Multiplicative scoring model: how a decision engine combines dimensions

What is a multiplicative scoring model?

A multiplicative scoring model forms its score as a product of factors. Each factor describes one aspect of the situation, such as funding, demand, permits or team, and each takes a value between 0 and 1. Multiply them and the result is the score. A situation scores high only when every factor is reasonably high, which suits decisions that depend on several conditions holding together.

The structure is familiar and practical. The chance that a chain of steps succeeds is the product of the chance of each step, and a funnel multiplies its conversion rates. A decision engine applies the same logic to the dimensions of a situation. The model is transparent: a reader sees each factor, multiplies, and obtains the same figure, which keeps the discussion on the inputs.

How does a multiplicative scoring model differ from an additive one?

An additive model sums the factors, often with weights. A gain on one factor can cancel a loss on another, because the changes are measured in the same absolute units. A multiplicative model measures changes in proportion. A fall of 20% in any factor lowers the score by 20%, whatever the other factors are doing.

This proportional behaviour makes each factor a partner of all the others. In an additive model, the effect of improving a factor is the same wherever the other factors stand. In a multiplicative model, improving a factor pays more when the other factors are strong, and pays less when one of them is weak. The model rewards a team that fixes the weakest link first.

Why must ratings in a multiplicative scoring model sit between 0 and 1?

Normalising each factor to a range from 0 to 1 keeps the product on the same range. A factor of 1 leaves the score unchanged, and a factor below 1 reduces it. Raw values such as revenue in US$ or headcount cannot be multiplied directly, because their units differ. Convert each raw value into a rating first, with a stated rule, for example 0 for the worst plausible case and 1 for the best.

The rule for conversion matters as much as the product. Write down what 0, 0.5 and 1 mean for every factor, with a source for each anchor. Two analysts who use the same anchors will obtain ratings that match closely, and a reviewer can see where a rating came from. The simulator in the EQOS Asia demo uses sliders from 0 to 100 for this purpose (EQOS Asia simulator page, as of October 2026).

What happens to a multiplicative scoring model when one factor falls by half?

Take three illustrative factors of 0.9, 0.8 and 0.7. Their product is 0.504. Cut the middle factor from 0.8 to 0.4, a fall of half, and the product falls to 0.252, also a fall of half. An additive model with the same factors starts at an average of 0.800 and falls to 0.667, a drop of about a sixth.

The multiplicative model therefore reacts strongly to a deterioration that an average dilutes. For a committee, this means an early warning: one factor that halves is visible at once in the score. The values are illustrations, not EQOS Asia results. The table below sets the two models side by side for the same change.

Why do chained steps suit a multiplicative scoring model?

Many decisions are chains in which each step must succeed. Suppose a plan has 5 steps, and each succeeds 90% of the time. The chance that the whole plan succeeds is 0.9 multiplied 5 times, which equals about 0.59. Each step looks reliable on its own, and the chain is far less certain. A multiplicative model shows this loss directly.

The EQOS Asia method page describes the Φ lens in these terms: step-by-step conversion with identification of the main bottleneck (EQOS Asia method page, as of October 2026). A product makes the bottleneck visible, since improving the weakest step lifts the chain the most. The same reading applies to a sequence of approvals, hires or launches.

How does the EQOS Asia decision engine combine its dimensions?

The EQOS Asia engine rates 9 structural dimensions: coherence, integrity, resonance, care, trust, transparency, clarity, stability and energy. It aggregates them through a weighted geometric product, not an average, so the lowest score drags the whole system down (EQOS Asia method and case study pages, as of October 2026). The score φ(S) runs from 0 to 1.

The master equation Λ consolidates 8 lenses, including φ(S) for alignment, Π for pricing power, Ξ for runway, Ω for unit economics and Β for brand authority. A value of Λ at 1.0 or above marks a viable system (EQOS Asia method page, as of October 2026). The expanded forms and coefficients are proprietary, and the core is deterministic.

How do you build a first multiplicative scoring model for a decision?

Five steps give a first model. Name the decision in one sentence. List 4 to 9 factors that must hold for it to work. Define 0, 0.5 and 1 for each factor. Rate each factor and note its source. Multiply the ratings, then change the weakest one and watch the product. A spreadsheet with one column per factor completes the model in under an hour.

To see a full engine at work, try the public Λ simulator, which accepts ratings for 9 dimensions and returns a live score (EQOS Asia simulator page, as of October 2026). A 30-minute pre-diagnostic by video, scheduled in Singapore time (SGT), applies the equation to a real case, and the pricing page shows the tiers that follow (EQOS Asia contact and pricing pages, as of October 2026).

Frequently asked questions

About the author

Stéphane Jambu

President, EQOS Forecast Asia

Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile