It has been 15 years since Irisys introduced the world’s first high accuracy ceiling mounted people counter and since then the use of people counting to provide essential retail performance metrics has grown in popularity. People counting enables a critical retail metric – conversion rate – to be easily calculated for a retail store. Conversion rate – the ratio of revenue against people count measures how effectively a retail premises turns customers into money and can hence be used to optimise store performance, measure the impact of advertising campaigns, understand seasonal variations in performance and compare performance store to store to promote best practice.
Over the years new features have been introduced into people counters – the ability to count based on customer height (a simple method for eliminating children from the count) and the ability to count families (a family of 5 should not be considered the same as 5 individuals) are just two examples. Irisys is at the forefront of such technological innovations, the latest Vector 4D product provides enhanced group counting and the ability to resolve height with world leading accuracy. However, both of these enhancements must be used with care – the use of height alone to discriminate children is open to significant error as any parents of 8-12 year olds will clearly understand. Similarly, whilst it is simple for a human observer to understand that a group of people looks like a family the same cannot be said for a people counter – rules are applied and unfortunately mistakes can be made.
There is a trend for users to only look at group counts or child free counts when comparing store performance and in doing this important information, insight and accuracy in the resulting conversion rate figures is lost. Here at Irisys we have always considered providing an accurate count of all of the people entering or leaving a store as the core metric – it is the baseline truth against which all other analytics can be interpreted. Use of height or grouping information should only be used as additional context to help understand differences in store performance and diagnose the root cause.
For example two stores may display different conversion rates based on the baseline truth people count – in understanding the reason for this difference it would then be useful to see if there were significant differences in the number of children or the number of groups entering the stores or whether the store was simply less efficient.
For this reason Irisys provides a flexible counting system allowing counts of every individual, counts of groups or counts according to height range or any combination of these. Similarly Irisys Vector counters accumulate statistics of height allowing the retailer to understand how the distribution of height in a store varies over time for example, are there more children at weekends? Does this change how a store should operate at these times?
The use of these enhanced metrics undoubtedly can provide useful insight – but don’t throw information away – start with ground truth and then use the extra information to help you understand what is going on.