A furniture and home furnishings retailer in Central Europe had its complete selected counting set verified at one store.

Four selected days in early March, followed by one more in May. A visitor crosses into one of the numbered zones of a home furnishings store, and a counting line records the movement.

Twenty-two interior crossing lines supplied data to nine zones. Another line measured people who passed the store without entering.

The entrance was the smallest part of what was measured.

The store

Nine zones in, the entrance stops being the only question

A retailer that measures nine interior zones wants to understand more than how many people entered. It wants to know where visitors went, which areas attracted attention, and what each part of the floor contributed.

An entrance is a defined directional crossing. Interior boundaries are less straightforward. Visitors can approach them from different angles, pause near them, double back, or cross the same line several times during one visit.

This deployment measures those interior movements. It also includes a line for people who passed the store without entering. That pass-by count separates a quiet period outside from a period when many people passed but few came in. The entrance count alone cannot show the difference. The dining area inside the store is measured in both directions, recording movement into the zone and back onto the wider store floor.

Every zone metric depends on a count taken at a line. If that count is wrong, the conclusions built on it are also affected.

An interior line is where a store begins to understand what its floor is worth. It is also harder to measure than the entrance most stores quote.

The check

The complete counting set was assessed together

The retailer decided where, when, and how much was tested.

The verification covered the entire selected counting set:

• 27 counting lines
• 13 cameras
• 9 interior zones
• 5,105 visitors
• 144 hours

Indivd’s count was compared with an independent count of the same hours. The complete set was assessed rather than selecting only particular lines or zones.

The store achieved 97.34% absolute accuracy, reported at a 99.99% confidence level.

Every counting error was included. A missed entry, a missed exit, and a person counted even though they did not cross were each recorded as errors. Errors in opposite directions were not allowed to cancel one another.

The customer decides where, when, and how much is tested, can run the check without Indivd present, and can put the same test in anyone else’s hands.

What the number carries

Store performance that can be read zone by zone

A home furnishings store in Central Europe achieved 97.34% absolute people counting accuracy against an independent count of the same hours. The verification covered 5,105 visitors and 144 hours across 27 counting lines on 13 cameras. The selected counting set included nine interior zones, a dining area measured in both directions, the entrance, and pass-by traffic.

This is a result about a complete deployment, not an isolated counting point. Interior boundaries behave differently from entrances, and the selected lines measured several types of movement across the store.

For the retailer, the value is the ability to use zone data with greater confidence. The retailer can assess which areas attract visitors, which are passed by, what the dining area returns to the wider floor, and how many people outside become store visitors.

Each of those measures depends on a count checked at the line where it was produced.

An error at an interior boundary can influence decisions far beyond that zone. An undercount can make an area appear less attractive, leading to reduced floor space or stock. An overcount can direct investment toward an area whose apparent performance was created by the counter.

When the complete counting set is verified, zone decisions are more likely to reflect visitor behaviour rather than differences in measurement.

In a separate engagement, KPMG audited Indivd’s counting accuracy at a major retailer’s store against manual ground truth. Every camera tested came within 1.17%, the strongest result of any system included in that audit.

The retailer can now ask a question an entrance count alone cannot answer: not only how many people came in, but how far into the store they went.

FAQ