What Makes Institutions Endure While Companies Don't

I grew up in Kalahandi in a district that had plenty of organisations — government offices, schools, local bodies — but very few institutions. The difference was visible even to a child. Organisations existed. Institutions functioned. Organisations had names, buildings, and staff. Institutions had norms that were stronger than any individual inside them, systems that ran when the people who designed them had long since left, purposes that outlasted the politics of any particular year.

The organisations without institutional architecture didn't survive transitions. When a good administrator left, the office reverted. When funding shifted, the programme disappeared. When the person who held something together was gone, the thing was gone too. The institutions — the ones that had built their purpose into how they made decisions, not just into who made them — kept going.

I think about that distinction constantly when I'm building now.

The numbers on lifespan

The average lifespan of an S&P 500 company has fallen from 61 years in 1958 to under 18 years today, according to McKinsey research on corporate longevity. At the current rate of churn, three-quarters of the companies on the S&P 500 today will be replaced by new entrants within the next decade.

Most of the companies that disappear from that list do not fail because of bad strategy or poor execution. They fail because they were built for the environment that existed when they were growing — and they did not have the institutional architecture to adapt when the environment changed. They were companies, not institutions. And companies are optimised for a specific moment.

Jim Collins and Jerry Porras, studying what distinguished visionary companies from merely successful ones across 70 years, found that the companies built to last had one thing in common: they preserved a core identity — a set of values and a fundamental purpose — while continuously renewing their methods, products, and strategies. The ones that disappeared had either no core (they drifted with the market) or a rigid shell (they preserved the methods instead of the values).

The enduring companies — 3M, Johnson & Johnson, Procter & Gamble, Merck — were not more successful in any given decade than their competitors. They were more architecturally sound across multiple decades. That is a different thing.

What institutions have that companies don't

Shell, Sumitomo, and a handful of other corporations have operated continuously for more than 200 years. Arie de Geus, studying these long-lived organisations for Royal Dutch Shell, identified four characteristics they shared: sensitivity to their environment, strong cohesion and identity, tolerance and decentralisation, and conservative financial management.

What strikes me about that list is what is not on it. No mention of exceptional founders. No mention of brilliant strategy. No mention of competitive advantage in their original market. The things that produced endurance were architectural — how the organisation related to its environment, how it sustained identity across generations, how it distributed authority.

This is what separates institutions from companies. Companies carry their operating logic in their people — particularly in their founders and senior leaders. Institutions encode their operating logic into their systems: decision architecture, accountability structures, cultural transmission mechanisms, knowledge management practices.

When the people leave a company, the operating logic goes with them. When people leave an institution, the systems remain.

The things that make institutions endure are not exceptional founders or brilliant strategy. They are architectural — how the organisation makes decisions when the founder isn't in the room, how it transmits culture without proximity, how it learns without depending on any single person's memory.

The three things that have to be built

Running executive search at Majhi Group and building the operational infrastructure of Majhi OS in parallel has given me an unusually specific vantage point on this. The VP and C-suite leaders I place most successfully into organisations are the ones who understand institutional architecture intuitively — who ask, in the first interview, not just "what is the strategy" but "how are decisions made, who owns what outcomes, and what happens when those two things conflict."

The organisations where those searches go best are the ones that can answer those questions clearly.

Decision architecture. The question is not who makes decisions — most companies have org charts. The question is who actually decides, under what criteria, with what information, and who can see the decision being made. Institutions have written decision frameworks. Companies have the founder's instinct. The founder's instinct is fine while the founder is present and their instinct is good. It does not transmit.

When I built the observability layer of Majhi OS — the system that monitors mandate health and surfaces recovery recommendations — the first design question wasn't "what should the system detect." It was "when the system fires an alert, who decides what happens next, and what do they need to know to decide well." That is institutional architecture in a technical system. The same question applies to human organisations. The answer needs to be written down and survive personnel changes.

Accountability architecture. Accountability is not the same as responsibility. Responsibility is assigned. Accountability is real when the person assigned responsibility actually bears the consequences of outcomes — reputationally, economically, structurally. Most organisations have plenty of responsibility assignment and very little accountability architecture.

The organisations built to last are the ones where accountability is genuine: where the person who owns an outcome is judged by whether it happened, where failure has visible consequences, and where success has meaningful recognition. Not because these organisations are harsh, but because accountability structures shape behaviour before the fact. People who know they will be judged by outcomes care about outcomes in a specific way.

Culture transmission. The cultures of companies with strong founders feel coherent while those founders are present. Culture is transmitted by proximity — by watching how decisions get made, by seeing what gets praised and what gets quietly ignored, by absorbing norms from the people who carry them.

Institutions that endure have deliberate culture transmission mechanisms that do not depend on proximity. Written operating principles that are specific enough to guide actual decisions (not vague values statements). Onboarding that teaches how things actually work, not just what the product does. Leadership development that explicitly trains the next generation of decision-makers rather than assuming culture will diffuse. The Jesuits, who have operated continuously for nearly 500 years, trained leaders in the same values and decision frameworks across generations not because their founders were exceptional, but because they designed a transmission mechanism. The transmission mechanism outlasted any individual.

The Odisha question

Building from Odisha — and before that, from Kalahandi — has shaped how I think about institution-building in a specific way. There were no models around me. The institutions I could observe were mostly government bodies: slow, captured by incentive structures that rewarded presence over performance, designed for durability rather than function. The private sector models I read about were from Delhi or Mumbai or Bangalore.

Building something real from a place without infrastructure means building institutional habits early or building nothing at all. You cannot rely on the ecosystem to carry you. There is no informal network that transmits norms, no community of practice that socialises the right behaviours, no density of similar organisations that keeps standards high by proximity. You have to design what other people inherit.

That constraint, which felt like a disadvantage at the time, turned out to be a design forcing function. It required me to be explicit about things that founders in denser ecosystems can leave implicit. Decision systems that other founders carry in their heads because they have seen them operate elsewhere — I had to write down because there was no one around me who had seen them operate.

The lesson, which I have tested enough times to trust: institutional architecture that has to be made explicit because it cannot be assumed is more durable than the implicit kind. Writing it down is not bureaucracy. It is the difference between a company that depends on you and one that doesn't.

The question every founder avoids

At some point in building, every founder faces a version of this question: do you want to build something that needs you, or something that works without you?

The honest answer, for most of the early years, is that you need to be needed. The company runs because you are running it. That is not a failure — it is what building looks like in its first phase.

The failure is not transitioning. Most founders are so good at being needed that they never build the architecture that would make them replaceable. The company stays a company. It grows, sometimes significantly. And it remains dependent on its founders in ways that make it fragile to exactly the transitions — succession, scale, market shift — that eventually reach every organisation.

The companies built to last are not the ones with the most exceptional founders. They are the ones whose founders built, alongside the business, the institutional architecture that would outlast them.

That is the harder work. It is also the more interesting work, if you can see it clearly enough to choose it.

Sources

McKinsey — Corporate Longevity: Turbulence Ahead for Large Organizations

Jim Collins — Built to Last: Visionary Companies and the Core Ideology Framework

Harvard Business Review — The Living Company (Arie de Geus, 1997)