies in two recurring confusions: treating “The Agile” as a universal method and equating agile teams with high-performance teams. The dynamic capabilities literature corrects this by showing that superior organizations sense changes, seize opportunities, and reconfigure resources and competencies (Teece et al., 1997; Teece, 2007). At the team level, this capacity manifests as delivery diagnosis, choice of the appropriate regime, and reconfiguration of heuristics, roles, protocols, and decision support systems. In the Drucker (1967) synthesis, efficiency means doing things well, while effectiveness means doing the right things. Here, effective performance combines delivery effectiveness, resource efficiency, and protection against critical risks.

The guiding question of the essay is as follows: how can Collective Delivery Demand (CDD) at work be classified and, for each typology, how can the team’s operating regime be selected so as to combine effectiveness, efficiency, and protection against critical errors? The objective is to construct the CDD Matrix relating demand typologies, operating regimes, heuristic metabolism, sociotechnical design, dynamic capabilities, and decision support.

The central contribution of the CDD Matrix relates to a second neurocognitive thesis: adaptive teams are distinguished less by their nominal method than by the collective capacity to generate, share, and refine heuristics during operations. This capacity, termed heuristic metabolism, designates the continuous conversion of experience, tacit knowledge, and learning into context-sensitive practical rules. Heuristics are action knowledge, know-how, and as-built knowledge that enable decision-making under pressure without recalculating everything in each situation. An intelligent team, in this sense, is one whose stock of useful heuristics grows, circulates, is refined, and corresponds to the delivery demand it executes.

The essay adopts, as its epistemological foundation, Powell’s (2011) neurostrategy program, which recognizes three legitimate contributions of neuroscience to strategy: construct validation, theory testing, and practice informing, and warns against three illegitimate uses: the mereological fallacy, reverse inference, and the explanatory seduction of neuroscientific evidence. Each association between a neural mechanism and a coordination practice is subjected to this filter, and a limitations section discusses the risks of the enterprise.

Methodologically, the article is a theoretical-conceptual essay, appropriate for organizing dispersed constructs, proposing conceptual relationships, and deriving research propositions (Gilson & Goldberg, 2015; Jaakkola, 2020). The synthesis integrates literature on team learning, naturalistic decision-making, somatic marking, working memory, trust, inter-brain synchrony, sociotechnical systems, and decision support. These constructs are incorporated into the CDD Matrix, which relates demand typology, operating regime, heuristics, sociotechnical configuration, and managerial error. Verifiable cases, such as Apollo 13, Miracle on the Hudson, Three Mile Island, and Tham Luang, are used as analytical illustrations rather than as primary empirical tests. The literature was selected purposively for conceptual relevance to the demand–regime correspondence rather than for exhaustive coverage, drawing in particular on recent systematic reviews of agile teams and organizational agility (Steegh et al., 2025; Nguyen et al., 2025). The CDD Matrix was constructed in three steps: defining the classificatory axes, deriving the demand typologies they generate, and mapping each typology onto its operating regime, heuristics, sociotechnical configuration, and characteristic error; the illustrative cases were chosen to exemplify these typologies rather than to serve as empirical tests. The entire endeavor remains subject to the critical filter of Powell’s (2011) neurostrategy, particularly regarding the mereological fallacy, reverse inference, and the explanatory seduction of neuroscientific evidence.

2. Literature Review and Theoretical Foundations

2.1. The Structure of Collective Delivery Demand (CDD)

The structure of Collective Delivery Demand (CDD) allows for distinguishing among routine, ill-structured, and complex deliveries. In the tradition of Simon (1947, 1977), programmed decisions are repetitive, tractable through procedures, and convertible into routines, coordination algorithms, and stopping rules; ill-structured decisions, in turn, require judgment, satisficing, and progressive construction of a response. In the vocabulary of this article, team maturity manifests as the capacity to convert initially open demands into repertoires of action, protocols, heuristics, and completion criteria, without losing reversibility.

In the field of contingency theory and strategy, Mintzberg (1978; Mintzberg & Waters, 1985) distinguishes deliberate decisions, oriented by intention, planning, and prior coordination, from emergent decisions, formed in the course of action. Demands that admit deliberation approximate planned regimes; demands that require emergence call for continuous adaptation. The dynamic capabilities literature extends this formulation by showing that sustainable advantage depends on sensing changes, seizing opportunities, and reconfiguring resources and competencies (Teece et al., 1997; Teece, 2007). At the team level, the New Agile translates this macro-logic into a micro-delivery capability: sensing the nature of demand, choosing the operating regime, and reconfiguring heuristics, roles, protocols, and decision-support systems.

In this essay, a complex collective delivery demand is one in which the team faces a large state space, long action sequences, successive decision branches, uncertainty about consequences, accumulated risks, and interdependent contingency plans. Each new state reached in the flow alters the available alternatives, redistributes risks, and requires updating heuristics, protocols, and coordination. In managerial terms, the complex demand operates as a sociotechnical algorithm oriented toward the objective-goal-result triad: objectives define direction; effective goals and OKRs define stopping rules; and the team updates its course of action according to emerging evidence, constraints, and risks. Within this framework, the problem typology is subordinated to the demand typology: a problem designates an occurrence, critical event, deviation, or bifurcation that arises within a collective delivery demand. When the flow is covered by heuristics, protocols, or stopping rules, the team responds through recognition and adjustment; when the occurrence breaks the available repertoire, the demand requires learning, recomposition, or a regime change.

This need for continuous reconfiguration aligns with recent multidimensional models of organizational agility (Asghar et al., 2026). Specifically, treating agility as a dynamic capability requires teams not only to sense environmental shifts but to structurally realign their sociotechnical algorithms to match the specific constraints of the new demand (Nguyen et al., 2025).

2.2. Naturalistic Decision-Making and Adaptive Expertise

Klein et al. (1986) and Klein (1993, 1998) documented, in studies with firefighters, nurses, and military personnel, that experts under pressure rarely compare alternatives: they recognize the situational pattern and immediately activate a consolidated course of action, mentally simulating it before executing it. This recognition-primed decision model describes the regime appropriate to critical response demands. Hatano and Inagaki (1986) distinguished routine expertise, which executes consolidated procedures efficiently, from adaptive expertise, which is capable of reconfiguring the repertoire when faced with the novel; the latter is the form of competence that heuristic metabolism presupposes. Gigerenzer and Gaissmaier (2011) added that simple, ecologically adapted rules frequently outperform complex models under uncertainty: the heuristic operates as the correct mode of action when the environment so requires.

2.3. Team Learning and Psychological Safety

Two complementary sources, introduced here in chronological order, ground the notion of heuristic metabolism. Nonaka and Takeuchi (1995) explain how tacit knowledge is converted into shared collective repertoires; Edmondson (1999) then shows that psychological safety is the condition under which such knowledge is safely surfaced and refined. Their integration is deliberate: the first supplies the content that circulates within the team, and the second supplies the condition that allows it to circulate and improve.

Edmondson (1999) demonstrated, in a field study with fifty-one teams, that psychological safety is associated with learning behavior and that the latter mediates the relationship between psychological safety and performance. The counterintuitive finding from her hospital research, in which better-performing teams reported more errors, illustrates the central argument: superior teams convert errors into shared heuristics because they operate in a regime that makes it safe to surface them. Psychological safety is a regime variable that, at the neural level, conditions the team’s availability for learning (Edmondson, 2018).

Nonaka and Takeuchi (1995) offer a complementary vocabulary: the SECI model describes the spiral through which tacit knowledge is Socialized, Externalized, Combined, and Internalized, converting individual experiences into collective repertoires. Heuristic metabolism can be understood, in this key, as a specific SECI spiral: initially tacit heuristics are shared, verbalized, combined with other knowledge, and subsequently internalized as cognitive and organizational routines. Psychological safety, in this logic, becomes a critical condition for the honest externalization of these heuristics, because without it, tacit knowledge remains hidden and the organizational learning spiral tends to stall.

2.4. Decision Neuroscience: Dual Systems, Somatic Marking, and Memory Under Pressure

Four neurocognitive contributions underpin the CDD Matrix without displacing the article toward a neurobiological discussion. The first is dual-process theory (Kahneman, 2011): System 1, fast and pattern-based, sustains the expert’s immediate recognition; System 2, slow and analytical, verifies, corrects, and handles the novel. The second is the somatic marker hypothesis (Damasio, 1994, 1996), according to which bodily signals associated with prior experiences guide choices under uncertainty. The third is working memory, whose limits make protocols, role division, heuristics, and Decision Support Systems (DSS) necessary in complex tasks (Miller, 1956). The fourth is procedural and motor memory, decisive in surgery, aviation, sport, and musical performance, for example (Yarrow et al., 2009). Taken together, these four contributions matter here only for what they imply about team design: which regime to favor, when to rely on trained intuition rather than deliberation, and how to protect memory and attention under load. The underlying neurobiology is invoked to explain mechanisms, not to make the argument depend on it.

In this article, stress is treated not as an autonomous object of investigation but as a contextual condition that alters the quality of individual and collective decisions. The literature distinguishes short-term sympathoadrenal activation, potentially useful in critical situations, from prolonged activation of the HPA axis, associated with reduced cognitive flexibility, cooperation, and emotional regulation (McEwen, 2007). In teams, this distinction is central: moderate pressures may favor attention and readiness, while chronic pressures tend to compromise communication, trust, coordination, and learning. Thus, different decisional typologies require distinct combinations of trained intuition, deliberation, memory, shared attention, and emotional regulation.

2.5. Biases, Error Asymmetry, and Cognitive Limits

The heuristics and biases literature (Tversky & Kahneman, 1974; Kahneman, 2011) has documented predictable deviations: anchoring, availability, overconfidence, loss aversion, planning fallacy, and confirmation bias. For the CDD Matrix, the relevance is functional: each typology is vulnerable to specific biases, and part of collective intelligence consists of developing safeguards targeted to the regime in which it operates. Critical response is vulnerable to tunnel vision; deadline-driven execution to the planning fallacy (Flyvbjerg, 2014); iterative discovery to sunk-cost and confirmation bias; the adversarial typology to projection and emotional reactivity; and stable operation to complacency and normalcy bias (Weick & Sutcliffe, 2007). Moreover, recent literature confirms that implementing agile practices does not automatically mitigate these biases across all task types (Steegh et al., 2025). The effectiveness of team adaptation is highly sensitive to the nature of the task, meaning that forcing iterative discovery onto a routine execution problem can actually exacerbate cognitive overload and degrade performance (Georganta et al., 2023).

The pedagogical value of error depends on its reversibility. The 2001 Manifesto thrived in contexts where failures can be recovered, discarded, or converted into information; applying the same logic to high-irreversibility contexts shifts learning to prior training, simulation, and subsequent debriefing.

Error and uncertainty also differ. Error may have an associated probability, allowing for contingencies; uncertainty, in Knight’s (1921) sense, has no known distribution. Since teams operate under limits of information, attention, working memory, and time, complex demands require cognitive load division, protocols, heuristics, DSS, and sociotechnical coordination. Training a team for complex deliveries means training cognitive, emotional, motor, and sociotechnical repertoires, not merely conveying conceptual instructions.

3. The Collective Turn: From the Individual Brain to the Team Brain

Teams transcend the sum of individuals: they constitute a level of analysis in their own right, with emergent properties. Those who decide as a team activate neurocognitive configurations distinct from those activated alone, including circuits of social cognition and mentalization (Frith & Frith, 2012), and the fear of social exclusion activates regions associated with physical pain (Eisenberger, 2012), altering what each member is willing to say or question.

The literature on hyperscanning offers emerging evidence of measurable synchrony between cooperating brains. Hasson et al. (2012) proposed a brain-to-brain coupling model; Reinero et al. (2021) reported an association between inter-brain synchrony, measured by electroencephalography, and collective performance in certain cooperative contexts. These preliminary findings, subject to replication, converge with what the sociotechnical tradition and the team cognition literature (Salas et al., 2008; Cooke et al., 2013) had postulated for decades: effective teams develop shared mental models and act as a distributed cognitive system. Social fear, in this framework, can be treated as a relevant neurocognitive variable rather than an incidental moral concern.

4. The Sociotechnical Tradition and Trust as a Regime Variable

The sociotechnical perspective originated in the Tavistock Institute studies (Trist & Bamforth, 1951) on the mechanization of British coal mines. The authors observed that the most efficient technology, in isolation, proved insufficient to generate performance gains; by disrupting the social subsystem of teams, it frequently degraded their performance. The founding lesson is that optimal work systems optimize both the technical and social subsystems simultaneously, and that treating either as given results in performance inferior to joint optimization. Cherns (1976, 1987), as reviewed by Pasmore (1988), systematized this insight into design principles: minimal critical specification, control of variances at source, role multifunctionality, and congruence between power and responsibility structures.

Sociotechnical systems thus coordinate tasks, technologies, flows, and affective states. Basic emotions, somatic markers, trust, and perceived threat modulate the willingness to share critical information, accept correction, sustain operational discipline, and preserve cooperation under pressure (Ekman, 1992, 1999; Damasio, 1994, 1996; Bechara et al., 1994).

Neuroeconomics offers a substrate for the thesis that trust is a structural variable (Fehr & Camerer, 2007). The experiment by Kosfeld et al. (2005) suggests that trust and cooperation also have psychobiological bases, without reducing the phenomenon to a single hormonal mechanism; in organizational terms, environments of repeated cooperation operate through cohesion mechanisms that go beyond cultural dispositions.

Taken together, this review provides the conceptual foundation of the CDD Matrix: decision theory explains the structure of the demand; naturalistic decision-making explains action under pressure; team learning explains the conversion of reversible error into knowledge; the sociotechnical tradition explains the integration among technology, roles, and relationships; and neurostrategy offers the filter that allows neurocognitive mechanisms to be used with conceptual parsimony.

Leadership, Team Types, and Leader-Team Empathy

Leadership integrates the social subsystem, and its alignment with the team’s operating regime is a variable in that regime. The relationship between the leader and team members depends on bidirectional functional empathy: the leader must read fatigue, fear, and overload; the team must read the leader’s intention, priority, and thresholds. This mutual mentalization, supported by social cognition circuits (Frith & Frith, 2012), is a coordination channel of the collective brain, and the protection of psychological safety determines whether errors surface in time (Edmondson, 1999, 2018).

The weight of leadership varies with demand. In critical response and the critical capsule, leadership tends to be directive: it maintains protocol and regulates panic. In synchronized execution, authority is preponderant but often non-verbal, such as the conductor’s gesture. In iterative discovery, leadership is facilitative and distributed.

In real-time adversarial situations, authority migrates to whoever holds the best situational reading. The same person may lead well in discovery and perform poorly in critical response; leadership must therefore also correspond to the regime.

Acculturation completes this argument. The leader is the carrier and transmitter of the regime’s culture: through choices in communication, corrections, and omissions, the leader signals what is valued, tolerated, or dysfunctional. Preparing the team for the regime is continuous cultural modeling work, supported by senior members and specialists. A consolidated regime culture sustains heuristic metabolism, integrates new members, and enables recognition of when the demand changes and reconfiguration is required.

5. CDD Matrix: Five Axes and a Taxonomy of Collective Delivery Demands at Work

The taxonomy derives from five axes that characterize the structure of a collective delivery demand at work: time pressure; error cost; degree of uncertainty; adversariality; and environmental munificence (Dess & Beard, 1984). Real complexity results from the interaction among the state space, available resources, and the team’s cognitive repertoire: the same demand may belong to different typologies depending on who executes it and the conditions under which it is executed.

The dominant combination of these five axes defines the demand typology and, with it, the appropriate operating regime. The axes are classificatory criteria; the resulting demand types are organized here into seven classes. Two typologies stress the scheme: synchronized execution, defined by extreme collective synchronization, and the critical capsule, in which confinement and collective lethality intensify the regime to the limit.

The five axes are not proposed ad hoc; each is grounded in an established research tradition and captures an analytically distinct property of a delivery demand. Time pressure and error cost derive from naturalistic decision-making, where the tempo of action and the consequences of error define the decision regime (Klein, 1993). The degree of uncertainty follows Knight’s (1921) classic distinction between measurable risk and genuine uncertainty. Adversariality captures the presence of an active opponent that adapts in real time, a property absent from purely technical tasks. Environmental munificence is drawn from the dimensions of organizational task environments (Dess & Beard, 1984), indexing the resources and slack a team can mobilize. Together the axes are intended to be conceptually independent yet jointly sufficient to characterize the structure of a collective delivery demand; their dominant combination, rather than any single axis, defines the demand typology. This contingency logic is consistent with recent evidence that executing a team’s adaptation process yields high performance only under certain task conditions, not others (Georganta et al., 2023), and that team capabilities translate into performance conditional on team decision-making (Singh & Jha, 2024). We treat commitment under existential risk as a latent sixth axis (Section Teams Under Confinement and Existential Risk: Commitment as a Latent Axis), salient only in extreme settings and therefore modeled as a boundary condition. This construction gives the taxonomy content validity, in that the axes span the properties the reviewed literature treats as decisive, and discriminant clarity, in that each axis isolates a different source of variation; it remains open to empirical refinement through the propositions in Section 8.

Table 1 organizes the seven typologies, from predictable error to collective existential risk. Each row relates typology, regime, dominant axes, and examples, allowing the mode of action to be derived from the demand structure.

Table 1. CDD matrix—Typologies of collective delivery demands, operating regime, and examples.

Typology	Operating Regime	Dominant Axes	Examples
Stable operation	Optimized routine	Low uncertainty; high repetition; predictable error	Mature store; stable production line
Iterative discovery	Classic agile iteration	Open requirements; learning by cycles; reversible and informative error	Digital prototyping; MVP; R&D; product design; exploration of new services
Deadline-driven execution	Planned coordination	Fixed deadline; costly but recoverable error; defined critical path	Large-scale construction; 30-day store opening; product launch
Synchronized execution	Real-time synchronization	Time fixed by the score; audible and irreversible error; no opponent; maximum synchronization	Symphony orchestra; choir; ballet company
Real-time adversarial	Intra-operation adaptation	Active opponent in zero-sum game; continuous real time; on-the-fly learning	Volleyball; soccer; negotiation table; trading
Critical response	Recognition-primed decision	Time in seconds; fatal error; uncertainty resolved by protocol	Emergency room; firefighters; cockpit in emergency
Critical capsule/existential risk	Emotional regulation under confinement	Fatal and collective error; confinement without exit; synchronization under threat	Submarine; aviation cockpit; rescue; orbital station

Note. Authors’ own elaboration. The five classificatory axes are criteria; the seven typologies are ideal types of collective delivery demand. Stable operation serves as a theoretical contrast, highlighting contexts in which intensifying agility tends to be dysfunctional. Conceptual sources: Simon (1947, 1977); Mintzberg (1978); Klein (1993); Dess and Beard (1984); Knight (1921).

Operationally, typological classification should consider three moderators: environmental munificence; the team’s learning curve and experience; and the availability of Decision Support Systems, from classical DSS to current AI, agent, and multi-agent resources (Dess & Beard, 1984; Gorry & Scott Morton, 1971; Klein, 1993, 1998; Hatano & Inagaki, 1986; Russell & Norvig, 2020; Wooldridge, 2009).

Such systems can expand diagnosis, simulation, alerting, organizational memory, and alternative verification; however, because rational artificial agents operate on performance measures, objective functions, or learned policies rather than on emotions and somatic signals, their contribution depends on sociotechnical integration, governance, and human accountability.

The typologies are ideal types in the Weberian sense: real situations may combine characteristics of more than one class, and their analytical value lies in making the dominant combination explicit.

Teams Under Confinement and Existential Risk: Commitment as a Latent Axis

The taxonomy classifies Collective Delivery Demand (CDD) based on its objective structure. There is, however, a latent axis that becomes dominant in certain teams: members’ existential commitment to the task. Two teams may be equally agile and effective but differ radically when the cost of error falls on the very lives of those who decide. This difference is one of emotional regime, beyond method.

Teams under confinement operate in a capsule, with no immediate exit and individual destiny coupled to the collective: submarines, platforms and orbital stations. Teams under existential risk make the risk of their own lives a daily part of the craft: firefighters, rescue workers, saturation divers, emergency responders. In these cases, collective emotional regulation ceases to be an individual virtue and becomes operational infrastructure.

Verifiable cases delimit the regime. On Air Transat Flight 236 in 2001, the crew glided a powerless aircraft to the Azores while maintaining technical competence under extreme threat (Transportation Safety Board of Canada, 2004). At Three Mile Island in 1979, ambiguous instrumentation, simultaneous alarms, and misinterpretation compromised the operational response and led to partial core meltdown (United States Nuclear Regulatory Commission, 1979). Both cases share the structure of critical response under existential risk but differ in their preservation of collective cognition.

The Tham Luang rescue in 2018 illustrates the overlap of typologies: critical response of the divers, extreme confinement, synchronized execution of submerged transport, and coordination against time and water (Beech et al., 2018). Its success depended on the coupling of teams with distinct specialties, namely technical diving, medicine, pumping, and logistics, and demonstrated that training the team is a condition as necessary as training those who lead it.

To operationalize this axis, it is proposed to classify the team along two diagnostic constructs: degree of confinement, that is, unavailability of exit and coupling of individual destiny to the collective; and error lethality, measured by the extent to which failure falls on the lives of the members themselves.

Crossing these constructs places the team in one of the four quadrants shown in Table 2. The central recommendation is that leadership explicitly name the quadrant before choosing the regime, because treating a high-lethality team as low-lethality, or vice versa, results in a regime error.

Table 2. Confinement and lethality matrix: Team quadrants and leadership conduct.

Quadrant	Confinement	Error Lethality	Examples	Priority Leadership Conduct
Critical capsule	High	High	Submarine in operation; orbital station in emergency	Selection for emotional stability; panic regulation training; role redundancy
Critical exposure	Low	High	Firefighters; rescue; emergency room; aviation	Protocol; structured debriefing; rotation against decision fatigue
Prolonged confinement	High	Low	Long-duration crews; expeditions without immediate risk	Management of cohesion and interpersonal conflict in closed space
Open operation	Low	Low	Administrative teams; software development; construction teams	Open discovery recording; no reinforced emotional infrastructure

Note. Axis of existential commitment. The costliest error on this axis is treating a high-lethality team as if it were low-lethality, or vice versa. Source: authors’ own elaboration.

Axes, typologies, and quadrants serve distinct functions: axes classify the demand; typologies correspond to the seven resulting types; and the quadrants of Table 2 derive specifically from the crossing of confinement and lethality.

6. The Correspondence Matrix: Correlates, Heuristics, Sociotechnical Design, and Cases

The core of the CDD Matrix is the correspondence between each demand typology and four elements: collective neurocognitive correlates, appropriate heuristics, sociotechnical configuration, and typical managerial error. Throughout, the terms are used consistently: an operating regime denotes the team-level configuration of processes, roles, and artifacts through which a demand is met; the neurocognitive correlates (or neurocognitive architecture) denote the individual and collective cognitive substrate that a regime mobilizes; and cognitive regime, where it appears, is treated as a synonym of operating regime and has been harmonized accordingly.

The subsections develop each typology, and Table 3 below adds a fifth operational column: referenced training methods. Reading row by row allows, for a given typology, the identification of the cognitive regime, heuristics, sociotechnical design, errors to avoid, and recommended training.

Table 3. Correspondence matrix: Typologies, correlates, heuristics, and sociotechnical design.

Typology	Neurocognitive Correlates	Appropriate Heuristics	Sociotechnical Configuration	Typical Managerial Error (Key Risks)	Referenced Training Method
Stable operation	Habit (basal ganglia); low prefrontal load; residual vigilance	Standardization; statistical control; exception alerts	Specialized roles; hierarchy; vertical flow	Introducing agility where routine is needed	HRO (Weick & Sutcliffe, 2007)
Iterative discovery	S1/S2 alternation; anterior cingulate; prefrontal updating	Short sprint; retrospective; WIP (Work in Progress) limit	Self-organized team; flat hierarchy; high safety	Fixing the sprint scope (importing planning)	Agile coaching; retrospectives (Schwaber & Sutherland, 2020)
Deadline-driven execution	Dorsolateral executive function; dopamine; moderate HPA	Critical path; milestones; slack; communication rhythm	Multidisciplinary team; overlapping roles; flow management	Treating a fixed deadline as negotiable	AAR (Klein et al., 2005); TeamSTEPPS (AHRQ, 2006)
Synchronized execution	Elevated inter-brain synchrony; motor simulation; anticipatory listening	Internalized score; conductor’s gesture; mutual listening; rehearsal	Conducting hierarchy; sections and sections; gestural and silent communication	Improvising in performance or deliberating during execution	CRM (Helmreich & Merritt, 1998); MBSR (Kabat-Zinn, 1990)
Real-time adversarial	Collective recognition; inter-brain synchrony; motor simulation	Reading the opponent; coded signals; micro-routines	Distributed leadership; gestural and silent communication; recalibration	Over-planning or iterating (losing speed)	AAR (Klein et al., 2005); simulation (Gaba, 2004)
Critical response	Recognition-primed decision; somatic marker; sympathoadrenal axis	Protocol, checklists, standard phraseology; debriefing	Small team; functional hierarchy; pre-trained trust	Deliberating or planning during the crisis	CRM; TeamSTEPPS; clinical simulation (Helmreich & Merritt, 1998; AHRQ, 2006; Issenberg et al., 2005)
Critical capsule/existential risk	Sympathoadrenal axis under prolonged threat; collective emotional regulation; risk of amygdala hijack	Emergency protocol; panic regulation; role redundancy; cross-checking	Selection for emotional stability; redundancy; exhaustive joint training; life history	Underestimating emotional load; failing in training collective panic regulation	AAR; CRM; emotional regulation; extreme simulation (Klein et al., 2005; Helmreich & Merritt, 1998; Gaba, 2004)

Note. Authors’ own elaboration. Neurochemical and structural attributions are construct validations in light of the literature, without causal determination status, per Powell (2011). Somatic marker: Damasio (1994, 1996); dual systems: Kahneman (2011); team learning: Edmondson (1999, 2018); recognition-primed decision: Klein (1993); inter-brain synchrony: Hasson et al. (2012), Reinero et al. (2021); sociotechnical: Cherns (1976, 1987); Pasmore (1988).

6.1. Critical Response: The Recognition-Primed Regime

In critical response situations, the team decides in seconds under acute stress, and an error can be fatal. The appropriate regime is recognition-primed decision-making (Klein, 1993, 1998): the expert reads the situation and immediately activates a consolidated course of action, sustained by somatic marking (Damasio, 1994) and automated recognition in the basal ganglia. Heuristic metabolism operates outside the operation: learning occurs before, through exhaustive training, and after, in debriefing; during the crisis, the repertoire is executed. The typical managerial error is attempting to deliberate or plan during the crisis.

6.2. Deadline-Driven Execution: The Planned Coordination Regime

In deadline-driven execution, the deadline is fixed, there are multiple interdependent fronts, and errors are costly though recoverable. The appropriate regime is planned coordination (Mintzberg, 1978), anchored in dorsolateral executive function and goal-oriented motivation (Locke & Latham, 2002). The central heuristics are critical path, verifiable milestones, strategic slack, and structured communication rhythm. The typical managerial error is importing the open iteration of discovery, treating a fixed deadline as negotiable.

6.3. Iterative Discovery: The Classic Agile Regime

In iterative discovery, requirements are still stabilizing, the work horizon is organized into learning cycles, and failure tends to be reversible and informative. This is the classic domain of the 2001 Manifesto (Beck et al., 2001), provided the demand allows learning during execution. Correlates include flexible alternation between Systems S1 and S2 (Kahneman, 2011), error detection by the anterior cingulate, and prefrontal model updating. Appropriate heuristics are exploratory and iterative: short sprint, retrospective, and work-in-progress limits. Here, heuristic metabolism operates during execution, at each cycle. The typical managerial error is transforming the sprint into a fixed-scope micro-project, importing the planning of deadline-driven execution.

6.4. Real-Time Adversarial: The Intra-Operation Adaptation Regime

In the adversarial typology, there is an active opponent reacting to the team’s actions, time is continuous, and learning occurs in the course of the action itself. The elite sports team is the paradigmatic example. Correlates combine recognition-primed decision-making (Klein, 1993) for individual speed, elevated inter-brain synchrony (Hasson et al., 2012) for coordination under pressure, and motor simulation circuits for anticipating the opponent (Yarrow et al., 2009). Heuristics are adaptive and adversarial: reading opponent patterns, coded signals, reconfigurable micro-routines. The repertoire must be simultaneously automated and reconfigurable in real time, which characterizes adaptive expertise in the full sense (Hatano & Inagaki, 1986). The typical managerial error is over-planning, losing the flow, or opening up to iteration, losing the speed of continuous action.

6.5. Stable Operation: The Optimized Routine Regime

In stable operation, uncertainty is low, repetition is high, and error is predictable and manageable through statistical control. The appropriate regime is an optimized routine, close to Simon’s (1947, 1977) programmed decision: value lies in consistency and variance reduction, with less emphasis on generating new heuristics. Correlates involve high basal ganglia automatization and low prefrontal load. Heuristics are maintenance- and control-oriented. This typology serves as a theoretical contrast: it demonstrates that there are demands for which intensifying agility is dysfunctional, and the typical managerial error is the inverse of the others, importing the discovery regime into a demand that requires routine.

6.6. Synchronized Execution: The Real-Time Coordination Regime

In synchronized execution, time is fixed by a common score; the error is audible and irreversible in the instant; the absence of an opponent and the demand for extreme collective synchronization define the regime. The symphony orchestra is the paradigmatic example, alongside the choir and the ballet company.

The regime requires many performers fused into a single result: the pure expression of the collective brain discussed in Section 3 (Hasson et al., 2012; Reinero et al., 2021). Its correlates combine inter-brain synchrony, motor simulation, and anticipatory listening, with low conscious deliberation during execution. Heuristics are internalized score, conductor’s gesture, mutual listening, and exhaustive rehearsal; the typical managerial error is treating the performance as a rehearsal, opening space for improvisation when the regime requires synchronized fidelity to the common plan.

6.7. Comparative Cases: Team Differences by Specialty

The CDD Matrix instructs leadership and the entire team. Table 4 therefore shows, side by side, how equally competent teams differ in their mode of thinking and acting according to the specialty of the demand they execute. Training a team means developing the repertoire and the emotional regime that its specialty requires, rather than transferring a single method to it.

Table 4. Comparative cases by typology: Specialty, mode of action, and error to avoid.

Typology and Case	Diagnostic and Delivery Specialty	Mode of Thinking, Deciding, and Delivering	Error to Avoid
Deadline-driven execution: large-scale construction	Coordinating interdependent fronts under deadline	Plan the critical path, sequence, and protect milestones with slack	Treating the fixed deadline as if it were open
Critical response: polytrauma emergency care	Rapid stabilization under primary assessment protocol	Recognize the pattern and execute the trained protocol with fixed roles	Deliberating or improvising at the moment of crisis
Critical response: Miracle on the Hudson River (2009)	Emergency piloting and coordinated evacuation	Decide in seconds for a water landing; coordinate the crew without hesitation	Seeking the unreachable airport instead of deciding
Synchronized execution: symphony orchestra	Gestural and silent coordination under a common score	Internalize the score and listen to the ensemble; follow the conductor’s gesture	Improvising or deliberating during the performance
Adversarial (zero-sum): volleyball team	Reading the opponent and recalibration in real time	Anticipate the block through shared reading; coded signals	Over-planning and losing the flow of action
Critical capsule: Apollo 13 (1970)	Technical improvisation under existential risk and confinement	Couple crew and mission control; use the lunar module as a lifeboat	Underestimating interdependence and time pressure
Critical capsule: Tham Luang rescue (2018)	Coupling of teams with distinct specialties	Integrate diving, medicine, and logistics under confinement and time	Underestimating emotional load and interdependence

Note. Authors’ own elaboration. Real and illustrative cases. Miracle on the Hudson (National Transportation Safety Board, 2010); Apollo 13 (Kranz, 2000); Tham Luang rescue (Beech et al., 2018).

7. The New Manifesto: Principles as Correspondence Rules

The 2001 Manifesto enunciates still-valuable preferences: individuals and interactions, working software, collaboration, and response to change. The New Manifesto preserves this legacy and reformulates it as correspondence rules: if the demand has structure X, adopt regime Y. The shift from universal rules to conditional rules makes the principles sensitive to the typology of the collective delivery demand at work.

To make the link between the CDD Matrix and the New Manifesto explicit: the Manifesto is not a second, independent artifact but the normative reading of the matrix. Each row of the matrix—a demand typology with its corresponding regime, heuristics, sociotechnical configuration, and characteristic error—translates into a conditional prescription of the form “if the demand has structure X, prioritize regime Y and guard against error Z.” The common base (Section 7.1) states the principles that hold across all rows, because they follow from properties that every demand shares; the differentiated layer (Section 7.2) states the priorities that change from row to row, because they follow from the axis values that distinguish one typology from another. In this sense, the typologies do not merely coexist with the Manifesto—they are what modulate it: moving along the axes (raising error cost, adding adversariality, tightening time pressure) shifts normative weight from one principle to another, so that the same spirit yields different operative priorities for a discovery team, a critical-response team, and an adversarial team. This conditional structure mirrors recent reviews reporting that agility’s contribution to outcomes is contingent on dimension and context rather than uniform (Nguyen et al., 2025). The Manifesto thus inherits its conditional structure from the matrix, and the matrix acquires its prescriptive force from the Manifesto.

7.1. The Common Base: Principles Valid for Every Team

The first level of the New Manifesto is the common base: seven principles that apply to every team, regardless of demand typology. This is the shared trunk on which subsequent differentiation rests.

First. Diagnose the delivery demand typology before choosing the regime. Leadership’s opening question shifts from which methodology to adopt to which combination of time pressure, error cost, uncertainty, adversariality, and munificence the demand presents. Method follows diagnosis, and regime follows the required delivery.

Second. Make the regime correspond to the demand rather than artificially adjusting the demand to the organization’s preferred regime. Applying open iteration to a critical response is as dysfunctional as applying a rigid protocol to a discovery demand. The costliest error lies in the incompatibility between regime and typology.

Third. Cultivate heuristic metabolism as a central asset. The enduring value of a team lies in generating, sharing, and refining heuristics appropriate to the demand typology, and leadership protects the conditions, above all psychological safety, that convert reversible errors into collective knowledge.

Fourth. Modulate the environment, because it modulates the collective brain. Predictability, goal clarity, interpersonal safety, and sustainable rhythm are regime variables that condition the sympathoadrenal and HPA axes, collective executive function, and inter-brain synchrony.

Fifth. Adjust the learning cadence to the typology. In critical response, learning occurs before and after; in discovery, during each cycle; in adversarial, during action; in stable operation, in exception monitoring. The reflection cadence varies with demand.

Sixth. Jointly optimize technical and social subsystems. Both are subject to design, and trust is the lubricant that makes this optimization viable.

Seventh. Treat stability as a legitimate regime rather than a sign of low agility. There are demands whose optimal regime is optimized routine, and recognizing this is part of the CDD Matrix’s maturity.

7.2. The Differentiated Layer: Priorities by Typology

The concrete application of principles is differentiated by typology. In critical response, priority is given to protocol discipline and panic regulation; in deadline-driven execution, to milestone commitment and interdependent coordination; in iterative discovery, to experimentation and welcoming change; in real-time adversarial, to situational reading and distributed authority; in synchronized execution, to fidelity to the common score and mutual listening; in stable operation, to consistency and variance reduction. When existential commitment intensifies, preserving collective emotional regulation takes precedence over all other values.

Table 5 synthesizes this normative architecture at two levels: the common base and the differentiated layer by demand typology.

Table 5. The New Manifesto in Two Levels: Common Base and Differentiated Layer by Demand Typology.

Typology	Differentiated Layer: Normative Priority
Stable operation	Consistency and variance reduction over innovation
Iterative discovery	Experimentation and welcoming change (original agile normative field)
Deadline-driven execution	Milestone commitment and interdependent coordination over scope renegotiation
Synchronized execution	Fidelity to the common score and mutual listening over individual initiative
Real-time adversarial (zero-sum)	Situational reading and distributed authority over prior planning
Critical response	Protocol discipline and panic regulation over open deliberation
Critical capsule/existential risk	Transversal clause: preserve collective emotional regulation, which precedes all other values

Note. Authors’ own elaboration. The first row assembles the common base; subsequent rows specify the differentiated layer, the value that receives priority in each demand typology; the last states the transversal clause, applicable to teams in which existential commitment intensifies. Common base, valid for every team: (1) diagnose the demand typology before choosing the regime; (2) make the regime correspond to the demand, not the reverse; (3) cultivate heuristic metabolism as a central asset; (4) modulate the environment, which modulates the collective brain; (5) adjust the learning cadence to the typology; (6) jointly optimize the technical and social subsystems; (7) treat stability as a legitimate regime.

7.3. When the Regime Fails: Substitution, Recomposition, and Protection of Collective Performance

The New Manifesto also specifies what to do when the regime breaks down. Adaptive leadership ensures alignment among demand typology, objectives, goals, action plans, accountable parties, deliverables, protocols, and synchronization. In the logic of the Balanced Scorecard, objectives coordinate collective action only when they unfold into goals, initiatives, accountable parties, and verifiable indicators (Kaplan & Norton, 1996). When a member or the team itself repeatedly compromises these elements, substitution, removal from the operation, or recomposition ceases to be a punitive choice and becomes a measure to protect collective performance.

This monitoring is team risk management and varies according to regime governance. In reversible-error typologies, self-managing teams can operate with empowerment, self-correction, and peer arbitration. In high-error-cost or critical capsule typologies, internal autonomy must coexist with defined command, external accuracy management, auditing, arbitration, and DSS or AI resource support. Algorithms, agents, and multi-agents can function as sensors, simulators, consistency checkers, case memory, and alerting mechanisms, with explicit restriction on unrestricted control in existential contexts: they operate without emotions, through performance functions and decision policies, and require veto rights, interruption criteria, and human accountability.

Firm intervention is justified when observable conduct degrades the regime: intoxication, loss of control, contagious panic, recurrent aggression, protocol disobedience, timing breakdown, sabotage of psychological safety, repeated non-fulfillment of feasible goals, or rupture of critical-path processes. The higher the error cost, time pressure, interdependence, and irreversibility, the lower the tolerance for collective disorganization should be. In extreme cases, failure becomes embedded in the team’s pattern, requiring role recomposition, an intermediate leadership change (contingency leadership), or replacement of the operational configuration.

7.4. EAD² Culture: Six Traits and Referenced Training Methods

The normative foundation of the New Manifesto presupposes a team culture that sustains it. The culture of Agile High-Performance and Dynamic Teams (EAD²) is recognized by six traits that operate as collective habits rather than mere declared values.

These six traits are: primacy of delivery over ritual; psychological safety practiced; heuristic metabolism as habit; orientation toward verifiable goals and results; reconfiguration as norm; and collective discernment of the correspondence between demand and regime. Together, they cause the team to judge its choices by what they deliver, expose reversible errors, convert experiences into shared heuristics, articulate objectives as public goals, adjust roles and protocols as demand changes, and signal early when the current regime no longer corresponds to the typology.

These traits depend on continuous training, standardization, and acculturation. The high-performance team literature converges on methods mappable to CDD Matrix typologies. The After Action Review (AAR), systematized by Klein et al. (2005), anchors heuristic metabolism in critical response and the critical capsule. Crew Resource Management (CRM), formalized by Helmreich and Merritt (1998) and evaluated by Salas et al. (2006), trains communication, hierarchical assertiveness, and checklists in aviation, surgery, and critical response. Mindfulness-Based Stress Reduction (MBSR), by Kabat-Zinn (1990), supports stress regulation under load. TeamSTEPPS, a team-training program developed by the Agency for Healthcare Research and Quality (AHRQ, 2006), integrates psychological safety, structured communication, and debriefing within healthcare teams. High Reliability Organizations (HROs), by Weick and Sutcliffe (2007), promote a culture of attention to minimal deviations, reluctance to simplify, and deference to expertise. Agile Coaching with retrospectives (Schwaber & Sutherland, 2020) sustains learning through iterative discovery. Navigating these cultural shifts is a recognized challenge in agile adoption, as the transition often exposes structural frictions within traditional management hierarchies (Perides & Vasconcellos, 2025). Therefore, developing absorptive and adaptive capacities at the team level is essential not merely for innovation, but to sustain these new operating regimes under pressure (Singh & Jha, 2024).

Realistic simulation and permanent training have a central role in emergency rooms, rescue, aviation, and critical capsules, since they create repertoires that activate automatically under pressure (Gaba, 2004; Issenberg et al., 2005). With each new technology incorporated into DSS, whether embedded AI, sensors, wearables, or collaborative platforms, the team must recalibrate heuristics and usage protocols, on pain of turning decision support into a new source of bias.

Standards, excellence models, maturity systems, certifications, and accreditations also function as technologies for standardizing and acculturating the regime. ISO 9001 and ISO 30401 offer institutional languages for quality management, knowledge management, documentation, auditing, continuous improvement, and evidence-based decision-making. Malcolm Baldrige, EFQM, CMMI, and World Class Manufacturing structure routines for evaluation, organizational learning, process maturity, and operational excellence. In healthcare, the Joint Commission International and ONA translate this logic into standards for quality of care, patient safety, integrated management, and operational reliability. Collectively, these mechanisms directly affect team performance by reducing variability, making responsibilities explicit, stabilizing protocols, orienting records, qualifying debriefings, and converting tacit heuristics into verifiable standards (ISO, 2015, 2018; CMMI Institute, 2018; JCI, 2021; ONA, 2022).

A technically competent team that is culturally misaligned with the regime tends to reproduce the typical managerial errors listed in Table 3. Acculturation is a first-order leadership function, prior to method selection and team composition.

8. Theoretical Propositions and Empirical Agenda

From the CDD Matrix, six testable propositions are derived and stated in conditional form to guide future empirical research.

P1. When a team’s operating regime corresponds to the typology of the delivery demand it executes, superior performance is expected compared to teams where regime and typology are incompatible, provided individual competence and resources are controlled. The reason is that the appropriate regime mobilizes the neurocognitive configuration necessary for delivery, while incompatibility forces the team to operate in a mode ill-suited to the demand’s structure.

P2. When the rate of accumulation and refinement of a team’s shared heuristics is measured through shared mental model instruments, that rate is expected to better predict adaptive performance than the sum of individual experience or the nominally adopted method. The reason is that adaptive intelligence resides in the circulating collective repertoire, above the sum of individual competencies and the method’s label.

P3. When a team has high psychological safety, the relationship between committing errors and improving performance is expected to be stronger, with a larger effect in discovery and adversarial typologies than in critical response and stable operation. The reason is that discovery and adversarial learning occur within the operation, where safe error exposure feeds iteration, while critical response and stable operation shift learning outside it.

P4. When the iterative discovery regime is applied to a critical response demand, performance degradation is expected, since the open deliberation that discovery requires overloads cognition under the extreme time pressure of emergency, precisely when recognition-primed decision-making would be the appropriate regime.

P5. When a team is subjected over an extended period to a regime inappropriate for the typology of demand it executes, chronically elevated HPA axis activation is expected to predict, over time, a decline in collective creativity, cooperation, and member retention, due to the documented effect of elevated cortisol on cognitive flexibility and social cooperation.

P6. When a team’s operating regime begins to degrade, explicit mechanisms of monitoring, arbitration, interruption authority, and recomposition are expected to better preserve collective performance than maintaining the dysfunctional configuration, especially in typologies of high error cost, high interdependence, and irreversible consequences. The reason is that regime failure requires risk governance that combines self-management, internal command, and, when criticality demands it, DSS support, AI agents, multi-agent systems, external accuracy management, and arbitration.

The empirical agenda combines quasi-experimental field designs comparing teams across different typologies, with process measures such as psychological safety (Edmondson, 1999) and shared mental models (Cooke et al., 2013), and, when appropriate, safe and non-invasive neurophysiological measures, always subject to Powell’s criteria.

More specifically, the propositions suggest complementary designs. P1 and P4 lend themselves to quasi-experimental or vignette/simulation studies that manipulate regime–typology (mis)match and measure team performance while controlling for individual competence and resources. P2 and P3 call for multi-wave field studies that track the accumulation of shared heuristics (via shared-mental-model similarity and accuracy) and test psychological safety as a moderator of the error–learning link across typologies. P5 requires longitudinal designs relating sustained regime–typology misfit to strain indicators (e.g., validated stress and burnout measures and, where ethical and non-invasive, HPA-axis markers) and to creativity, cooperation, and retention outcomes. P6 is suited to comparative case and process studies of teams undergoing regime degradation, contrasting the presence versus absence of monitoring, arbitration, interruption authority, and recomposition mechanisms. Across all designs the team is the primary level of analysis, and neurophysiological measures remain auxiliary and subordinate to Powell’s (2011) criteria.

9. Limitations Under Powell’s Filter

Epistemological honesty requires making limits explicit, following Powell’s (2011) warnings. Regarding the mereological fallacy, expressions such as “team brain” and “collective brain” were used as abbreviations; strictly speaking, people and teams decide and learn, while brains sustain processes. Regarding reverse inference, the associations between neural systems and coordination functions were presented as construct and plausibility validation, never as proof that a regime derives from an activation; correlational studies counsel caution regarding causal claims about organizational behavior.

Regarding explanatory seduction (Weisberg et al., 2008), much of the CDD Matrix’s recommendations are justified in classical behavioral terms, with neuroscience’s role being to elucidate mechanisms rather than solely underpin the prescription. It should be added that the inter-brain synchrony literature is recent and requires replication, and that the prescriptive-conceptual character of the CDD Matrix requires longitudinal designs to test its propositions, which the present essay lacks. Recognizing these limits precisely delimits what the CDD Matrix prescribes. In practical terms, the framework warrants diagnostic and design guidance (matching regime to demand and protecting the conditions for learning), but not causal or predictive claims about outcomes, which require the empirical tests set out in Section 8.

10. Recommendations for Teams and Leadership

The CDD Matrix translates into practical guidance for leadership and all team members. Three recommendations synthesize the model: diagnose demand across the five axes before choosing the method; compose and allocate people by regime correspondence, recognizing that those who excel in discovery may underperform in critical response; and protect psychological safety in discovery and adversarial typologies, and cultivate collective emotional regulation in existential-risk typologies. Added to these is a transversal imperative: maintain permanent training and update the heuristic repertoire with each technological innovation incorporated into the sociotechnical environment.

In operational terms, these recommendations imply a repeatable leadership routine. Before a delivery, leaders diagnose demand across the five axes, identify the dominant typology, and select the corresponding operating regime rather than defaulting to a familiar method. During composition, they match people to the regime—pairing those who thrive in open-ended discovery with iterative work and those who excel under time pressure and high error cost with critical-response work—and make interruption authority and decision rights explicit. During execution, they protect psychological safety, where learning happens within the operation (discovery and adversarial typologies), and cultivate collective emotional regulation, where commitment under existential risk prevails. After delivery, they conduct a structured debrief (for example, an After-Action Review) to convert reversible errors into shared heuristics, and they recalibrate those heuristics whenever a new technology enters the sociotechnical environment. When indicators show that the regime no longer meets demand, leaders treat regime substitution, recomposition, and the protection of collective performance as a governance decision rather than a personal failure.

11. Concluding Remarks

The 2001 Agile Manifesto was a creative response to a real delivery demand, and its success carried its principles far beyond the original domain. This essay preserves that legacy but proposes an additional conceptual step: treating the New Agile as a correspondence theory between Collective Delivery Demand (CDD) and team operating regimes. The practical consequence is direct: agile teams can promote adaptation, but high-performance teams also require effectiveness, efficiency, safety, learning, and sociotechnical coherence.

The CDD Matrix articulated the classical decision structure of Simon and Mintzberg; the dynamic capabilities literature of Teece, Pisano, and Shuen; Klein’s naturalistic decision-making; Edmondson’s team learning; Powell’s safeguard; decision neuroscience; and the sociotechnical tradition. The result is a correspondence theory for EAD², Agile High-Performance and Dynamic Teams: distinct demands require distinct regimes, and team maturity manifests as the capacity to diagnose the delivery, mobilize appropriate heuristics, draw on referenced training methods, and reconfigure resources, roles, and decision supports.

The intended contribution is to offer a criterion for discerning which regime each demand requires and how the team must be structured to sustain it. This discernment, more than adherence to a method, distinguishes the truly adaptive team. The New Manifesto shows that teams should be designed based on CDD typologies, objectives, goals, risks, and accountabilities. Two imperatives close the essay: permanent capability development, since the heuristic repertoire degrades if not exercised, revised, and expanded through training, simulation, and debriefing; and attention to technological innovations, since each new DSS, AI agent, platform, or sensor requires recalibration of heuristics and protocols. The dynamism of digital transformations does not threaten the model; it is the condition that makes it necessary.