People Counting vs. Wi-Fi Analytics: Which Should You Use?

If you've started researching foot traffic analytics, you've probably come across two very different approaches: dedicated people counting sensors, and Wi-Fi analytics that track devices connecting to (or simply searching for) your store's wireless network. They sound similar but measure fundamentally different things.

How each technology actually works

A people counting sensor like Nano uses depth-based computer vision, mounted above the doorway, to directly detect and count people crossing a specific point. It's measuring the physical event of someone walking through the door.

Wi-Fi analytics works differently: it detects smartphones broadcasting Wi-Fi probe requests as they search for known networks, then estimates visitor counts and dwell time from that device activity. It's measuring a proxy for people — phones — rather than people directly.

Where Wi-Fi analytics falls short

Not everyone's phone participates. Wi-Fi scanning behaviour varies by device and operating system, and many modern phones randomise their Wi-Fi identifiers specifically to prevent this kind of tracking, which undercounts visitors.
One person, multiple devices (or zero). Someone carrying a phone and a tablet can register as two visitors; someone who left their phone in the car registers as zero.
Range is imprecise. Wi-Fi signals can be picked up from the footpath outside, the store next door, or the level below in a multi-storey building, making it hard to know exactly who was actually inside.

This is why Wi-Fi analytics typically lands well below the 99% accuracy that dedicated depth-sensing hardware achieves — it's estimating from indirect signals rather than directly observing the doorway.

Where Wi-Fi analytics does make sense

Wi-Fi analytics isn't without merit. If you already have the network infrastructure in place and want a rough, low-cost signal of relative traffic trends — not a precise count — it can be a reasonable starting point, especially across a large number of sites where installing dedicated hardware everywhere isn't yet feasible.

The privacy angle

It's worth being direct about this: Wi-Fi analytics tracks device identifiers, which raises more privacy questions than a sensor that never captures identifiable data in the first place. Nano's approach — on-device depth processing that only ever transmits anonymous counts, detailed on our AI & Technology page — sidesteps this issue entirely, since there's no device identifier or personal data involved at any point.

Our honest take