SVIE OS is a comprehensive, evidence-based framework for evaluating ventures at every stage — from the earliest idea through investor readiness. It was built for African market realities, not Silicon Valley assumptions.
Browse the full Knowledge Base →SVIE OS is not a pitch template, a checklist, or a scoring rubric. It is an operating system — a structured, sequential intelligence layer that a venture, a studio, or an investor runs on to make better decisions.
It codifies what experienced investors, operators, and advisors know intuitively — the questions you must be able to answer, in what order, with what quality of evidence — and makes that knowledge accessible to every founder, regardless of network or background.
The framework is structured across 9 phases and 50 modules, built on 19 Global Evaluation Standards. Every module has a defined purpose, guiding questions, evidence requirements, and a Decision Gate outcome. Nothing is subjective. Every output is scored and flagged.
SVIE OS v1.1.0 adapts its depth and evidence requirements to who is running the evaluation. Every track draws on the same 50 modules and evidence-weighted scoring (GES-18) — only the required scope and thresholds change. Modules are not available individually — they are run as part of a track to ensure evaluation integrity and comparability across ventures.
Solo founders and early teams (fewer than 5 members, no institutional funding)
Run the five core modules and receive a scored report, confidence rating, and prioritised action plan. Governance standards are calibrated for early-stage realities under GES-14 Minimum Viable Governance. This is the entry point for most first-time users.
Venture studios, accelerators, and grant-making programmes evaluating multiple ventures for selection
A 13-module evaluation covering founder, market, business model, and early financial signal — deeper than the MVE, targeted rather than exhaustive. Ventures reaching Conditional Go or better proceed to full evaluation. Module 8.6 (Portfolio Intelligence Framework) activates once three or more ventures have completed evaluation in the same cohort.
Founders preparing for investor meetings, grant applications, or strategic partnerships
A comprehensive, self-directed diagnostic covering all nine phases. No minimum evidence threshold — the output is a frank readiness assessment identifying your strongest evidence, your gaps, and what you need before approaching investors. Multi-session; save and resume at any point.
Investors and grant-makers using SVIE OS as their primary due diligence framework
A targeted 29-module set spanning founder, market, business model, product, and financial evaluation, plus the Integrated Decision Framework. Phase 6 (Financial Evaluation Sprint) and Phase 8 (Integrated Decision Framework) must reach minimum Level 3 evidence for a Go recommendation. Does not include Phase 5 (Operations), Phase 7 (Strategy), or Modules 8.4–8.6.
These are the three most recently added Global Evaluation Standards. SVIE OS v1.1.0 now spans 19 Global Evaluation Standards (GES-01 through GES-19) — GES-15 (Inter-Rater Reliability Protocol) governs how scores are consolidated when a venture is assessed by more than one evaluator.
Ventures operating in complex regulated sectors (healthcare, education, energy) or specialist sectors (biotech, medtech, mining) carry regulatory and credentialing prerequisites that are material to commercial viability. GES-17 assigns a Sector Complexity Class (A through D) to every venture and applies consequence rules for Class C and D: confidence capped at Moderate until Sector Readiness is verified; two or more unaddressed Class C requirements trigger a Fatal Flaw.
SVIE OS uses evidence-weighted aggregation: Level 4–5 evidence at full weight (1.0), Level 3 at 0.75, Level 1–2 at 0.5. Modules with insufficient evidence are flagged as Evidence Gap and trigger a Pause — not scored as zero. Phase scores aggregate to an Overall Confidence Rating (Very High / High / Moderate / Low / Very Low). A single Fatal Flaw Condition overrides the aggregate and produces a mandatory No-Go.
Before any market signal is weighted as evidence, the evaluator identifies the Ecosystem Role Map: who is the User, who is the Payer, and whether they are the same party. Signals are classified as Tier 1 (Interest: sign-ups, clicks), Tier 2 (Intent: LOIs, pilot agreements), or Tier 3 (Transaction: payment). A Tier 1 User signal cannot substitute for Payer demand evidence. Signal Misattribution — citing User signals as Payer evidence — triggers an Evidence Level downgrade on the affected module and a Revenue Model Risk flag.
The SVIE OS phases are sequential but not rigid — you can enter at your current stage and work forward. Each phase builds on the evidence and decisions of the one before it.
Before evaluating the idea, evaluate the evidence behind it. This phase establishes how much of what the founder knows is proven versus assumed, and scopes which track and modules apply.
Most ventures are solutions looking for problems. This phase forces the discipline of validating that the problem is real, painful, and persistent, and that real customers — not just users — will pay to solve it.
More ventures fail from founder fracture than from bad markets. This phase evaluates founder stability, relevant background, and the external accountability mechanisms that keep decision-making honest.
How does money move? This phase stress-tests the mechanics of value creation and capture — including M-Pesa-native revenue models, pricing, and informal sector economics.
Validate that the solution actually addresses the validated problem — not the assumed one. Includes technical architecture, IP defensibility, and product quality under real local conditions.
What is built must also be reliably produced, delivered, and legally compliant. This phase evaluates the operational machinery — from SOPs and vendor resilience to regulatory licensing and business continuity.
Build a financial model grounded in validated assumptions — not hockey sticks. For the Investor Due Diligence Track, this phase must reach minimum Level 3 evidence for a Go recommendation.
Does this venture have a defensible, scalable future? Impact is evaluated on four criteria — Directness, Scale, Durability, and Additionality. Solo-founder ventures may complete this phase at Level 1–2 evidence without triggering an automatic Pause, given a credible evidence-building plan.
By the time a venture reaches Phase 8, the evidence has been assembled across all prior phases into a single, defensible recommendation. For the Investor Due Diligence Track, this phase must reach minimum Level 3 evidence for a Go recommendation.
Every module in SVIE OS ends with a Decision Gate — a structured outcome that tells the founder (or evaluator) exactly what to do next. There is no ambiguity. No “interesting, but...”. Just a clear, evidence-based verdict.
The evidence meets or exceeds the threshold for this module. The venture is ready to move to the next gate. Strengths are documented for the investor profile.
The core foundation is sound but specific gaps must be addressed within a defined timeframe. The venture can progress while resolving the conditions.
Critical assumptions are unvalidated or evidence quality is too low to proceed with confidence. Continuing without resolution wastes resources and misleads investors.
A fatal flaw has been identified — a condition that cannot be resolved within the current venture structure. This is not failure; it is intelligence that prevents greater loss.
SVIE OS does not just ask what you know — it asks how you know it. Every claim a founder makes is rated against a five-level evidence hierarchy. The quality of evidence is as important as the content of the answer.
Audited financials, third-party due diligence, independent market research, regulatory approvals, signed contracts with major partners.
Pilot revenue data, signed LOIs, cohort analytics from real users, documented customer interviews with specific named sources.
Unpaid pilot users, documented customer discovery interviews (5+), observable market behaviour, cited industry reports from credible sources.
Surveys, analogous market data, competitor analysis, informal conversations that are not documented or systematically conducted.
Personal belief, anecdote, intuition, or "I think people would pay for this." This is the starting point — not the evidence.
Flags are the shorthand output of every evaluation — a colour-coded signal that tells all three tiers at a glance where a venture stands. They are derived from the score and the Decision Gate, not assigned subjectively.
This module is a venture strength. Evidence is solid, assumptions are validated, and the Decision Gate supports progression.
This module has promise but gaps. The foundation exists but validation is incomplete or evidence quality is low.
This module represents a material risk. Core assumptions are unvalidated, evidence is insufficient, or a structural problem exists.
The full SVIE OS covers 50 modules. The MVE Track is the 5-module subset that covers 80% of the risk with 20% of the effort — the questions that, if answered badly, kill most ventures before they scale.
The founder is the highest-risk variable in any early-stage venture. Misalignment between self-perception and reality in this module predicts more failures than any market factor.
The single most common failure mode: the problem is real, but the founder doesn't actually understand it deeply enough to solve it. This module separates assumed problems from validated ones.
Knowing the problem exists is not the same as knowing who experiences it most acutely, why, and what they've already tried to do about it. This module defines the beachhead customer.
A market size claim is a chain of four separately-evidenced steps, not one projected figure. This module scores the weakest link — problem reach, paying capacity, distribution, and conversion — not the founder's TAM slide.
A second-pass founder assessment at the market intelligence stage — testing whether the founder's self-assessment holds up once market realities are on the table.
No revenue model, no business. This module tests whether the logic of how the venture makes money actually works — unit economics, pricing power, and path to profitability.
SVIE OS is not a point solution for founders. It is the shared intelligence layer across an entire venture ecosystem. The same framework that evaluates a venture for a founder also generates the scored reports that studios use — and the discovery data that investors browse.
A founder completes the MVE Track (or full SVIE OS). Their answers are evaluated against the framework by Castalia. They receive a scored venture profile with flag, Decision Gate outcome, strengths, risks, and next actions.
Once the evaluation is complete, the venture profile becomes discoverable to Tier 2 and Tier 3 partners who have defined matching criteria. The founder's data remains private — only the scored profile is shared.
A studio running an application call defines their criteria (stage, sector, minimum score, flag requirements). Castalia pre-screens every applicant against those criteria and delivers a ranked, scored shortlist — not an inbox.
An investor defines their thesis. Castalia surfaces ventures from the ecosystem pool that match — including ventures that never pitched them. The investor sees standardised scores, not pitch decks, as the first filter.
The MVE Track takes 25–35 minutes. You will walk out with a scored venture profile, specific action items, and an honest picture of where your venture stands.
Start your free evaluation →