Every impact, risk, and opportunity (IRO) you score gets reduced to a single number between 0 and 100: its Materiality Score. This score is what a threshold profile compares against to decide whether an IRO is material, and it's also what gets aggregated up to give each sustainability topic its position on the materiality matrix. This article explains how the score is calculated and why severity counts for more than likelihood.

Score = 100 × Severity^0.75 × Likelihood^0.25
Severity and likelihood are each normalized to a 0-1 scale before entering this formula (see Scoring an impact, risk or opportunity for how your answers map to these values). The result is always a number between 0 and 100.

The 0.75/0.25 weighting means severity contributes three times as much as likelihood to the final score. This reflects a simple principle: a highly severe impact shouldn't be dismissed just because it's unlikely. A catastrophic but rare event can still score higher than a mild but near-certain one.
This weighting is fixed. It never changes based on which threshold profile you've selected, or which company, sector, or topic you're looking at.

The Materiality Score is calculated the same way regardless of which threshold profile is active. Only the threshold, the minimum score required to be material, changes between profiles. A score of 65 always means the same thing; what changes is only whether 65 clears the bar you've set. See Management of Materiality Thresholds for how profiles work.
This separation is what makes scores comparable: across IROs, across topics, across companies, and across reporting years, a given score always carries the same meaning.

Using the Balanced profile (threshold 56.2):

Individual IRO scores are aggregated up to the sub-sub-topic level using a maximum-score rule: each sub-sub-topic takes the highest score among its constituent IROs, separately for impact and financial materiality. This is deliberately conservative. A single highly material issue within a topic isn't diluted by averaging it with less material ones, so the group-level result always reflects the most critical exposure. The same maximum-score logic is used again when consolidating results across multiple entities in a group.

Not necessarily. Severity contributes three times as much as likelihood, so a highly severe IRO can outscore a more likely but milder one.

No. The score is calculated the same way regardless of profile. Only the threshold it's compared against changes.

Yes. The formula and its weighting are fixed, so a given score always means the same thing, regardless of company, sector, or year.

Averaging would dilute a single highly material issue among less material ones. Taking the maximum ensures the topic-level result always reflects its most critical exposure.

100, which requires both maximum severity (1) and maximum likelihood (1).