How ValuFai Works
From public observation to a capital-ready estimate
ValuFai turns a messy reality — several assets, a team, a set of dependencies — into a structured, evidence-aware, deterministic estimate of value. The process has six stages.
1. Discover
ValuFai identifies the assets in a portfolio and performs legitimate public observation where you ask it to — for example, fetching a public pricing page or documentation page to gather evidence. Discovery respects the boundaries of what is publicly available and never invents facts.
2. Question
A structured, adaptive diligence interview asks about the things that materially change value: revenue and its sources, dependencies, people, ownership, and relationships between assets. The interview adapts to what has already been answered and to the types of assets involved.
3. Measure
Assets, people, teams, capabilities, and relationships are structured and scored against a versioned framework. Every input carries an evidence state, so the model knows what was confirmed, what was owner-supplied, and what remains unknown.
4. Analyze
The engine examines the relationships among assets: shared audiences, shared infrastructure, concentration in one traffic source or one platform, cross-selling, and dependencies. These relationships drive the portfolio adjustment — a premium, roughly additive value, or a discount.
5. Value
A deterministic engine produces the low, base, and high range for the portfolio, the standalone asset values, the portfolio effect, the operating-capability value, and an evidence confidence score. The numbers are reproducible: the same inputs and methodology version produce the same result.
6. Package
Results are presented so a founder can understand what is driving the number and what is holding it back: major strengths, major risks, missing evidence, and prioritized value-creation opportunities. Formatting the same evidence for a specific audience — a bank, an investor, a strategic acquirer — is the capital-readiness function ValuFai is building toward.
Throughout, the principle is the same: AI assists with analysis, questioning, classification, and explanation; it never decides the numbers.
Last updated: 2026-08-08.