Differential privacy

Differential privacy is a formal technique used when reporting on a group, designed so that the presence or absence of any single person's data makes no meaningfully detectable difference to the aggregate result. In practice, this typically involves adding carefully calibrated statistical noise or requiring a minimum group size before any pattern is shown.

The purpose is to allow useful group-level insight, such as noticing that a whole team's average recovery has dropped during a demanding project, without exposing or allowing the reconstruction of any individual's underlying data from that aggregate.

Sloane applies this kind of thinking to leader-facing views. Leaders see anonymised, group-level patterns, never an individual's wearable readings, calendar detail or messaging metadata. Group views are only shown where the underlying group is large enough that no single person's contribution could reasonably be inferred.

This is a meaningful design constraint, not a marketing claim: it shapes what leader dashboards can and cannot show, and it is the reason Sloane will decline to surface a pattern for a team that is too small to protect anonymity, even if a leader would find the detail interesting.