Indivd
•
•
Intro
A footwear and accessories retailer in Southern Europe achieved at least 98.03% absolute counting accuracy in each of its three stores, each one tested as its own deployment rather than assumed from the first.
A specialty footwear and accessories retailer in Southern Europe had its people counts verified at three stores between March 2025 and May 2026.
At every location, Indivd’s count was compared with an independent count of the same hours. Each store achieved at least 98.03 percent absolute accuracy.
The result is the minimum across the three stores, not their average. The verification covered 1,394 visitors and 98 hours across six counting lines on three cameras.
Each store relies on one entrance line and one exit line. Together, those two lines produce the store’s complete visitor count.
There is no second measurement point inside the store to correct an error at the door.
The stores
Two counting lines carry the whole store
Each of the three stores is measured using one entrance line and one exit line on a single camera.
Those two lines supply the visitor data used to calculate conversion, plan staffing, and evaluate whether display or merchandising changes bring more people into the store. If the doorway count is wrong, every metric built on that count is also affected.
Accuracy is particularly important in footwear retail. A sale may require several sizes, alternative products, and assistance from a member of staff. Each visit can take significant time to serve.
For that reason, stores need to understand not only how many people visit, but also how effectively those visits become sales. Both conversion and staffing decisions depend on an accurate visitor count.
Accuracy is not something a model has. It is something a deployment produces.
At these stores, two counting lines carry the whole visitor dataset.
The check
Three stores assessed using one standard
The retailer selected the locations, dates, and hours included in the verification. The checks took place in different months and covered different amounts of time at each store.
Across the three locations, 1,394 visitors were included in the comparison over 98 hours. Indivd’s count was compared with an independent count of the same hours.
The verification covered six counting lines on three cameras. Each store’s entrance and exit lines were assessed together.
Every store achieved at least 98.03 percent absolute accuracy. This figure is the lowest result among the three stores.
Reporting the minimum matters because it describes the accuracy achieved at every tested location. An average could allow a stronger result at one store to offset a weaker result at another.
DEFINITION
Absolute accuracy is 100 percent minus the absolute error rate.
Every counting error is included. A missed entry, a missed exit, and a person counted even though they did not cross are each recorded as errors. Errors in opposite directions cannot cancel one another.
Absolute accuracy is 100 percent minus the absolute error rate. Every counting error is included. A missed entry, a missed exit, and a person counted even though they did not cross are each recorded as errors. Errors in opposite directions cannot cancel one another.
What the number carries
Comparable visitor data across three stores
Verified between March 2025 and May 2026: at least 98.03 percent absolute accuracy at each of three footwear and accessories stores in Southern Europe.
The verification covered:
• 1,394 visitors
• 98 hours
• 6 counting lines
• 3 cameras
Applying one measurement standard across the three stores gives the retailer a more dependable basis for comparing performance.
A difference in conversion between two locations is more likely to reflect a difference between the stores when both visitor counts have been assessed using the same accuracy measure.
Without verification, a counting error at the door can remain hidden. An overcount can make a store appear to be converting fewer visitors than it actually is. That could lead to unnecessary changes in staffing, stock, promotions, or merchandising.
FAQ
How do you choose a people counting system for retail?
Compare on measured performance rather than advertised accuracy. Ask how the system was tested at locations like yours, during representative hours, against what independent reference, and under which error measure. Every vendor can answer those four questions, and an advertised figure answers none of them.
What accuracy should a retailer expect from a people counter?
There is no standards body that certifies people counting accuracy, so the figure alone means little without the measure behind it. These three footwear stores each reached at least 98.03% under absolute accuracy, which counts every miss and every overcount in full rather than letting them cancel.
How accurate is footfall counting in footwear retail?
Across three verified footwear and accessories stores, every location reached at least 98.03% absolute accuracy against an independent count, covering 1,394 visitors and 98 hours on six counting lines.
What questions should you ask a footfall analytics vendor?
Which stores were tested, who chose the hours, what the count was compared against, whether errors in opposite directions were allowed to cancel, and whether the test can be repeated by someone else. The answers separate a measured result from a marketing figure.
Does a people counting system need to be tested after installation?
Yes, because accuracy is produced by a deployment rather than held by a model. Each of these three stores was tested separately for that reason, since a result at one location is evidence about that location only.
Can a third party verify Indivd's counting accuracy?
Yes. 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.