The Architecture of Returns

What the research asks

Most empirical work on venture returns asks which funds performed well and looks for the attributes they shared. The Architecture of Returns asks a different question: whether the way a fund is configured, the choices a firm makes about size, portfolio count, ownership, stage, scope, and the structure of the firm around them, relates systematically to how that fund performs.

The distinction matters because the first question tends to find attributes that cannot be acted on. Knowing that top-performing funds were often in the right place at the right time is true and useless. Configuration, by contrast, is chosen, and a relationship between configuration and outcome is a relationship a General Partner can do something about.

The dataset

The research is built on 2,471 United States venture capital funds spanning vintages from 2000 through 2024. That range deliberately crosses several complete market cycles, including the aftermath of the dot-com correction, the 2008 financial crisis, the long expansion that followed, and the correction after 2021, so that findings are not an artifact of a single environment.

Twenty-five vintage years is also close to the minimum needed to observe venture funds through to maturity. A fund raised in 2020 has not finished, and treating its interim marks as an outcome is one of the more common errors in this literature.

The finding the platform rests on most

The result that does the most work inside the Colibrí Architecture model concerns emerging managers.

The gap is large, it is persistent across vintages, and it runs against the intuition that scale and track record should compound into better outcomes. It is the empirical basis for the model treating lifecycle stage as its most important contextual variable, and for the model calibrating separately for each stage rather than holding every firm to one standard.

It is worth being careful about what the finding does not say. It does not say an emerging manager will outperform. It is a statement about medians across a population over twenty-five years, and the dispersion within each group is far wider than the gap between them. What it supports is the narrower claim the model actually makes: that emerging and established managers occupy different frontiers, and that a configuration efficient on one can be inefficient on the other. What gives that dispersion its shape, and whether it is growing more extreme, is a separate question this research does not settle, and the power law debate is where the evidence on it sits.

How the research reaches the platform

The Institute produces the research. Colibrí Strategies licenses the operationalized model from the Institute under a formal agreement that preserves the research firewall, and builds the platform that applies it. The Institute does not build product, and Strategies does not direct the research.

In practice this means the platform's baselines, its lifecycle calibration, and the structure of what it evaluates trace back to published empirical work rather than to a set of opinions about how firms ought to be run. The thresholds and weights that turn those findings into scores are proprietary to Strategies and are not published, which is a separate matter from the research itself.

Availability

The Institute publishes its research openly, and Why Emerging Venture Capital Managers Matter: Rethinking Institutional Portfolio Construction is available through the Institute directly. The Architecture of Returns is referenced here by title and finding, in the way the platform references it elsewhere, and is not distributed from within this library.

Sources

• Colibrí InstituteColibrí InstituteThe Institute's published research, including Why Emerging Venture Capital Managers Matter: Rethinking Institutional Portfolio Construction.