Build only when the market has not produced the right builder

Intellectual Frontiers Studios is a last-mile experiment capability. When the evidence remains compelling and a serious search finds no better-positioned founder or company, Studios builds the cheapest useful test before it considers building a company.

Thought moves through evidence to a decisionAn observation becomes a Native Alpha hypothesis. Research & IP, Press, Capital, and Studios test it in different ways while Network connects people across the whole system. Evidence supports a decision to continue, change, back someone else, build, or stop. Every result returns to revise the hypothesis.ObservationNative AlphahypothesisEvidenceNetwork connects every testResearch & IPPressCapitalStudiosDecisionContinue · Change · Back · Build · Stop
The units are not a pipeline. Each encounter with reality changes the thesis.

Building companies is expensive, distracting, and full of execution risk. If somebody else can credibly pursue the thesis, Capital should consider backing them. Studios becomes important when the thesis is early, sits between categories, depends on a capability unusual to the firm, or has not attracted the right builder.

The first build is usually an experiment: a prototype, a manually delivered service, a design partnership, an open-source project, a paid proof of concept, or something simpler that answers an important question cheaply.

Paid work beats pilots. A hospital that put a purchase order against an operating budget line has told us more than ten that signed a letter of intent, and a research site paying monthly tells us more than a free vendor pilot. Ventures that cannot earn revenue against a real budget line get closed early and cheaply.

The ventures worth chartering do work nobody was staffed to do. A company that reads every denial a plan issues the day it lands, or one that keeps a device submission's evidence assembled while the product runs, is doing a job that did not exist to be automated. That is a harder company to start and a much harder one to displace.

Experiments before entities

Modern AI makes software cheaper to build. That makes judgment more important, not less. The ability to build something in a week is not evidence that the week should be spent building it.

Customers provide better evidence than code. Usage provides better evidence than demonstrations. Payment provides better evidence than compliments. Renewals and repeatable economics are stronger still. Studios spends as little as reasonably possible to discover whether an idea deserves to exist.

What a partner gets on day one

A company that starts above zero, because the hard parts are already in place before the first hire.

A licensed rights position

Clinical workflow, device data, payer automation, or retail systems claims licensed into the venture, with the field of use written down.

Operators with scar tissue

People who have carried an FDA submission, an insurance integration, a trials platform rollout, or a large system replacement to the finish.

A first buyer in reach

Our research relationships are where most ventures find their first paying customer, rather than starting cold.

A written exit path

Entity, cap table, closure condition, and independence plan are set at the charter and held to when the venture gets uncomfortable.

The experiment does not have to stay here

A Studios experiment may prove the market and reveal a founder or company better positioned to take the work forward. The firm can invest, partner, license, merge efforts, spin the work out, or transfer the opportunity to a stronger operator.

The goal is not to defend ownership of an implementation. It is to preserve and compound the Native Alpha while putting the work in its best possible form.

Where ventures get built

Healthcare delivery, where the problem is usually getting outside data into a clinician's hands without another interface project. Medical devices, where the software organization decides whether the hardware ships on time. Insurance and digital authorization, where who may read what, and the proof it happened, consume operations budgets. Clinical trials, where the study still runs on binders and reconciliation. Digital marketplaces and payments, where many parties have to settle and keep the evidence.

Each of those is a market we have worked in. We do not charter ventures in markets nobody here has operated.

How to engage

Operators bring a market they have run and a problem they know is worth paying for. Health systems, manufacturers, insurers, and research organizations bring an internal project that should be a company but will never get funded inside. Investors bring capital alongside Intellectual Frontiers Capital once a venture has contracted revenue.

Come prepared to say what has already been paid for, what right the venture would need, and who would own the outcome.

Before bringing a venture

These pages show the standard a proposal is measured against.

  • The ventures and companies

    What Studios has chartered and maintains, with each entry's status on the record.

  • Ventures

    The operating companies we have chartered, each with an entity, an operator, and a closure path.

  • Software

    The software we build and maintain, with each entry's licence and status recorded.

  • Portfolio of substantial works

    Every venture and shared service we have chartered, including the ones we closed.

  • Use cases

    Deployments we have run in provider, device, trials, insurance, and marketplace settings.

  • Patent register

    Rights positions available to license into a new company.

  • Market potential analyses

    How we size a market before committing people to it.

  • Contact Studios

    Send the market, the paying buyer, and the operator who would run it.

Every venture is chartered to stand on its own. If it still needs the parent after a few years, we got something wrong at the start and we say so.

Put this to work

Venture experiment brief

Turn a broken workflow into the cheapest credible test of a company worth building.

For
Operators, institutions, founders, and design partners.
What you keep
A venture experiment brief you can review, revise, and send.
What counts as sound
  • Names a real buyer
  • Shows a measured problem
  • Has an accountable operator
  • Uses a bounded experiment
  • Includes a closure condition

A useful problem does not by itself justify forming a company.

Nothing entered here is stored or sent. Review the prompt before sharing confidential, personal, patient, or privileged information.

Review the prompt

You can leave any field blank. The prompt will mark it as not provided.

If the record survives your review, send the question, evidence, unknowns, and requested next step.

Bring us a venture