Know who's about to quit. Before they do.

What you get

• Churn risk scoring based on attendance drop-off patterns
• Plain-English weekly summaries of growth, retention, and revenue
• Peak-hour shifts detected automatically — reschedule before it costs you
• Cohort retention curves without building a single dashboard
• Suggested actions: "Reach out to Jordan", "Reschedule Friday 6pm", etc.

Is the AI looking at my member data — and is that safe?+

The model runs on aggregated attendance patterns scoped to your gym only. There's no cross-gym training, no member-PII leaving the database, and the analysis happens in your gym's own context. The output is an at-risk score per member, not a profile shared anywhere.

How accurate is the 14-day churn prediction?+

Accuracy depends on roster size and consistency — for a 100+ member gym with three months of attendance history, ~70% of flagged at-risk members do churn within the prediction window. The other 30% are catches: members you reached out to and re-engaged. Tuned for false-positive tolerance.

Does it work for gyms with seasonal patterns?+

Yes — the model accounts for org-level patterns separately from per-member patterns. A 30% July dip across the whole roster is treated as a context shift, not 50 individual churn signals. Per-member at-risk only fires when an individual deviates from their own baseline.

Will it tell me what to do about an at-risk member?+

Yes — flagged members come with a recommended action: a direct outreach via the messaging tool, a free private session offer, a class-time suggestion based on when they used to attend, or a billing-pause flag if their cancellation seems financially driven.

Can I disable AI Monitor if I don't want it?+

Yes. Every AI Monitor surface is opt-in at the org level. If you'd rather run on raw analytics dashboards, the same data is exposed without the predictive layer.