Retail analytics, both online and in-store, begin at the very same place – traffic. Traffic has always been a critical metric as it represents opportunity to serve and ultimately sell to shoppers. Traffic is also absolutely critical to determining very fundamental performance metrics, like conversion, for example, giving store operations professionals direction on what managerial levers to push and pull upon to achieve greater results.
Store traffic used to be calculated with simple human observation and a pencil and pad. Luckily, those days are long gone. Today, when it comes to determining store traffic, there are a variety of technologies that can count systematically. Important for retailers everywhere are two factors when counting store traffic: 1) Accuracy, and 2) Consistency.
Traffic needs to be accurate because store managers will base staffing decisions around peak traffic demands. Staff wages are simply too high of a variable cost to waste on slow traffic hours and days, and the opportunity cost – realized from a drop in conversion and sales – of being too leanly staffed during high traffic demand are prohibitive. Plus, of course, accurate traffic helps retailers best understand conversion, highlighting their ability to meet shopper needs once in-store.
Traffic counting also needs to be consistent, meaning the variables that affect it or could cause error are consistent. Traffic counting needs to be consistent or a retailer won’t be able to confirm, for example, if traffic decreased due to a layout change in the store or if some other variable, like the door clicker took extra bathroom breaks that day, affected traffic counts.
Recently, I’ve been involved in discussions about Wi-Fi/MDD (mobile device detection) as an inexpensive replacement for state-of-the-art traffic counting systems, and as someone who has spent her entire career in retail and knows the importance of accurate data and comprehensive insights to what’s going on in store, I want to explain the capabilities and limitations of this technology.
How mobile device detection works – a crash course for non-engineers
Sensors in a retail store are used to pick up Wi-Fi signals from smartphones. Those smartphones aren’t necessarily connected a Wi-Fi hotspot: a phone in your pocket or purse may periodically look for a Wi-Fi connection so that it always has a good data connection available. When a device is found, retail analytics systems will keep track of it by its MAC address, which is like a unique serial number for the phone.
However, not every smartphone in the store will be detected, and not everyone in the store is carrying a smartphone, so Wi-Fi sensing can only give you a sample of data about your shoppers.
The key benefit of Wi-Fi detection is that unique identifier, which provides two key data points that you can’t get from traffic counting alone: 1) the approximate time a shopper spends in the store and 2) how frequently shoppers visit. In some settings, Wi-Fi detection can also be used to approximate passby traffic.
Combining data points
Traffic counting is meant to give you precise, accurate data so that you can compare traffic data with POS transactions and staffing allocation down to the minute.
Wi-Fi data alone can’t be used this way: it’s just not granular enough to calculate critical retail KPIs like conversion and shopper-to-staff ratio.
On the other hand, having Wi-Fi data along with traffic counts can help add context to an analysis: if traffic is consistent but conversion is down, understanding a change in the mix of new vs. repeat shoppers may reveal the cause.
Better solutions, better insights, better decisions
While accurate traffic counting (combined with POS data) is the foundation of retail analytics, it can’t do the complete job by itself, like, for example, segregating employee tracks from shoppers, and excluding employees from critical data streams like traffic counts, dwells, etc. For that, a store needs to integrate additional technologies – in the case of RetailNext’s Traffic 2.0 solution, it’s a combination of stereo video analytics, Bluetooth and Wi-Fi, integrated seamlessly into a single, comprehensive software platform.
Other interesting technologies that help add greater texture to traffic counts include data from solutions like Foursquare and Verve. They don’t help out with accuracy – as proof, go to an app like Swarm or even Google Maps and see where the app thinks you are currently (if you’re in a mall, it could be one of dozens of stores!). My Google Maps often thinks I’m in the Bay in San Francisco, about half a mile from my apartment. Literally in the water. But, their real strength lies in the rich amounts of user information that is readily shared through an opt-in environment.
Join the #retail, #inspiringretail and #SmartStore conversations on Twitter @Lauren_isRockin & @RetailNext, as well as at www.facebook.com/retailnext.
What Every Retailer Must Know About Traffic 2.0
Traffic 2.0 is the latest generation of traffic counting, and it doesn’t stop at “how many.” Today, it includes “who,” “what,” “where,” and “when,” and when brought all together, goes a long way to determining “why” and “why not”. Download this ebook to learn more about Traffic 2.0 and how you can implement in your stores.