Creator score and the decision engine
3 min read
The creator score is a 0–100 health number computed daily from your live data. The decision engine turns that analysis into a ranked list of up to 5 actions to take. Both update every day automatically.
How the creator score is calculated
The score is a weighted average across five dimensions:
- ●Growth (30%) — follower growth rate across connected platforms
- ●Engagement (25%) — average engagement rate relative to your historical baseline
- ●Revenue (20%) — monthly revenue trend and consistency
- ●Consistency (15%) — posting and sync regularity over the last 30 days
- ●Diversity (10%) — number of active platforms and revenue sources
Score labels: Strong (80–100), Growing (60–79), Building (40–59), Needs attention (below 40). The dashboard shows your current score, weekly change (+/- since last week), and the weakest dimension to focus on.
The creator score requires at least one successful platform sync to calculate. If you've just connected your first platform, allow the first sync to complete before expecting a score.
The decision engine
The decision engine scans your live data daily and produces a ranked list of up to 5 actions. These are specific, data-grounded — not generic suggestions. Examples:
- ●"Reconnect Twitch — sync has been failing for 3 days" (Risk, High priority)
- ●"Engagement rate is below your segment median — see benchmark analysis" (Growth, Medium priority)
- ●"Your follower growth is in the top 25% of your segment — keep the current cadence" (Growth, Low priority — reinforcement)
- ●"YouTube is your highest-revenue platform — it's also your only source. Consider diversifying." (Revenue, Medium priority)
Decision priorities
- ●High (red) — something is broken or at immediate risk. Act soon.
- ●Medium (amber) — a gap or opportunity worth addressing this week.
- ●Low (slate) — informational or long-term consideration.
The decision engine is most useful when you act on high-priority items quickly. Ignoring a "failed sync" decision for weeks means benchmark and score data will drift from reality.
Was this helpful?