FuseEngagement Score
Engagement Score

A score you can read

No black box. The engagement score is the sum of a handful of named factors — how often they show up, how much of the product they use, how recently — with a trend as the tie-breaker. It never looks at spend, which is exactly the point: it is the axis RFM leaves out.

The factors

±40
Session rate
How often they show up, independent of what they spend
±30
Feature depth
How much of the product they actually use, not just how much they buy
±20
Recent activity
How fresh their last active session is, the early-warning half of the signal
±10
Trend
Whether their engagement is climbing, holding steady or fading

Points sum to a 0 to 100 score. Session rate carries the most weight — showing up is the strongest engagement signal — then feature breadth, then recency, with a climbing trend adding a bonus and a fading one subtracting. Illustrative weights; the production model calibrates them on your data.

The bands

Power
score 80+
Engaged
score 60 to 79
Regular
score 40 to 59
Light
score 20 to 39
Dormant
score 0 to 19
What feeds it in your world

No new tracking to invent. These are events you already capture; the model reads them as they are.

  • ·App and web sessions per customer, the show-up signal
  • ·Interaction events: opens, views, saves, feature use
  • ·Products touched, the depth axis
  • ·Recency of activity, separate from spend
How it sharpens over time

Models are tuned, not installed. Each step is earned on evidence, and the previous step keeps running until the next one beats it.

  1. v0
    Heuristic, day one
    Transparent rules on the data you already have. Live in Braze in weeks, and everyone can see why it fired.
  2. v1
    Calibrated to your history
    Thresholds and weights re-fit to your own base and vertical, backtested against what actually happened.
  3. v2
    ML where it earns it
    A trained model replaces the rules only when it beats them on holdout data. Explainability stays a requirement.
  4. v3
    Monitored and re-tuned
    Drift watched, thresholds reviewed on a cadence, and every change proven with holdouts before it ships.

Engagement starts the moment session events flow, before any spend history exists. Factor weights are re-fit per vertical (a daily-app QSR guest and a seasonal hotel guest show up very differently), and the band cut-offs are calibrated to your base's real distribution rather than fixed percentiles.