The Eye in the Sky
By David Dorf-Oracle on Aug 03, 2009
Not long ago I was in a Walmart with my three kids to pick up a few bath items as a gift for my wife. For some reason the self-checkout lanes had been removed, so we were forced to use the express lane. There were three people in front of me, and the person checking out was trying to use their credit card. Swipe after swipe it would not work. The line got longer, and after 5 minutes I set my items down and walked out. I don't know what the problem was, and I didn't really care. If a retailer can't get me checked-out within a reasonable amount of time, I walk.
Queue management is a huge part of retail operations, especially for scan-and-bag retailers. Too many open checkouts and you're overspending on labor; too few and you risk frustrating customers leading to lost sales. Regardless of whether the cashier paged for help, the manager should be aware when lines get long and take corrective action.
Tesco uses Irisys to count the people in line using infrared technology. When the queue gets long, the system alerts management to open more checkout lanes. Under Amour recently deployed Sensormatic's people counters, which also monitor traffic throughout the store.
Bu there's so much more data to be collected. Just as sophisticated video systems can detect theft as I described in this posting, it can also be used to mine customer statistics, such as queue depth, traffic patterns, and consumer behavior.
We collect sales data and mine it for price affinity, cross elasticity, and product affinities, so why can't we mine video to better understand how customers shop. Which displays attract them? When end-caps hold their attention longest? Which floor models draw the most attention? Which items do they pickup, then put back?
It will be interesting to see how retailers use video analytics to improve the shopping experience, or a least get me checked-out faster.