By Smart Urban Sensing | April 2026
If you’re still relying on a legacy people counter — whether it’s an old infrared beam sensor, a basic camera system, or a WiFi probe setup — this guide is for you. Yesterday we published our complete Q&A guide to AI people counting sensors, covering the technology, accuracy, privacy, and ROI of every modern sensor type. Today we tackle the next question: how do you know when your legacy people counter needs replacing?
Thousands of UK retailers, shopping centres, councils, and venue operators are still running people counters installed five, eight, even ten years ago. The hardware still powers on. The dashboard still shows a number. But that number is quietly undermining every decision built on top of it — from staffing rotas to marketing ROI to funding reports.
Here are seven clear warning signs that your legacy people counter is costing you more than it’s saving — and what to upgrade to.
Sign 1: Your Legacy People Counter Accuracy Is Below 85%
The problem: Traditional infrared beam counters work on a simple break-beam principle — anything that interrupts the beam increments the count. Three people walking abreast? One count. A pushchair followed by a parent? Two counts. A member of staff stepping outside for a delivery? Another count added to your visitor total.
The result is accuracy between 60% and 85%. If your true daily footfall is 1,000 but your counter records 780, your conversion rate calculation is inflated by 28%. Your marketing team thinks the last campaign underperformed on traffic. Your ops team thinks staffing levels are fine. Both are wrong.
The fix: 3D stereo vision and Time-of-Flight sensors use depth mapping and AI tracking to identify and count individual people — not beam breaks. They handle groups, resolve direction of travel, and consistently deliver 99% accuracy. According to the Mordor Intelligence people counting market report, the global shift toward AI-powered counting is accelerating rapidly. We explain exactly how the technology works in our AI sensors guide.
Sign 2: You Can’t Exclude Staff From Your Data
The problem: Every time a member of staff walks through the entrance — arriving for a shift, stepping out for a break, crossing between zones — they inflate your visitor count. In a store with 10 staff each entering and exiting twice per shift, that’s 40 phantom visitors per day contaminating your data.
If your raw count reads 540 when the true customer count is 500, your reported conversion rate drops by 7.4%. Over weeks and months, this corrupts staffing models, marketing attribution, and space utilisation reports. A legacy people counter based on beam technology has no mechanism to distinguish staff from customers.
The fix: Modern AI people counters offer multiple staff exclusion methods — from physical tags detectable by ToF sensors, to AI-powered uniform recognition, to anonymous movement signature learning. The system subtracts staff crossings in real time, giving you a clean customer-only count. We cover all six methods in detail here.
Sign 3: No Bidirectional Counting
The problem: A single-beam IR sensor counts interruptions. It doesn’t know whether someone walked in or walked out. Without bidirectional counting, you cannot calculate live occupancy, produce accurate entry-only visitor counts, or identify net flow patterns across different times of day.
Some legacy systems attempt directionality using two parallel beams — but accuracy degrades significantly in busy doorways where multiple people cross simultaneously.
The fix: AI-powered sensors track each individual’s trajectory across the full field of view, assigning unique tracking IDs and calculating directional vectors. Entry and exit are logged separately, and live occupancy is maintained as a running net total — foundational to fire safety compliance under the Building Safety Act 2022.
Sign 4: Your Legacy People Counter Supplier Has Gone End-of-Life
The problem: The people counting market has consolidated significantly. Several brands common five to eight years ago — Irisys legacy models, Brickstream, certain Axis integrations — have discontinued product lines, stopped firmware development, or been acquired.
If your legacy people counter manufacturer no longer issues firmware updates, you face three escalating risks:
- Security vulnerabilities go unpatched on a device connected to your network.
- Accuracy degrades over time without AI model improvements.
- Spare parts dry up — when a unit fails, you’re sourcing replacements on eBay or cannibalising from other sites.
One retailer was paying a premium maintenance contract on hardware that hadn’t received a firmware update in four years. The annual service fee alone exceeded the cost of a full modern replacement.
The fix: Our Trade-In Programme handles the migration from your existing legacy people counter to the current AI 3DPro2 platform, including data continuity planning.
Sign 5: You Can’t Integrate With POS, BMS, or Analytics
The problem: Legacy people counters were designed as standalone systems. They count. They store a number in a proprietary dashboard. That’s it.
But footfall data only becomes genuinely valuable when combined with other datasets: POS transaction data (for conversion rate), staff scheduling data (for traffic-to-labour ratio), BMS data (for occupancy-driven HVAC control), or marketing calendars (for campaign attribution).
If your current system doesn’t support MQTT, REST API, Modbus RS485, or BACnet — or locks your data behind a proprietary portal — you’re sitting on a number that can’t talk to anything else.
The fix: Modern AI sensors are built for integration. The Smart Urban Sensing platform publishes real-time data over MQTT, exposes a full HTTP REST API, and supports BMS protocols including Modbus RS485 and BACnet. Data flows directly into Power BI, Google Sheets, or any third-party platform. See our integration guide for technical detail.
Sign 6: Your GDPR Compliance Is Uncertain
The problem: If your legacy people counter uses WiFi probe counting, it is almost certainly non-compliant. WiFi probe counting relied on capturing MAC addresses — classified as personal data under GDPR Article 4. Following universal MAC address randomisation in iOS 14 and Android 10, accuracy collapsed below 50%.
Even 2D camera systems with cloud-based video processing carry a significant GDPR burden. Many legacy deployments were installed before GDPR came into force in 2018 and have never been audited.
The fix: Depth-based sensors (ToF and 3D stereo) output distance measurements only — no colour data, no facial features, no biometric information. Edge AI cameras process video on-device and transmit only anonymous count metadata. This is privacy by design. We cover this in our Privacy-First AI guide.
Sign 7: You’re Paying Too Much for Basic Data
The problem: Many legacy people counter contracts bundle hardware, software, and analytics into an annual SaaS fee that escalates at each renewal. Five years of recurring fees often total two to three times the original hardware value — for a system delivering nothing more than a basic daily count.
We’ve seen annual contracts of £1,200–£2,500 per store for legacy systems that deliver less functionality than a modern sensor costing £795 outright.
The fix: The Smart Urban Sensing model is transparent: you own the hardware outright from day one. Basic counting, bidirectional tracking, and real-time occupancy data come included. Use our Pricing Calculator to build a quote, and our ROI Calculator to see the payback timeline.
How to Replace Your Legacy People Counter: The Upgrade Path
If you’ve recognised two or more of these signs, here’s the practical next step:
- Audit your current system — Document accuracy, integration capability, GDPR compliance status, annual cost, and remaining contract term.
- Benchmark against modern alternatives — Our AI sensors guide covers every sensor type. Our Buyers Guide 2026 compares specific products.
- Calculate the ROI — Use our ROI Calculator to model savings from improved accuracy, staff exclusion, and eliminated SaaS fees.
- Explore the Trade-In Programme — Our Trade-In Programme provides a structured legacy people counter replacement path with data continuity.
- Book a site survey — We assess your entrance geometry, network infrastructure, and integration requirements at no cost.
The Bottom Line
A legacy people counter that was state-of-the-art in 2018 is not state-of-the-art in 2026. The technology has moved on — from beam breaks to 3D depth mapping, from standalone dashboards to real-time IoT integration, from GDPR grey areas to privacy by design.
The data from your people counter feeds your conversion rate, your staffing model, your marketing attribution, your energy management, and your compliance reporting. If that data is wrong, everything built on top of it is wrong too.
If any of the seven signs above sound familiar, the question isn’t whether to replace your legacy people counter — it’s how quickly you can stop making decisions on unreliable data.
Ready to See It in Action?
Discover how our AI-powered people counters can transform your space.
Book a free 20-minute footfall review | Trade-In Programme | Pricing Calculator
Retail and shopping centres
Find out what your entrance data should be telling you
In 20 minutes we review your entrances, current counting setup and conversion questions, then tell you what accurate footfall data would change. No obligation, no hard sell.
