SENTINEL monitors credit already granted and flags deterioration before the first missed payment. It is the module that moves an institution from reaction to anticipation.
What the module does
Each line describes a function by what it produces, never by the promise it carries.
- Activity: decision volume, distribution of outcomes, average handling time, movement over time.
- Decision quality: manual review rate, human override rate and its grounds, score distribution.
- Portfolio performance: default rate by score cohort, prediction against actual, discriminatory power.
- Data: average completeness, most frequently missing features, quality by source.
- Internal comparison: by branch, by product, by segment, by analyst.
Three questions a management team asks of its portfolio
Each view answers a precise question: how risk is spread, where the activity concentrates, and across which territories. A figure that answers no question has no place here.
Low risk96
Moderate risk148
High risk61
Critical risk22
What it never does
What a module refuses to do can be checked; what it promises to do well can only be argued. Its limits are therefore written next to its capabilities.
- VISION never shows an indicator without its period, its scope and its population size.
- No indicator computed on a population too small to be significant is shown without saying so.
What this changes
What having this module changes in the real work of a credit team.
The richest indicator in the system is not the default rate: it is the percentage of files where the analyst departs from the recommendation, and why. A high rate signals a calibration flaw, a blind spot in the data, or a training need. It is the loop that improves the platform.
Keep reading
SAGE answers the analyst's questions about a file, rephrases an explanation and drafts summary notes. It explains. It never decides.