26 August, 2015 | Nader Fathi
The people-counting industry is at a crossroads. Declining foot traffic in brick-and-mortar stores has made retailers much more discerning about the accuracy of their traffic counters. The data can arm them with crucial insights for more efficient staffing, store layouts, inventory management, and more. But the question remains: What is the most reliable traffic counting source?
Recently, more and more in-store analytics companies are pushing camera solutions and abandoning Wi-Fi analysis due to Apple iOS randomization. Kiana Analytics set out to put the question to rest and ran a study for a large-format women’s apparel retailer, comparing the accuracy of cameras to Wi-Fi. Kiana found that people exiting the purview of cameras when they went out of range, e.g. into a restroom or a dressing room, were generating multiple traffic counts. The camera could not distinguish whether the same customer just re-entered the store and often double counted the customer’s path.
Notably, traffic counting via Wi-Fi only counted that shopper as ONE person in the store.
And the retailer’s store managers were much more interested in learning how many actual people were in their stores versus erroneous traffic counts. Wi-Fi-enabled traffic counting does not depend on customers staying within a camera’s range; and while it does depend on shoppers having Wi-Fi, chances are they will. IBM Mobile found that 91% of mobile users keep their devices within arm’s reach 100% of the time, and 90% of all smartphones are equipped with Wi-Fi capabilities (HuffPo). What’s more, there will be more than 7 billion new Wi-Fi enabled devices in the next 3 years (Strategy Analytics).
So, for those 91%, Wi-Fi detection is comprehensive and accurate, while cameras are blind to invisible zones and prone to duplicating traffic counts. Wi-Fi discerns “new vs. repeat” customers. The sum of “new and repeat” customers presents the number of distinct people who visited a store during a specific timeframe. This also eliminates double counting, making it a more reliable source to calculate conversion. Moreover, cameras cannot “see” pass-by traffic while Wi-Fi can detect it in order to evaluate “entry vs. pass-by” ratios. These analytics feed store managers valuable data about how many shoppers passed by compared to how many entered the store, which gives them insight into efficacy of, for instance, window promotions.
If the study above is not a clear indicator that retail traffic counting methods can have varied accuracy, simply take a look at the discrepancies between retail traffic indexes compiled by leading analytics solutions. During the 2014 holiday period, one traffic index reported that retail store traffic was down 14.6 percent compared to the prior year. Another index reported that during that same holiday timeframe retail traffic was down 6.5 percent; less than half of the other report! While there may be varied retailers and store footprints within these indexes, along with other factors that contribute to starkly different results, with millions of holiday visits counted, we have to ask “how many unique paths may have been missed or counted multiple times?”
Kiana’s study revealed that it was difficult to determine how many distinct customers visited a store on a specific date using only cameras. Wi-Fi-enabled analytics uses advanced data mining to capture the number of unique visitors, how much time they spend in a store, and the frequency of their visits. This accuracy and in-depth understanding of traffic counts and flow provide more reliable metrics to improve operations and deliver a better shopper experience.
To learn more about Kiana’s Wi-Fi enabled analytics, visit www.kiana.io