Our Philosophy

The Primacy of Human Judgment over Generative Speculation

If we have better technology than ever before, why do projects still end up with the same problems?

Over 95% of construction data goes completely unused, with only roughly 0.5% used effectively. Poor data quality contributed to over $1.8 trillion in lost value globally in a single year. As of 2026, 92% of respondents still report budget overruns of 6% or more, and poor communication and coordination is now cited by 41% of global AECO teams as a leading cause of rework, according to Revizto's 2026 Digital Design and Construction Report.

This is not a talent problem and it certainly is not an expertise problem. An integrative process based upon professional judgment and stakeholder collaboration will forever be the most important part of construction. I did not build Siloplot to change that or to replace human judgment. I built it because I believe the people doing this work deserve a clearer picture to work from.

The Theory

In the AEC industry, most of the information professionals need to evaluate a design choice is dispersed throughout technical documents and data sources that are often heterogeneous in nature. No single actor can simultaneously see how a given design choice will affect direct costs and schedule, interact with site specific conditions such as water stress or local climate risks, impact emissions throughout the building's lifecycle, satisfy ever changing code and regulatory constraints, and be interpreted under different reporting standards, all while meeting project specific client and community preferences.

Any coordination mechanism in sustainable construction must therefore operate under conditions of inherently partial, local information rather than full centralized knowledge.

This exemplifies what Hayek identified decades before the construction industry had a software problem. To quote Hayek, the knowledge relevant for coordination exists "solely as dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess." In Rules and Order, he explains how price signals convey this dispersed knowledge and guide market decisions to help it continuously return to equilibrium.

The question is not whether the knowledge exists, but whether the right people have access to it at the moment it matters.

In the AEC industry, those channels are broken. As I argue in my thesis, the same institutions intended to correct market failures often make coordination even more difficult. Each new regulatory layer, rating framework, and software platform adds to the clutter of information without improving the channels through which that information reaches the people who need it. There is not one coordinating authority at the top of the pyramid, but rather a cluster of partially overlapping bureaucracies with their own information and compliance requirements. We don't lack guidelines or tools, but each new layer of information tends to make the coordination problem worse.

"The knowledge relevant for coordination exists solely as dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess."

— Friedrich Hayek, The Use of Knowledge in Society, 1945

The Evidence

I believe the financial sector offers the clearest example of dispersed knowledge handled well. As written in my thesis, new information is almost instantly incorporated into valuations, which allows decentralized actors to adjust according to prices that already reflect the latest information. No single person needs to hold all of the information because the system aggregates it continuously.

For example, the Bloomberg Terminal did not centralize all financial knowledge or completely eliminate complexity or regulation. It did, however, aggregate data from multiple sources and provide tools for querying, comparing, and acting on that information in near real time. It allowed participants to interpret and respond to dispersed signals with a degree of precision that would be impossible otherwise. The AEC industry does not have an equivalent mechanism.

Instead, the knowledge required for coordination is distributed across regulatory environments, investment priorities, and operational project workflows. This problem of digital information fragmentation has persisted for decades, suggesting that the industry may benefit from stronger ways of connecting those information channels and evaluating tradeoffs while decisions are still reversible.

Siloplot was built as an exploration of that idea.

The Thesis

The ideas behind Siloplot are developed in full in the paper below.

The thesis examines the structural causes of data fragmentation in sustainable construction, the coordination challenges that result, and explores what kinds of institutional and technical conditions may support better project decisions.

To read further on these ideas, here is my full thesis.