The Three Phases of Venture Decentralization
Move venture funding from institutions to individuals to automated decisions
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 92%
The Three Phases of Venture Decentralization describe a progression from institutional gatekeeping toward broader and more automated capital allocation. Phase one builds administrative infrastructure that lets smaller funds and special-purpose vehicles form efficiently. Phase two enables skilled individuals to leave established firms and operate as solo general partners, expanding the number and diversity of investment decision-makers. Phase three connects standardized company data from systems such as payment and accounting platforms to investment interfaces that can evaluate businesses and issue checks with fewer pitch meetings. The sequence matters: automation is useful only after accessible infrastructure and diverse independent investors exist. The intended output is faster funding, more competition for deals, less dependence on prestige networks, and eventually more objective capital allocation.
Origin
Extracted from Masters in Business, where Chamath Palihapitiya described AngelList, the growth of solo general partners, and a future of API-driven venture decisions.
Core principles
- 01Infrastructure must become accessible before participation can broaden
- 02Individuals can replace institutional investing teams
- 03Operational data can reduce dependence on pitch meetings
- 04Automation should lower bias rather than encode it
- 05Competition for deal flow improves access to capital
How to run it
- 1
Build accessible infrastructure
Automate the legal, administrative, reporting, and capital-call work needed to create small funds and special-purpose vehicles.
Pro tip Make the simplest compliant vehicle usable without a large back office.
Watch out Administrative access alone does not diversify actual investment decisions.
- 2
Enable independent investors
Give credible individuals the tools to raise, deploy, and report capital without joining a traditional venture partnership.
Pro tip Pair operational tooling with transparent track records and references.
Watch out Replacing firms with individuals can preserve bias if the same networks select every investor.
- 3
Standardize company data
Collect permissioned operating data from payment, accounting, sales, and product systems in comparable formats.
Pro tip Prioritize data that reflects durable behavior rather than vanity metrics.
Watch out Young or unconventional companies may lack enough standardized data for fair evaluation.
- 4
Automate repeatable decisions
Use transparent criteria to evaluate routine opportunities and issue funding decisions without unnecessary meetings.
Pro tip Start with narrow company types where the signals are well understood.
Watch out A black-box model can make bias harder to detect rather than removing it.
- 5
Audit and escalate exceptions
Monitor approval patterns, performance, and demographic or geographic disparities. Route unusual but promising cases to informed human review.
Pro tip Compare automated outcomes with alternative decision processes.
Watch out Optimizing only for historical winners can systematically reject genuine innovation.
In the wild
Automated infrastructure lowers the cost of creating funds and SPVs. Skilled investors then use that foundation to leave institutional firms, build independent track records, and compete directly for startup investments.
→ Capital allocation expands from venture partnerships to a larger group of individual decision-makers.
A software startup authorizes access to payment, accounting, retention, and sales data. A funding platform evaluates the standardized signals, provides transparent terms, and escalates atypical features to a specialist rather than requiring a conventional pitch process.
→ The company receives a faster decision while unusual businesses retain a path to contextual review.
Common mistakes
Automating too early
A model built before reliable data and accessible infrastructure exist will merely accelerate weak decisions.
Encoding historical bias
Training criteria on conventional winners can exclude founders and business models that differ from the old system.
Eliminating every human judgment
Pure automation may miss technically novel or pre-revenue companies whose most important evidence is qualitative.
Is it for you?
Best for
Fund-platform builders, solo general partners, fintech teams, and allocators redesigning early-stage finance.
Not ideal for
Highly novel companies whose value cannot yet be represented through meaningful operating data.
From the transcript
“Phase one was things like Angelists, which was building the infrastructure so that um, smaller funds and SPVs like single purpose vehicles or special purpose…”
“Step two is happening right now, which is that individuals are taking the leap and saying, I don't need to be a part of my…”
“And then the third phase after this as I suspect that you'll do a little bit more automated so that it's less biased than that.”
From the episode
Chamath Palihapitiya on Venture Investments (Podcast)
Chamath Palihapitiya