Industry
Retail Chains and Standalone Stores
Footfall analytics, queue intelligence, loss prevention, and heatmaps — across single stores and multi-store chains — with the per-store autonomy that brand-format chains depend on.
Typical client profile
Multi-store retail chains, large-format standalone stores, premium boutiques, and quick-commerce dark stores. Typical store deployment is 8–40 cameras; chain deployments scale to thousands across stores.
Indicative scope
Retail Discovery focuses on the store-format template: what works for one store must scale to all. We invest heavily in the per-site profile so a chain rollout is a configuration exercise, not a re-engineering one.
Compliance considerations
- DPDP Act 2023 — customer data minimisation, no individual identification without clear opt-in
- Consumer Protection Act considerations for loss-prevention escalations — never identify a customer as a loss-prevention target on AI alone
- Per-store autonomy so a store can keep running if the chain's central platform is unreachable
Retail stack
Modules typically deployed in this vertical.
The platform supports more — this is the configuration we most often land on for clients in this category.
People Counting
Footfall and occupancy by zone, by hour, with per-site calibration. Counts are de-identified at the edge.
Accuracy: 97% at zone entries with overhead camera; 92% at oblique angles.
Heatmap & Dwell Analytics
Aggregate dwell-time and traffic-pattern visualisation. All trajectories aggregated; no individual tracks persisted.
Accuracy: Visualisation quality is qualitative — we publish methodology, not vanity scores.
Queue Management
Wait-time estimation, queue-length thresholds, and abandonment alerts — wired to staff escalation flows.
Accuracy: Wait-time estimates within ±20% under steady throughput.
Multi-Camera Tracking
Handoff of a tracked entity across overlapping camera fields, with behavioural threading for incident reconstruction.
Accuracy: Track integrity is scene-dependent. We publish per-deployment integrity numbers rather than a single headline metric.
Loitering Detection
Zone-based dwell-time thresholds with operator-tunable durations; separates queue-waiting from anomalous loitering.
Accuracy: 88% on zones with consistent lighting and unobstructed lines of sight.
Predictive Analytics
Forecast capacity peaks, staffing needs, and anomaly windows from historical occupancy and event data.
Accuracy: Forecast accuracy is reported as MAPE per site after the first 90 days of training data.
ANPR / Licence Plate Recognition
Read Indian commercial and private plates at vehicle gates — single-lane, multi-lane, and toll-style installations.
Accuracy: 94% on standard high-security plates in daylight · 88% on aged/dirty plates · 85% at night with IR fill.
Fire / Smoke Early Detection
Visual smoke and flame detection as an early warning layer that complements — not replaces — fire-panel sensors.
Accuracy: 85% on visible flame within camera FOV; smoke detection accuracy is highly scene-dependent and is treated as supporting, not primary.
AI-Prioritised Alert Routing
Severity ranking, dedupe and suppression rules, escalation chains with on-call windows, and SLAs per alert class.
Accuracy: We measure mean-time-to-acknowledge and false-positive rate per class, monthly.
Run a retail site through Discovery.
Tell us about the site count, the existing camera estate, and the compliance perimeter. We'll come back with a scoping outline.