Walk-in · demo background
Production surface — everything below is what the live dashboard would show. This fold is for the presenter.
Talk track.
Who actually came inside? Camera walk-in is 41%; card-match is 24%. The
gap is cash, a second card, and passengers. Lunch converts; the morning
commute fuels and leaves. Unlinked is unresolved, not “didn’t come in.”
Claims.
Continuous presence tracking + sampled AI review. Cross-camera links use
topology and transit windows, not face match. No person imagery on this
page. Rate denominators are resolved customer journeys. “Left without
reaching the counter” does not claim whether a purchase occurred.
Numbers are illustrative, on real mechanism.
Build readiness.
Same- and cross-camera journey linkage is a shipped primitive. Walk-in
rate, fill-ending split, daypart conversion, and the door-to-served
funnel are derivable. Left-before-counter is not built.
41% walked in.
113 entered the queue; 102 reached a serving state; 11 left the queue without reaching the counter. No purchase outcome is inferred. Same 102 as Checkout. Checkout →