Video-based customer-counting system increased data accuracy by 42 percent
- By Rich Mellott
- Aug 01, 2011
For The Limited, an established U.S. women’s fashion retailer with
220 stores and 2,600 employees, finding better ways to manage and
improve customer service, staffing levels, merchandising effectiveness
and profitability is key to continued success.
As with all retailers, The Limited maintains a careful balance between
staffing and cost, ensuring enough salespeople are on hand so that customers
aren’t kept waiting but without overstaffing to the point of inefficiency and
“We need enough staff at any given time to communicate with customers, help
them find their size and maybe try on some different outfits,” said Roger Coville,
CEO of the The Limited. Such personal attention and overall experience helps
convert store traffic into sales and builds customer loyalty.
The more timely and accurate the customer traffic data is, the easier it is to
maintain proper staffing levels. For years, The Limited generated this critical information
manually, via store estimates and hand counters. The main goals: keep
the staff-to-customer ratio fine-tuned to minimize labor costs while maximizing
the customer experience; compare the traffic count to sales to understand how well
each store converts visitors to buyers; and, ultimately, be more profitable.
With new technology opportunities, it became clear that there were better solutions
than manually gathering and analyzing data.
The Need for a precise Automated System
The Limited wanted an automated in-store traffic counting and analysis system
that could help managers keep stores staffed with the optimal number of employees,
enhance the customer experience and maximize sales.
Because its stores are small, with just a few salespeople at any given time, The
Limited sought a system that delivered highly accurate traffic counts to manage
personnel optimally throughout store hours.
In addition, management wanted the ability to correlate traffic information with
key data, such as sales results, to gain better insight into customers’ purchasing habits
and to track shoppers-to-buyers conversion ratios. They wanted a “dashboard”
that consolidated multiple data inputs into a real-time view of business operations to
better manage store efficiencies and maintain competitive advantage.
Clearer Images, Greater Accuracy
The Limited’s first attempt at automation was a thermal imaging system set up
as a one-store pilot program. The system detected shoppers entering the store by
sensing body heat. At night, raw data transmitted to the thermal imaging vendor’s
headquarters were converted to a traffic count and posted on a website the
following day. But this system was limited by both time delays and questionable
customer count accuracy.
Coville discovered a better solution at a trade show, where video and data
management software provider DIGIOP Inc. demonstrated its video-based retail
counting system calculating the number of passing trade show attendees in real
time. The integrated system included DIGIOP software, powered by a purposebuilt
Dell computer, and a video camera.
“The real-time images meant the traffic count could be independently verified
by looking at the visual record. This capability excited us,” Coville said. “An instore
comparison of the DIGIOP system to the thermal imaging system revealed
a dramatic difference. There was a 30-person discrepancy between the two counts,
and the video record proved that the DIGIOP system was accurate. We were sold.”
Based on those results, The Limited implemented DIGIOP’s video management
system solution in all 220 of its stores.
The integrated system enables The Limited to capture and analyze in-store data
to optimize the customer experience while minimizing labor costs. Additionally,
The Limited’s employees are compensated partly on sales conversion rates, making
traffic counting accuracy very important to all. The DIGIOP system has satisfied
this requirement, delivering images with accuracy rates far superior to the thermal
A Powerful, Scalable Solution
Today, video cameras are connected to a DIGIOP Digital Video Management
System in each store to gather intelligence. The Limited uses up to six cameras per
store—one positioned at the store entrance to count traffic, and the others aimed
at the checkout area to help prevent theft and monitor checkout lines and customer
wait times. A video capture board converts the analog video signals to digital,
enabling a number of useful analytics. For example, the digital video can be correlated
with other data sources, such as point-of-sale transactions, and the DIGIOP
system can automatically classify objects, such as people, within the video.
Each DIGIOP system produces an in-store traffic count every 15 minutes and
transmits this data to a secure website for access and analysis by both store managers
and corporate managers.
The system provides traditional surveillance benefits of increased security and
risk management, as well, and The Limited plans to add complex loss-prevention
tools in the future.
“Our checkout desk cameras can be used for loss-prevention management,” Colville said. “If cash drawers open
without corresponding sales, the system
would alert us so that we could
quickly check for shortages.”
The Limited also can link the system
to third-party systems for employee
“It’s easy to add capabilities because
the system is digital, PC-based, and
simply ties into our back-office systems,”
Coville said. “New capabilities
are modular, so we can grow as we go.
We are exploring additional capabilities
to create stores of the future.”
The Limited has been impressed
with the system’s clear and crisp video
quality, which is important for surveillance
purposes as well as capturing images
accurately so corresponding data
are also reliable. And because the DIGIOP
solution is scalable, The Limited
can easily add more analog, IP and/or
megapixel cameras to any given store
without having to switch to a new hardware
In the retailer’s initial test of recognizing
and counting customers, Coville
noted that the video system measured 91
percent accuracy, versus the 64 percent
accuracy of the thermal imaging system—
a 42 percent improvement. These
accuracy levels have been further confirmed
in the full installation at all 220
stores, where The Limited is now scheduling
staff and calculating sales conversion
rates based on this intelligence.
Retail officials have confirmed labor-
saving benefits, as well. Labor is the
retail industry’s number-one cost and
a huge profitability factor. The videobased
system has empowered The Limited
to match staffing levels to customer
levels accurately, helping each store optimize
Additionally, The Limited has seen
an uptick in conversion rates of shoppers
to buyers since it installed the DIGIOP
“The traffic data comes back and
is available immediately to our corporate
office and to the stores,” Colville
said. “We aggregate that data with our
sales information to provide conversion
rates per store. We also integrate
the data with point-of-sale data systems
and generate conversion rates almost
instantly. We have improved rates
nationwide by a couple of percentage
points—on that scale, 2 percent adds
significantly to the bottom line.”
The system delivers business intelligence
in real-time, enabling managers
to quickly react to store performance.
For example, a low conversion rate on a
particular day might prompt increases
in staffing levels or training. Similarly,
managers might see a midday traffic
pattern related to nearby office workers
shopping during lunch breaks and assign
This article originally appeared in the August 2011 issue of Security Today.