Causal engine

The next cycle is decided by looking at what the current cycle measured. The screen reads signals from the balance (real revenue, pent-up potential, gates, adoption), applies each candidate against them and ranks them — with the full calculation in view.

The NEXT BET declared BASED ON DATA — the strategic close of the cycle. It scores the candidates (2nd country, deep verticalization up to our own product, MGA/embedded insurance with risk priced by us, agent sold separately, consolidation) against the SIGNALS read from the balance (real traction of forecast × actual, solid calibration, risk we can price) and ranks them. Each candidate carries its declared ALIGNMENT with each criterion — the assumptions, in plain view. 🔴 The choice COMES FROM THE BALANCE, not from wishful thinking: the board SCORES and RANKS by explicit criteria, and does NOT decide on its own — people decide, with the criteria visible. Read-only.

Each candidate opens its own calculation breakdown: signal × alignment = share, line by line. No score on this screen needs to be accepted on trust — it is the same principle as the magnifying glass of the projected cash flow.

Score = sum of (signal × declared alignment). The SIGNAL comes from the data; the ALIGNMENT is an explicit assumption — how much that candidate benefits from that signal — declared, not hidden. Disagreeing with the ranking is legitimate, but the disagreement points to a specific number, not to a feeling.

The next bet is DECLARED based on the balance; people decide, with the criteria in view. If the final decision diverges from the ranking, great — as long as you can say which assumption was wrong in the model, and it is corrected for the next cycle.

What is the next bet — the one the balance points to, with the criteria in view? The screen scores the candidates (2nd country, deep verticalization up to our own product, MGA/embedded insurance, agent sold separately, consolidation) against the signals read from the balance: real traction of forecast × actual, solid calibration, priceable risk. Each candidate declares its alignment with each criterion — the assumptions are exposed. The board scores and ranks, but does not decide on its own: people decide, with the criteria visible.