Everything About People Counting Solutions & Features

people counting software — FAQ About People Counting Software | Vemco Group

Written by Admin | Jul 23, 2026 10:21:38 AM

Most buyers don't struggle with the concept of counting people — they struggle with trusting the number. The question underneath almost every demo request is the same: can I make a staffing or leasing decision on this data without getting burned? That's the lens this FAQ uses. These are the questions that actually come up when retailers, shopping centres, airports, universities, and facility teams sit down to specify a system, based on what Vemco Group has seen across 2000+ customers processing more than 85 million counts a day.

How accurate is people counting software, really?

Be suspicious of any vendor who quotes a flat 99% on a brochure. Accuracy depends on physical conditions — lighting, entrance width, ceiling height, and how visitors actually move. Vemco commits to a contractual minimum of 96%, and in practice sees 98–99% when conditions allow. The gap between those numbers is where installation quality lives. A 3D sensor mounted correctly over a clean single-file doorway will sit at the top of that range. The same sensor over a wide sliding-door entrance with backlighting from a glass façade at sunset will not, until the mounting and angle are corrected. The honest answer is: accuracy is a range you engineer toward, not a fixed spec you buy.

Won't my own staff inflate the counts?

This is the single most common reason a store manager loses faith in a dashboard. If your five-person team walks in and out forty times a shift, your conversion rate collapses and nobody believes the system again. Staff-exclusion algorithms handle this by filtering employee movement — often via staff-entrance logic, wearable tags, or behaviour patterns — so the traffic figure reflects customers, not colleagues. Confirm during setup that exclusion is switched on and tested, because a system installed without it produces numbers that look plausible and are quietly wrong.

Am I locked into one sensor brand?

You shouldn't be. Vemco's platform is sensor-agnostic, working across partners including Xovis 3D AI, Milesight, Hikvision, and AXIS. That matters for two reasons. First, you can match sensor cost to the job — a small library branch doesn't need the same hardware as a multi-terminal airport. Second, if you've already installed counters in some sites, you're not forced to rip them out. A practical note from the field: mixing sensor types across a portfolio is fine, but keep the analytics layer unified. The moment you have three vendor dashboards, nobody compares sites and the whole exercise stalls.

What can AI sensors actually detect?

Modern AI sensors go past a raw headcount. They can estimate age and gender bands and, importantly for shopping centres and family attractions, separate children from adults. That distinction changes decisions: a centre reporting 10,000 "visitors" behaves very differently when it knows a third are children who don't shop independently. Universities and libraries use the same capability to understand who is using study spaces versus passing through. Demographics are aggregated patterns, not individual identification — a point worth raising early with privacy officers and works councils.

How does it fit with the systems I already run?

Counting data is nearly useless in isolation. Its value appears when traffic sits next to sales, staff hours, and space data. Vemco integrates with ERP and BI tools and offers hosted or private cloud deployment, so a facility manager can keep data inside their own environment where policy demands it. The module set is built around specific jobs rather than one generic dashboard:

  • VemCount — core footfall and conversion analytics.
  • VemTrack — movement and flow through a space.
  • VemTenant and VemLease — traffic reporting and lease context for landlords and centre managers.
  • VemSpace — occupancy and space utilisation.
  • VemFusion — combining data sources into one view.

What does each type of site actually use it for?

The same platform answers very different questions depending on who's asking. Retailers watch conversion rate — the share of visitors who buy — and align staff rosters to the busiest hours instead of a fixed schedule. Shopping centres use tenant-level traffic to justify rents and prove marketing campaigns brought people in. Airports track flow to open lanes before queues form. Universities and libraries measure occupancy to plan opening hours and space, and increasingly to defend or reclaim square metres in budget reviews. Facility managers use occupancy data to tune cleaning and HVAC to real use rather than a fixed timetable.

How long before the data is trustworthy?

Hardware goes live in days; confidence takes a few weeks. You need a baseline period to see normal weekday and weekend rhythms before the numbers mean anything for decisions. A practitioner habit worth adopting: cross-check the first week of counts against a manual count at one busy hour and one quiet hour. If the system reads within its stated range against your manual tally, the site is calibrated. If it doesn't, the fix is almost always physical — sensor angle, a reflective floor, or a door held open by deliveries — not the software.

Is it worth it for a smaller operation?

A single-site retailer often gets faster payback than a chain, because one clear insight — say, that Saturday afternoon traffic peaks two hours before you scheduled extra staff — pays for the system quickly. The scale question is less about cost and more about whether someone will act on the reports. Data that nobody reviews returns nothing, regardless of accuracy.

Why does vendor track record matter here?

Counting looks simple until you're troubleshooting a site with odd geometry or comparing thirty locations that were installed by different teams. Vemco has worked on this since 2005 — twenty years in 2025 — from its Fredericia R&D centre, with partners across 95+ countries. That history matters less as a marketing line and more because the awkward cases have been seen before, so the answer to your specific problem usually already exists.

Talk to someone who's specified your kind of site

If you're weighing accuracy guarantees, sensor choice, or how counting data will feed your existing ERP and BI stack, get specifics for your building rather than a generic quote. Contact the Vemco Group team to discuss your site layout, the questions you need the data to answer, and how to reach the top of that 98–99% accuracy range.