Kevin Dickerson.

Regenerative Intelligence

The Nine Capacities of a Regeneratively Intelligent Company

A working synthesis of organizational psychology, complexity science, and the post-2020 research on hybrid work and AI — assembled into a framework practitioners can deploy and refine.

By Kevin Dickerson 18 min read Framework Introduction Nine Capacities

An abstract nine-capacity architecture on textured paper: nine connected chambers, each holding a vividly patterned fantastical Bay Area alebrije made from Alameda-area animals and native edible plants.
On this page7
  1. Extractive, sustainable, regenerative
  2. Four modes, four clocks
  3. The nine capacities
  4. The peer-reviewed grounding
  5. How we think about structural levers
  6. What this framework is, and what it isn't
  7. How to read this arc

Alex Edmans tracked the stock returns of the “100 Best Companies to Work For” against the broader market for 28 years.1 Adjusted for industry, factor exposure, and standard controls, they outperformed by 2.3 to 3.8 percent per year. Compounded over a working lifetime, that’s the difference between a portfolio that doubles and one that quadruples.

The signal is consistent across multi-decade horizons. The mechanism is contested in the literature. What the framework that follows does is synthesize how we think about what produces that signal — at the organizational scale, in environments of continuous perturbation, with the same first principles that Adaptive Resilience operates on at the personal scale.

The frameworks of corporate strategy most managers learned were written for a market that no longer reliably exists — environments that could be perturbed and returned to, customers whose preferences shifted slowly, employees whose tenure compounded across decades. Each of those assumptions is now wrong in most industries most of the time. What we have instead is continuous perturbation. Not crisis — crisis is a moment a system passes through; continuous perturbation is the steady state. The academic literature has converged on the term polycrisis to describe the same condition: multiple crises causally entangled rather than merely co-occurring, where the interactions between them generate effects no single crisis would produce.2 Companies that scale by extracting surplus from people, customers, and ecosystem inputs run out of substrate. Companies that optimize for stability against the current backdrop optimize for the wrong target.

We call it Regenerative Intelligence. It is the organizational-scale analog of Adaptive Resilience — the personal-scale framework I’ve developed at this site over the last several years. Both are drawn from the same first principles. Both apply to systems operating under sustained pressure rather than stable conditions. RI extends the pattern from individuals to organizations.

A company operates regeneratively to the extent that its ordinary practices — the things it does every day to ship product, serve customers, hire and promote people, decide what to build — increase the capacity of every system they touch. A regeneratively intelligent company makes its employees more capable over time, not less. It builds customer capability rather than locking customers in. It strengthens supplier ecosystems rather than squeezing margin from them. It returns surplus to the systems that produced it.

None of this is sentimental. It is structural. The company that does these things has a substrate that compounds. The company that doesn’t has a substrate that depletes.

Extractive, sustainable, regenerative

Three positions a company can take relative to its substrate:

Extractive. Surplus is captured outward as rents. Over time, the substrate degrades. The company grows by depletion. This is the default for late-stage capitalism — the model most companies operate within whether they consciously chose it or not.

Sustainable. Surplus is maintained at parity. The substrate is held constant. This is the framing of most ESG and corporate sustainability work — net-zero impact, no degradation. It is treading water, indefinitely.

Regenerative. Surplus flows back into capacity. The substrate gets richer over time. The company and its ecosystem grow together.

The test is structural. Where does the surplus go? An extractive company sends it out as rents. A sustainable company maintains it. A regenerative company reinvests it — in employees, in customers, in suppliers, in learning, in the substrate the company depends on for its own continued operation.

You can run this test on any company without asking how it markets itself. The answer is in the audited cash flow statement, the equity ownership structure, the long-term retention and development of employees, the customer net-promoter trend, the supplier longevity, the open publication record. Words don’t matter; structure does.

Four modes, four clocks

In 2022, Yvon Chouinard converted ownership of Patagonia into a perpetual-purpose trust, directing the company’s profits into a permanent climate fund. The structural move was the culmination of nearly fifty years of practice: 1% of sales donated to environmental causes since 1985, an all-organic-cotton supply chain since 1996, the highest employee retention in its category for over two decades. Industry analysts had called the early commitments a death sentence. By 2022, Patagonia was generating over a billion dollars in annual revenue and had built one of the most durable consumer brands in retail.

In the same five decades, dozens of fast-fashion competitors operating on the opposite logic produced spectacular short-term returns and then collapsed. Forever 21 filed for bankruptcy in 2019 after two decades of squeezing every link of its supply chain. The substrate it built its empire on — workers paid below minimum wage in unsafe factories — was depleted.

Two companies. Same industry. Same window. Different positions on a 2x2 that explains the divergence.

Extractive
Regenerative
High Intelligence
High Intelligence · Extractive

The Optimizer

Wins this decade. Loses the next.

High Intelligence · Regenerative

The Gardener

Wins on every horizon.

Low Intelligence
Low Intelligence · Extractive

The Strip-Miner

Wins until the bill comes due.

Low Intelligence · Regenerative

The Idealist

Outlasts the Optimizer.

Four modes of organizational operation. The Gardener compounds across decades; the other three each run on a clock.

Two questions place any company in one of the four cells:

  1. Where does surplus go? Back into the substrate (regenerative), or out as rents (extractive).
  2. How clearly does the company see its system? Accurately, or distorted by identity-as-blindfold, motivated reasoning, and short-horizon pressure.

The four modes describe companies most readers can name.

The Strip-Miner is Forever 21 in its final decade — extraction without awareness, optimization for current conditions, no model of substrate depletion until the substrate is gone. Legacy extractive industries operate here by inertia. The clock is biological: substrate runs out.

The Optimizer is the 3G Capital playbook at Kraft-Heinz — sharp sensemaking, sophisticated cost-out, optimization with full awareness that the substrate is what’s being captured. Effective in the medium term; Kraft-Heinz absorbed a $15 billion writedown in 2019 when the strategy hit its limits. The clock is regulatory and social: the bill comes due when externalities become visible.

The Idealist is TOMS Shoes in the late 2010s — strong regenerative intent (the celebrated “buy one, give one” model), insufficient systems understanding of the local economies the donations were affecting. The company pivoted in 2019, publicly acknowledging that the donation model wasn’t delivering the impact it had promised. The clock is financial: generosity without diagnosis runs out of runway.

The Gardener is Patagonia. Sees the system clearly, reinvests in substrate, builds capacity that compounds across decades. The clock doesn’t run.

The Edmans signal is the strongest empirical anchor. Henderson and Mayer have built the theoretical case for why employee-aligned companies outperform.34 Bloom’s 2024 hybrid-work RCT supplies a clean modern demonstration: a 33% reduction in attrition with no performance loss when a company structures its workplace policy regeneratively.5

The skeptical reading has been argued most carefully by Lucian Bebchuk and Roberto Tallarita: the outperformance is selection rather than causation, and stakeholder governance is largely performative.6 Take it seriously — and then notice that even on the skeptical reading, the long-horizon outperformance is still there. Better companies attract better stakeholders, which still compounds. The mechanism is contested. The signal is not.

The structural question is separate from the empirical one. Whether a company should organize regeneratively, even before the empirical comparison resolves, is closer to a values commitment than a prediction. The framework holds both open: it uses settled science as settled, engages contested findings as contested, and labels normative claims as normative rather than smuggling them in.

The trap most companies are stuck in is mistaking the Y-axis for the strategic axis. AI sharpens sensemaking, memory, and coordination — real upgrades to the Y. But if a company remains structurally extractive on the X, AI only makes the extraction more efficient. A faster Strip-Miner is still a Strip-Miner. An Optimizer with a frontier model is still on the same trajectory; it just executes it sooner. The clock changes; the direction doesn’t.

The path is right before up. Most companies try to move UP first — acquire intelligence by adopting AI, hiring consultants, getting sophisticated. Then RIGHT later, becoming more regenerative once they can afford to. Patagonia did the opposite. The regenerative commitments came first, in the 1980s; the intelligence systems to sustain them got built over the four decades after. The Gardener is the upper-right destination, and the path to it almost always runs through the bottom-right first. Disposition before strategy.

The nine capacities

A company operates regeneratively to the extent that it deliberately cultivates nine capacities. They are listed here in narrative order — the developmental sequence in which they tend to need to be established — not in order of priority.

1. Safety. Can people speak truth without penalty? Most organizations treat psychological safety as the ceiling — the destination, the goal to achieve. It is actually the floor. Without it, sensemaking is corrupted, variation is suppressed, and memory is unreliable; none of the other eight capacities can operate. Amy Edmondson’s three-decade research program at Harvard established the construct,78 and Google’s Project Aristotle confirmed the prediction at scale. The strongest recent meta-analysis (Frazier et al. 2017) refines this:9 the effect is real but conditional on team composition, task type, and leader behavior. Safety is necessary but not sufficient — and treating it as sufficient is its own kind of failure.

2. Identity. Does the company know who it is clearly enough to filter signal? Identity acts as a cognitive filter on what the company can perceive. The classic case is Mary Tripsas’s Organization Science paper on Polaroid:10 the company’s identity as “the instant photography company” prevented it from acting on the digital-photography evidence its own researchers had collected. Stuart Albert and David Whetten first defined identity as a research construct;11 the most active modern line is Wendy Smith and Marianne Lewis on paradox — the insight that high-performing organizational identities are not stable singletons but dynamic equilibria.12

3. Sensemaking. Are we perceiving the environment accurately? Karl Weick’s foundational work on organizational sensemaking established the field.13 Gary Klein on premortem analysis operationalized one of its most practical applications.14 The COVID era added a new dimension: Marlys Christianson and Michelle Barton on sustained-crisis sensemaking — the cognitive demand of continuous re-perception when uncertainty doesn’t resolve.15 And the post-2023 dimension: Fabrizio Dell’Acqua and colleagues’ “jagged frontier” research on knowledge workers using generative AI,16 which shows the company now needs a continuously updated map of where AI extends perception and where it actively harms it.

4. Variation. Are we generating real options rather than converging to the mean? C.S. Holling’s adaptive cycle framework establishes the principle:17 regenerative systems pass through exploitation → conservation → release → reorganization. Eric Beinhocker’s Origin of Wealth applied complexity economics to companies;18 W. Brian Arthur’s work provides the rigorous foundation.19 The practice is running multiple bets concurrently, embracing deliberate constraints, and treating the production of variety as a first-order practice — not an afterthought.

5. Coordination. Are we aligned without bureaucracy crushing speed? Sandy Pentland’s MIT social-physics work shows that patterns of communication predict team performance better than IQ or experience.20 Edward Deci and Richard Ryan’s Self-Determination Theory establishes the substrate of motivation — autonomy, competence, and relatedness as the durable inputs.21 Longqi Yang and colleagues’ 2022 natural experiment on 61,000 Microsoft employees during the COVID transition shows what happens when network structure changes: cross-team networks become more siloed and static.22 Coordination after hybrid work cannot be the same as coordination before it.

6. Memory. Do we learn from our work, or relitigate? Chris Argyris on double-loop learning identified the failure mode — companies that learn within their existing frame but never question the frame.23 Peter Senge’s Fifth Discipline synthesized the practice.24 Ikujiro Nonaka and Hirotaka Takeuchi on tacit knowledge added the dimension of what cannot be written down but can be transmitted.25 The newest dimension comes from Erik Brynjolfsson, Danielle Li, and Lindsey Raymond’s 2025 Quarterly Journal of Economics paper showing that generative AI can extract expert tacit knowledge and distribute it to novices.26 This is a fundamental change in what organizational memory can be.

7. Stewardship. Where does our surplus go? This is the regenerative test, made operational. Bill Reed’s foundational paper on regenerative design and Daniel Wahl’s Designing Regenerative Cultures establish the conceptual frame.2728 The modern empirical case is in Henderson, Edmans, and Mayer, cited above. The skeptical counter is Bebchuk and Tallarita, cited above. One of the few clean modern demonstrations is Nicholas Bloom’s hybrid-work RCT: a 33% reduction in attrition with no performance loss when a company structures its workplace policy regeneratively.5

8. Generativity. Does our work build capacity in our ecosystem? Mark Granovetter’s 1973 paper on the strength of weak ties is the foundation;29 Ron Burt’s structural holes work updates it for the modern company.30 Damon Centola’s How Behavior Spreads sharpens the model — weak ties carry simple contagions; complex contagions need redundant social reinforcement.31 The practice involves open thought leadership, structured customer-developmental relationships, and treating the company as a node in a learning network rather than an island.

9. Cadence. Do we pace through expansion–release cycles, or expand until breaking? Lance Gunderson and C.S. Holling’s panarchy framework — coordinated adaptive cycles operating at different time scales — provides the model.32 Christina Maslach and Michael Leiter’s work on burnout shows what happens at the company scale when cadence is ignored: cumulative exhaustion that no individual choice can fix.33 Cadence is the least-developed of the nine capacities in the literature and one of the most important in practice.

The peer-reviewed grounding

The framework draws from roughly seventy peer-reviewed sources, organized in the Foundations of Regenerative Intelligence corpus on this site. The corpus deliberately includes papers that ground the framework’s claims and papers that contest them: Bebchuk and Tallarita’s skeptical case on stakeholder capitalism sits next to Henderson, Edmans, and Mayer. Frazier’s meta-analytic conditional on psychological safety sits next to Edmondson. Dell’Acqua’s “jagged frontier” sits next to Brynjolfsson’s productivity gains. The inclusion of skeptical work is part of the work, not a hedge against it.

Five paradigms anchor the framework, each with decades of replication:

  • Psychological safety (Edmondson plus the post-COVID hybrid-work updates)
  • Adaptive cycle theory (Holling’s panarchy and its extensions)
  • Sensemaking (Weick, Klein, Christianson & Barton)
  • Self-Determination Theory (Deci & Ryan on motivation and autonomy)
  • Social physics and network science (Pentland, Burt, Centola)

These are the rigor floor. The framework draws from the small set of researchers whose work has either replicated meta-analytically or held up across paradigm shifts. It does not draw from the popular business literature.

How we think about structural levers

A framework that depends entirely on cultural commitment will drift back to extractive practice the moment cultural attention moves on. The question isn’t whether structure matters — it does — but which levers are actually available to you given the entity, governance, ownership, and donor-or-investor constraints already in play.

Most readers don’t get to choose their company’s legal structure. What they can do is study the levers that exist in this structure and pull the ones that make the regenerative choice the easier choice over time. Different forms have different levers:

  • Governance levers (board composition, charter language, decision-rights documents) — available in nearly every structure
  • Ownership levers (employee equity, ESOPs, profit-interest units, cooperative member shares, perpetual-purpose trusts) — vary by entity type
  • Fiduciary-duty levers (Public Benefit Corporation charters, B-Corp certification, mission-lock provisions) — available where the legal entity supports them
  • Tax-code levers (Section 41 R&D credits, charitable contribution structures, donor-advised funds, fiscal sponsorship, opportunity zones) — vary by entity status and jurisdiction
  • Contract levers (operating agreements, employment-agreement clauses, supplier agreements, IP licensing patterns) — available everywhere

The discipline isn’t picking one. It’s mapping which levers are actually available given the company’s constraints, then deciding which combination addresses the specific failure modes that values-only commitments produce. A Delaware PBC is one strong move; a worker-cooperative member-equity structure is a different strong move; a donor-funded research org operates with mission-locked grants that produce yet another set of levers. The framework’s contribution isn’t the recommendation. It’s the discipline of asking what’s possible here, and which of what’s possible is worth pulling for which capacity.

What this framework is, and what it isn’t

This is a working synthesis, intended as a practitioner’s guide.

It is not a peer-reviewed scientific paper. It does not claim to have established new findings. It synthesizes existing findings into a coherent framework for action.

It is not consulting jargon. Every claim has a citation, and the citations are in the Foundation. Frameworks built without a citable foundation are doing performance, not work.

It is not a values manifesto disguised as research, nor research disguised as a values manifesto. The framework distinguishes among different kinds of claims:

  • Empirical predictions — “psychological safety predicts learning behavior” — have replicated evidence behind them. The framework states them at the confidence level the literature supports.
  • Operational principles — “surplus should flow back into capacity” — guide action. They are coherent arguments and values commitments, not empirical predictions.
  • Architectural choices — “Public Benefit Corporation plus ESOP encodes the operational principle in law” — constrain future drift. They are design decisions with rationales.
  • Heuristics — “safety is necessary but not sufficient” — compress experience. They are guides that hold across most but not all cases.

This taxonomy matters because most contested-empirics debates are debates about the first kind. The framework also uses the other three. Distinguishing the kinds is the actual rigor move.

How to read this arc

Nine capacity essays follow as they clear review, each structured the same way: research, what it does, how we do it. Each stands alone. The framework as a whole emerges from reading several.

Practitioners: as the essays publish, start with the capacity most relevant to your current bottleneck; the closing essay’s diagnostic will help you assess where each capacity sits in your operation. Scholars: the Foundation is where to start; every citation there resolves to a peer-reviewed source. Critics: contested empirics will live in the body of each essay, not a disclosure section. Engage them where they live.

The mapping to Adaptive Resilience — the personal-scale framework these draw their first principles from — is not one-to-one. Some AR domains map cleanly to RI capacities. Some don’t. Companies need explicit Coordination and Memory practices in ways individuals don’t, because brains handle both internally. The closing essay will work the crosswalk in full.

Try this week — Locate your company on the four-mode matrix above. If you’re not the Gardener, name what’s pulling you toward another cell — and which capacity’s failure mode best describes that pull. That capacity is where the work starts.

References

  1. Edmans, A., Grow the Pie: How Great Companies Deliver Both Purpose and Profit, Cambridge University Press, March 2020, ISBN 978-1-108-49926-2. ↩

  2. Lawrence, M., Homer-Dixon, T., Janzwood, S., Rockström, J., Renn, O., and J. F. Donges, “Global Polycrisis: The Causal Mechanisms of Crisis Entanglement”, Global Sustainability, vol. 7, Article e6, DOI 10.1017/sus.2024.1, February 2024. ↩

  3. Henderson, R., Reimagining Capitalism in a World on Fire, PublicAffairs, April 2020, ISBN 978-1-5417-3015-1. ↩

  4. Mayer, C., Prosperity: Better Business Makes the Greater Good, Oxford University Press, November 2018, ISBN 978-0-19-882400-8. ↩

  5. Bloom, N., Han, R., and J. Liang, “Hybrid Working from Home Improves Retention Without Damaging Performance”, Nature, vol. 630, no. 8018, pp. 920–925, DOI 10.1038/s41586-024-07500-2, June 2024. ↩ ↩2

  6. Bebchuk, L. and R. Tallarita, “The Illusory Promise of Stakeholder Governance”, Cornell Law Review, vol. 106, no. 1, pp. 91–177, December 2020, scholarship.law.cornell.edu/clr/vol106/iss1/2. ↩

  7. Edmondson, A., “Psychological Safety and Learning Behavior in Work Teams”, Administrative Science Quarterly, vol. 44, no. 2, pp. 350–383, DOI 10.2307/2666999, June 1999. ↩

  8. Edmondson, A., The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth, Wiley, November 2018, ISBN 978-1-119-47724-2. ↩

  9. Frazier, M., Fainshmidt, S., Klinger, R., Pezeshkan, A., and V. Vracheva, “Psychological Safety: A Meta-Analytic Review and Extension”, Personnel Psychology, vol. 70, no. 1, pp. 113–165, DOI 10.1111/peps.12183, Spring 2017. ↩

  10. Tripsas, M., “Technology, Identity, and Inertia Through the Lens of ‘The Digital Photography Company’”, Organization Science, vol. 20, no. 2, pp. 441–460, DOI 10.1287/orsc.1080.0419, April 2009. ↩

  11. Albert, S. and D. Whetten, “Organizational Identity”, in L. Cummings and B. Staw (eds.), Research in Organizational Behavior, vol. 7, JAI Press, pp. 263–295, 1985, ISBN 978-0-89232-497-2. ↩

  12. Smith, W. and M. Lewis, “Toward a Theory of Paradox: A Dynamic Equilibrium Model of Organizing”, Academy of Management Review, vol. 36, no. 2, pp. 381–403, DOI 10.5465/amr.2009.0223, April 2011. ↩

  13. Weick, K., Sensemaking in Organizations, Sage Publications, May 1995, ISBN 978-0-8039-7177-6. ↩

  14. Klein, G., “Performing a Project Premortem”, Harvard Business Review, vol. 85, no. 9, pp. 18–19, September 2007, hbr.org/2007/09/performing-a-project-premortem. ↩

  15. Christianson, M. and M. Barton, “Sensemaking in the Time of COVID-19”, Journal of Management Studies, vol. 58, no. 2, pp. 572–576, DOI 10.1111/joms.12658, March 2021. ↩

  16. Dell’Acqua, F., McFowland, E., Mollick, E., Lifshitz-Assaf, H., Kellogg, K., Rajendran, S., Krayer, L., Candelon, F., and K. Lakhani, “Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality”, Harvard Business School Working Paper 24-013, DOI 10.2139/ssrn.4573321, September 2023. ↩

  17. Holling, C., “Understanding the Complexity of Economic, Ecological, and Social Systems”, Ecosystems, vol. 4, no. 5, pp. 390–405, DOI 10.1007/s10021-001-0101-5, August 2001. ↩

  18. Beinhocker, E., The Origin of Wealth: Evolution, Complexity, and the Radical Remaking of Economics, Harvard Business School Press, May 2006, ISBN 978-1-57851-777-0. ↩

  19. Arthur, W., Complexity and the Economy, Oxford University Press, October 2014, ISBN 978-0-19-933429-2. ↩

  20. Pentland, A., Social Physics: How Good Ideas Spread — The Lessons from a New Science, Penguin Press, January 2014, ISBN 978-1-59420-565-3. ↩

  21. Deci, E. and R. Ryan, “The ‘What’ and ‘Why’ of Goal Pursuits: Human Needs and the Self-Determination of Behavior”, Psychological Inquiry, vol. 11, no. 4, pp. 227–268, DOI 10.1207/S15327965PLI1104_01, October 2000. ↩

  22. Yang, L., Holtz, D., Jaffe, S., Suri, S., Sinha, S., Weston, J., Joyce, C., Shah, N., Sherman, K., Hecht, B., and J. Teevan, “The Effects of Remote Work on Collaboration Among Information Workers”, Nature Human Behaviour, vol. 6, no. 1, pp. 43–54, DOI 10.1038/s41562-021-01196-4, January 2022. ↩

  23. Argyris, C., “Double Loop Learning in Organizations”, Harvard Business Review, vol. 55, no. 5, pp. 115–125, September 1977, hbr.org/1977/09/double-loop-learning-in-organizations. ↩

  24. Senge, P., The Fifth Discipline: The Art and Practice of the Learning Organization, Doubleday/Currency, October 1990, ISBN 978-0-385-26095-4. ↩

  25. Nonaka, I. and H. Takeuchi, The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, May 1995, ISBN 978-0-19-509269-1. ↩

  26. Brynjolfsson, E., Li, D., and L. Raymond, “Generative AI at Work”, Quarterly Journal of Economics, vol. 140, no. 2, pp. 889–942, DOI 10.1093/qje/qjae044, April 2025. ↩

  27. Reed, B., “Shifting from ‘Sustainability’ to Regeneration”, Building Research & Information, vol. 35, no. 6, pp. 674–680, DOI 10.1080/09613210701475753, November 2007. ↩

  28. Wahl, D., Designing Regenerative Cultures, Triarchy Press, May 2016, ISBN 978-1-909470-77-4. ↩

  29. Granovetter, M., “The Strength of Weak Ties”, American Journal of Sociology, vol. 78, no. 6, pp. 1360–1380, DOI 10.1086/225469, May 1973. ↩

  30. Burt, R., “Structural Holes and Good Ideas”, American Journal of Sociology, vol. 110, no. 2, pp. 349–399, DOI 10.1086/421787, September 2004. ↩

  31. Centola, D., How Behavior Spreads: The Science of Complex Contagions, Princeton University Press, June 2018, ISBN 978-0-691-17531-7. ↩

  32. Gunderson, L. and C. Holling (eds.), Panarchy: Understanding Transformations in Human and Natural Systems, Island Press, May 2002, ISBN 978-1-55963-857-9. ↩

  33. Maslach, C. and M. Leiter, The Burnout Challenge: Managing People’s Relationships With Their Jobs, Harvard University Press, November 2022, ISBN 978-0-674-25101-4. ↩

Loading…