🛍️ Staff Exclusion in People Counting: From Manual Burden to Behavioural Precision

For over two decades, people counting has formed the backbone of decision-making for retailers and mall operators. From understanding store performance to shaping staff plans and optimizing layouts, traffic data is treated as gospel. But that gospel has long carried a flaw: staff are often counted as customers, skewing the very insights it’s meant to provide.

While footfall tech has evolved, the staff exclusion challenge has remained frustratingly persistent. Until now.

🧭 Legacy Methods: A Heavy Lift for Store Teams

Traditional exclusion methods rely on wearables, tags, or lanyards—seemingly simple tools, but in practice they’re a compliance minefield.

Method	Concept	Challenges
Reflective Lanyards	Visible markers for thermal systems	Wear & tear, concealment by clothing or hair, no failure alerts
BLE Tags	Battery-powered exclusion devices	Silent battery drain, forgetfulness, added burden on store manager
Barcoded Name Tags	Scannable visual IDs	Relies on consistent wear and placement, easily obscured or ignored

Each method adds operational friction at a time when retailers are trying to free frontline staff to focus on customers—not manage tech protocols. And when compliance drops, so does data reliability.

🧠 Rethinking Accuracy: From Counting to Understanding

The industry benchmark of “98% accuracy” often stems from lab conditions—a clean metric that masks a messy reality. Real retail environments introduce passers-by, repeat entries, delivery drivers, and of course, staff movements. Once these are factored in, usable accuracy can fall closer to 70–75%.

This isn't just a tech issue—it’s a strategic one.

✅ Passive Exclusion Through RE-ID and Behavioural Analytics

New exclusion models shift the burden away from store teams and towards behavioural intelligence. These methods use anonymized Unique User IDs (UUIDs) and pattern recognition to classify individuals without the need for wearables.

Zone-Based LogicIf a UUID crosses into a staff-only area, like a storeroom, they’re tagged as staff for the entire day.

Time-Based LogicVisitors lingering well beyond average dwell times (e.g., 3+ hours vs. 20-minute shopping average) are flagged as staff—even with intermittent exits.

AI Uniform RecognitionWhere store staff wear distinctive attire, computer vision can distinguish them from customers or even identify delivery roles.

Together, these create a layered exclusion model that’s automated, passive, and scalable across locations.

📊 Why It Matters: Impact on KPIs

Clean exclusion directly influences the reliability of key metrics:

• Conversion Rate: Exclusion prevents staff from inflating footfall, giving a true read on customer productivity.

• Zone Heatmaps: Eliminating staff traces from analytics uncovers authentic shopper paths and hotspots.

• Dwell Time & Engagement: Corrected averages help retailers design better in-store experiences.

🏢 The Portfolio Advantage for Mall Operators

At scale, the implications grow deeper:

• Lease Validation: Clean data prevents staff-heavy stores from appearing underperforming.

• Cross-Tenant Benchmarking: Standardized exclusion logic supports consistent comparisons.

• Investment Decisions: Precise traffic insights support smarter redevelopment, planning, and asset management.

✍️ RFPs Need a Reset: What Buyers Should Ask

Procurement teams and RFP authors have the power to shape this shift. Here’s what to consider:

• Request breakdowns: Accuracy by unique vs. total visits

• Push for real-world contamination filtering

• Ask for exclusion logic: Time-based, zone-based, or visual

• Demand GDPR-compliant architecture with on-premises processing and anonymization

When these questions are built into the tender, better systems rise to the top.

💬 Final Thought

People counting is no longer just about traffic volume—it’s about decision confidence. Retailers, mall groups, and portfolio managers deserve metrics that reflect reality, not optimistic lab scenarios. As exclusion technology advances, it’s time the way we assess it does, too.