Reading Notes

Eric Ries on Incorruptible, Financial Gravity, and Mission-Controlled Companies

Source: Eric Ries on Incorruptible, Financial Gravity, and Mission-Controlled Companies

Notes — Eric Ries on Incorruptible, Financial Gravity, and Mission-Controlled Companies

Four questions [Adler frame]

Q1 — What is it about? A theory of why good companies go bad, and a practical framework for founders who want to prevent it. The core claim: organisations are subject to a structural force (“financial gravity”) that pulls them toward short-term extraction regardless of founder intent, and only deliberate structural defences — not good intentions — can resist it.

Q2 — How is it argued? Primarily through historical case study: Novo Nordisk (founded 1923, still mission-aligned a century later), Vectura (a British inhaler company that sold to Philip Morris because its charter required accepting the highest bid), Cloudflare (principled choices compounding into $70B of value), Groupon (email-frequency death spiral as a case study in unprincipled ROI optimisation). Ries also cites academic research: Harvard Law data (only 20% of founders still CEO three years post-IPO), and literature showing industrial foundations are 6× more likely to survive to year 50 with superior ROIC. The argument structure is: horror story → diagnosis → structural solution → track record of the solution.

Q3 — Is it true? The Harvard Law 20% figure and the 6× longevity finding are concrete anchors. The Vectura story is verifiable and illustrates the fiduciary duty problem without exaggeration. The Cloudflare SSL story (drive-cost-to-zero, top-of-funnel up by an order of magnitude) is well-attested. The Groupon death spiral is documented public record. The weakest claim is the magnitude of the implied counterfactual — how much better would these companies have done with the structural protections from day one? Ries doesn’t prove this directly; he infers from the correlation between governance structure and longevity/ROIC in the foundation literature.

Q4 — What of it? Founders should act immediately, before success makes them a target:

  1. Incorporate as a Public Benefit Corporation. Two-page filing. No meaningful cost.
  2. Write a mission you’d rather die than betray. Adversarial-prompt test it.
  3. Implement mission drive: audit whether anyone can profit by betraying the mission.
  4. Establish a mission guardian (perpetual purpose trust, nonprofit foundation, or employee ownership) — ideally encoded in the charter even if the structure isn’t operationalised yet.
  5. Install a director’s oath (the corporate equivalent of a Hippocratic oath). The structural window closes as leverage transfers to investors, bankers, and lawyers who all say “do it later.” Later becomes never.

Glossary

Financial gravity — The structural force within capitalism that pulls organisations toward short-term extraction of value from stakeholders, regardless of founder intent. It operates on the incentives of investors, bankers, lawyers, and growth equity holders simultaneously, making it a systemic force rather than a matter of individual bad faith.

Mission drive — The property of an organisation where it cannot profit except by achieving its stated mission. Distinguished from “mission hopeful” (good intentions, no structural enforcement). Test: is there anyone in this company who could make money by betraying one of our stated principles? If yes, you’re mission hopeful.

Public Benefit Corporation (PBC) — A US legal entity type available in Delaware and most states. A two-page filing that replaces the default “any lawful act or activity” (= maximise shareholder returns under shareholder primacy law) with a stated beneficial purpose. Changes the fiduciary duty to also encompass the stated mission.

Shareholder primacy — The theory, dominant since roughly the 1980s, that corporations exist solely to maximise returns to shareholders. Ries notes this is a ~40-year-old idea that replaced centuries of beneficial-purpose doctrine. Under it, “any lawful act or activity” = maximise shareholder value.

Spiritual holding company — Ries’s omnibus term for the category of structures that holds the animating mission of a company above financial extraction. Includes nonprofit foundations (Novo Nordisk), perpetual purpose trusts (Patagonia, Anthropic LTBT), employee ownership trusts (John Lewis, Mondragon), and employee voting trusts (Alibaba). Related: Financial Gravity.

Mission guardian — The entity (person or structure) whose job is to protect the mission from financial gravity. Options: founder control (temporary bridge), nonprofit foundation, perpetual purpose trust (with a “purpose protector” who can sue the trustees), or employee ownership.

Long-term benefit trust (LTBT) — Anthropic’s specific implementation: a non-economic entity (no equity, no financial interest) whose trustees are AI safety experts accountable only to the mission. Appointed to the for-profit board but with accountability to an outside group. No financial incentive to grow, every incentive to ensure it’s done properly.

Harder is easier — The leadership principle that principled commitments (to quality, design, safety, ethics) generate trust that compounds into competitive advantage: lower cost of customer acquisition, higher retention, better employee alignment, greater resilience after mistakes. Counterintuitive because trustworthiness is an intangible benefit with tangible costs, so ROI-based thinking always de-prioritises it.

Culture bank — The organisational asset of trust, built by making deposits (doing the right thing with a sacrifice, publicly visible) and eroded by withdrawals (self-interested extraction). The Todd Park rule (from Howard Schultz): only make deposits, never make withdrawals intentionally.

Invisible leader — Mary Parker Follett’s (1920) concept: the true leader of an organisation is its common purpose — the animating ethos — not the person at the top. Most consequential decisions are made when no manager is present. Only the invisible leader is present. If you don’t cultivate common purpose, you have no control over those decisions.

Torchbearer — Ries’s term for the rare person in an organisation committed to doing the right thing regardless of ROI pressure: the designer who won’t ship slop, the engineer who won’t sacrifice performance, the PM who prioritises the right thing even when people complain. In a mission-aligned company, everyone becomes a torchbearer; in a conventional company, they are isolated and beleaguered.

Emergent intelligence — The scientific principle by which a collective of individually simple agents (ants, transformer layers, employees) produces behaviours that no individual component could produce. Corporations are the oldest form of artificial intelligence: they are emergent super-organisms. The same design principles that make AI alignment hard also make corporate alignment hard.

Key argument: ethos + integrity

Ries’s formula is ethos + integrity:

  • Ethos = internal alignment: character, choices, purpose. Who would you rather die than betray? Write it down.
  • Integrity = structural resistance: two senses — (1) the ability to keep promises, and (2) structural integrity (stainless steel vs. corroded bolts). An organisation that is structurally weak cannot keep its promises because the person making them won’t be there.

Neither alone is sufficient. An organisation with ethos but no integrity is “mission hopeful.” One with integrity but no ethos has rules without a soul.

Key mechanism: it is always too early until it’s too late

The standard advice from lawyers, VCs, bankers is: “Do it later.” Each time:

  • Pre-seed: “Get product market fit first.”
  • Series A: “Don’t spook investors.”
  • Series B/C: “Growth investors want clean cap table.”
  • IPO prep: “Land the plane, house in order.”
  • Post-IPO: “Now it’s too late.”

The advisor incentive structure: all parties making this argument profit from the transaction volume on the way up and on the way down. They are not structurally aligned with the founder’s long-term mission. Related: Financial Gravity.

Key case: Novo Nordisk

Marie Krogh (diabetes patient, scientist wife of Nobel laureate August Krogh) incorporated Nordisk Insulinlaboratorium in 1923 as a for-profit company owned by a nonprofit foundation — a two-tiered “industrial foundation” structure. The structure pre-dates Martin Shkreli by 50 years and was explicitly designed to prevent insulin price-gouging. The trustees once exercised their power to prevent an extractive sale and created $500B+ of value. Novo Nordisk is still the same structure today — one of the largest companies in the world.

Key case: Vectura

UK inhaler therapeutics company, spun out of University of Bath. Philip Morris offered 165p/share vs 155p from private equity vs independence. The board had a fiduciary duty to accept the highest bid under standard governance. They accepted. Philip Morris wrote down £900M within three years. Company doesn’t exist anymore.

Key case: Cloudflare

Engineers organically developed “make a better internet” as the mission (it was emergent, not top-down). When a junior engineer pointed out that giving away SSL encryption for free would advance the mission, Matthew Prince said “let’s figure it out” rather than “that’s our most profitable product, get lost.” Result: drove down SSL cost through hand-rolled assembly language and biz dev with certificate authorities, gave it away for free, top-of-funnel increased by an order of magnitude, company is now worth $70B. The Groupon contrast: email frequency optimisation destroyed the core product one experiment at a time.

Anthropic structure

PBC from inception. Long-term benefit trust implemented at Series C but reserved in charter from day one. Trustees are AI safety experts with no equity. Appointed to and accountable to an outside group, not the for-profit board. Economic incentive structure: trustees benefit only from the mission being done properly, not from growth. This is why Anthropic can refuse to release a model they think is too dangerous — there’s a structural cost to doing so, but no structural incentive to abandon the constraint.

Mary Parker Follett

Feminist management theorist, 1920s, contemporary of Frederick Winslow Taylor but largely erased from history. Rediscovered in 1990s; Peter Drucker called her the “Prophet of management.” Key ideas relevant to Ries: “power with, not power over”; the invisible leader as common purpose; “the hallmark of a leader is, can they create more leaders?”; the superior and subordinate together obey the law of the situation.