What CSI is, how to read it, and what you get

CSI is a structured, analyst-reviewed read on global systemic risk, published every week. It compresses a fixed basket of ten indicators — spanning geopolitics and geoeconomy, conflict, the economy and markets, and systemic resilience — into a single resilience reading and a named regime, so you can see the state of the system, the direction it is moving, and what is driving it, without reassembling it from scattered headlines.

Decision-makers with real global exposure: executives, investors, manufacturers, supply-chain and treasury leaders, strategic planners. It is a decision-support layer, not a newsfeed — it tells you when the operating environment is shifting in a way that should affect commitments, hedging, liquidity or contingency planning.

It is not a market-timing tool, a recession or collapse predictor, a trading-signal generator, or a news aggregator. It will not tell you what a market does tomorrow. It gives you a disciplined, repeatable assessment of how fragile or resilient the system is right now, and how that is changing.

Every week we publish two readings of the same world. CSI-C (Conservative) draws on mainstream, institutional, broadly verifiable sources — what a prudent institutional observer would conclude. CSI-S (Sensitive) adds independent specialists and analysts — what informed observers may see before consensus catches up. C is slower and harder to dispute; S tends to move earlier.

The Consensus Gap is CSI-S minus CSI-C — the distance between the early, specialist reading and the mainstream one. A narrow gap means broad agreement. A widening gap means specialists are registering stress the mainstream has not yet recognised, and historically that divergence tends to precede regime transitions. The gap itself is a signal worth watching.

The CSI-S reading maps onto a five-step scale, each with a suggested posture: Stable (0–40) — orderly, routine monitoring; Elevated Stress (40–60) — visible strain in one or more systems, not yet systemic; Managed Instability (60–75) — serious stress being actively managed by authorities; Intervention Risk (75–85) — the system is near an extraordinary-intervention point; Declared Systemic Crisis (85–100) — a systemic crisis is underway. The full band table lives on the methodology page.

No. Confidence (0–100%) reflects how well-sourced and analyst-reviewed a reading is — the quality of the process behind it — not a probability that an event will occur. A high-confidence reading is well-corroborated; it remains an assessment of current conditions, not a prediction.

Momentum is the direction and pace of change in the index over the trailing weeks, tracked separately for C and S. It answers "which way, and how fast" — a stable-but-rising reading and a stable-but-falling one call for different postures.

CSI measures fragility — how close the system is to trouble. The CTI (Confirmed Transmission Index) is a separate gauge of whether elevated stress has actually been confirmed as broad transmission across systems, rather than remaining confined to sentiment. When confirmed stress is high enough it can floor the regime regardless of the fragility reading, so a live crisis is never under-stated.

The reading is built from a fixed taxonomy of ten indicators spanning four domains — geopolitics and geoeconomy, conflict, the economy and markets, and systemic resilience. Each is scored on a 0–10 scale and weighted by structural significance. Premium subscribers see the full per-indicator breakdown, the weekly change, and the evidence behind each score.

Yes. Premium subscribers see the exact weight of every indicator — shown as a percentage in the dashboard breakdown and in the appendix of each weekly report, alongside its score and weekly change. Structural drivers carry more weight than early-warning thermometers. What stays proprietary is the calibration behind the weights — how they were tuned against our historical reconstruction — not the weights themselves.

We present them as one section. The main risks are the developments most likely to move the reading now; the watchlist is the decision-trigger layer beneath them — the specific conditions that would escalate or de-escalate each risk, so you know in advance what would change the picture. The watch-trigger detail is part of the Premium note.

One reading is published every week as a Weekly Intelligence Note, delivered by email to confirmed subscribers and available on your dashboard. Each note shows the current regime, what changed, why it matters, the top risks, and — for paid tiers — recommended postures and the evidence layer.

Readings before the live-publication boundary are a reconstructed historical baseline — estimated after the fact under the methodology in force at the time, from available source data. They were not published live and are clearly labelled as such everywhere they appear. Live readings begin at the publication boundary.

Enterprise subscribers receive a customised alert system tuned to their own variables, regions and thresholds, plus direct analyst calls. Free and Premium tiers receive the scheduled weekly note.

Free gives you the current regime, both readings, the Consensus Gap, Confidence and Momentum, a public trend, and the methodology. Premium adds the full executive briefing, the per-indicator breakdown, sector and regional implications, the watchlist, full history and report archive, and the complete evidence layer with analyst reasoning. Enterprise adds monthly analyst calls, a customised alert system and priority desk access.

Yes — every new account begins with a 2-week Premium trial: full access, no commitment. It expires automatically and continues on Free unless you upgrade.

Premium and Enterprise access is provisioned per institution. Contact the desk at contact@novostok.com and we will set up the right tier.

Each score traces back to underlying signals — sourced observations, corroborated across independent sources wherever possible. Premium subscribers can open the evidence layer on any indicator to see the source links, supporting signals and analyst tickets behind the number. Nothing is asserted without a traceable basis.

Both, by design. Machine collection and scoring widen the surface, but every published reading passes through an analyst review and an approval gate before it goes out. Humans propose and dispose; automation never publishes an unreviewed reading.

Published readings sit on an immutable trail. When a correction is needed it is issued openly, with the change and its materiality disclosed, and — when material — notified to affected subscribers. The record is corrected, not quietly overwritten.

No. CSI is informational and analytical content only — a structured risk assessment with a clearly-labelled model-generated outlook. Outcomes are not guaranteed, and nothing on the platform is investment, legal, tax, financial or security advice. You remain responsible for your own decisions.