The Distributed Studio.
Why physical concentration is a sunk cost — and how to build a high-output operating company without geography.
Silicon Valley, the New York office tower, the London financial district — these were the physical infrastructure of high-output operating in the twentieth century. They are the largest sunk cost in the operating economy of the twenty-first. The Distributed Studio is the architectural response.
The geographic premium.
An operator who builds in San Francisco today pays roughly a 3-5x cost premium across rent, salaries, recruiting expense, and cost-of-living adjustments compared to a comparable operator who builds in a tier-2 U.S. city or in a major international hub like Lisbon, Buenos Aires, or Tallinn. This premium is documented, well-known, and accepted by Silicon Valley founders as the cost of access to the talent ecosystem.
The bet implicit in accepting this premium is that physical proximity to other operators, capital, and talent produces operational advantages that outweigh the 3-5x cost. For thirty years, this bet was demonstrably correct. The serendipity of the right meeting at the right coffee shop in the right month was sometimes the difference between a company that scaled and one that didn't.
The bet became substantially weaker between 2015 and 2023, and is now demonstrably wrong for many operating categories. Remote infrastructure works. Asynchronous decision-making works. Distributed engineering works. The premium remains; the advantage that justified the premium has eroded.
What a Distributed Studio actually is.
A Distributed Studio is not a remote-first company with the same operating assumptions as a co-located company. It is an operating company architected from the ground up around the assumption that no two team members will be in the same physical location, and that the operating system itself must absorb whatever coordination cost that produces.
The architecture has three layers. The first is communication infrastructure: asynchronous-by-default tools, written-first decision-making, recorded video for everything that would have been a meeting in a co-located company. The second is talent topology: hire from any geography, accept that team members may never meet in person, and design compensation and equity to be neutral to geography. The third is brand and methodology: the company's identity is built around its written outputs — manifestos, indices, methodologies, articles — rather than around its physical offices or social events.
A company that has built all three layers is structurally different from a remote-first company. It does not lose anything by being distributed; it has been designed to produce its output without any element that would benefit from physical concentration.
The capital efficiency case.
The Distributed Studio's most obvious advantage is cost. An operator who can pay $80K for engineering talent in Buenos Aires that would cost $300K in San Francisco can build a 3.75x larger engineering team for the same payroll. The talent at $80K is not 3.75x worse than the talent at $300K. In many categories, the talent is comparable, and in some categories the international talent is meaningfully better because the candidate pool is larger.
This is a one-time arbitrage that will compress over time as remote work becomes more standard and salary expectations equalize. But the window remains open today, and operators who exploit it are building larger, more capable operating teams at the same cost as their geographically concentrated competitors.
The talent ceiling case.
Less commonly understood: the Distributed Studio also has a higher talent ceiling, not just a lower talent floor. The best operator in your category may live in Tallinn. The best engineer for your problem may live in Lagos. The best brand strategist may live in Tokyo. The geographically concentrated company can only hire candidates willing to move. The Distributed Studio can hire candidates who would never have moved — and that pool, properly counted, contains substantially more world-class talent than the moving pool does.
Operators who have made this transition consistently report the same outcome: the quality of senior hires improves once geographic constraints are removed. Not because remote talent is generically better, but because the relevant talent pool is structurally much larger when geography is not a filter.
The methodology case.
There is one further advantage of the Distributed Studio that founders rarely articulate: it forces the operator to build methodology explicitly. A co-located team can coordinate through proximity, hallway conversations, and casual context-sharing. A distributed team cannot; everything has to be written down, codified, made explicit. The forced explicitness produces a methodology that the company can scale, train, and ultimately publish.
This is why the Distributed Studio is the natural operating architecture for the Parallel Operator. The methodology that is forced into existence by the distributed structure is the same methodology that compounds across the portfolio. Co-located companies often have implicit methodology that lives only in the heads of senior team members; distributed companies are required to have explicit methodology that can be transferred, taught, and reused.
Geography as a strategic decision.
The choice to build geographically is now a strategic decision, not a default. Operators who make it deliberately, with a clear understanding of what the premium buys them, can still build great geographically concentrated companies. Operators who make it by default, because that is how their predecessors built, are accepting a 3-5x cost penalty for diminishing strategic returns. The Distributed Studio is the architectural alternative that the era now favors.
Continue reading the Doctrine.
Seven whitepapers elaborate on the theses in the Sina Doctrine manifesto. The Library catalogs all of them, alongside the manifesto, the Annual Letter, and the reading list.