How to Optimize Store Performance in Real Time | NTT DATA

Wed, 21 September 2022

How to Optimize Store Performance in Real Time

NTT DATA in store tracking allows physical stores to track and influence shopping journeys with the same ease offered by online tracking

Optimize Store Performance in Real Time with In-Store Tracking

Physical stores have long been at a big disadvantage in comparison with digital channels because they have little detailed knowledge of how consumers behave when they are inside a store.

Unlike an online retail business, physical stores lack cookies to track where shoppers go and what they do and recognize them on their return visits. So a store’s options for increasing basket size or promoting specific products or brands are limited to basic marketing tactics based on customer profile and sales.

By analyzing the sales data for products on the planogram, a retailer can optimize the use of space. But a planogram is based on aggregate, historic sales data and doesn’t give the retailer any insight into what products are selling to a particular demographic, for example.

This is a problem for retailers in today’s era of hyperpersonalization as they lack the sophisticated measurement tools and granular metrics that allow them to easily see just who is buying or watching what inside their stores and when.

In an industry in which the line between physical and digital is becoming increasingly blurred, retailers want a way to be able to track and influence shopping journeys in their stores with the same ease and unobtrusiveness as that offered by online tracking.

Obtaining real-time detailed metrics with NTT Data In-Store Tracking

NTT DATA In-Store Tracking provides an easy-to-use solution to this problem by analyzing the real-time images of consumer traffic generated by a store’s security cameras. This enables a retailer to identify the gender and age of their customers and anonymously track their journey through the store to the checkout.

The anonymized data is processed in real time to enable the purchase journey of each customer to be analyzed in detail. The processed data is then displayed on dashboards that a retailer can customize to obtain detailed metrics on the demographic make-up of their shoppers, how they move through the store, and what they buy at any given time or day of the year.

The ability to do near real-time analysis of shopping journeys is particularly important if a retailer wants to judge the effectiveness of targeted or time-limited promotions - a 20% discount just on cosmetics and parapharmacy at certain hours during the day, for example.

It’s important to stress that the video images are processed inside of the store to guarantee privacy and the identities of shoppers cannot be reconstructed from the abstracted data. The data from the various cameras is aggregated and combined with point-of-sale data before uploading to the cloud, so ensuring GDPR compliance.

The anonymized data is then processed in the cloud to generate a wealth of detailed metrics. In the case of customer traffic, these metrics include: the duration of each visit, time per section, cluster by gender and age, store capacity, the performance of exterior displays, and so on.

Other In-store tracking technologies

Other indicators for real-time customer behavior include: correlation between sections, ROI by section, the "temperature" of specific points of the store displayed as heat maps, and the propensity to cross-sell between product categories. These performance metrics can be obtained with the following technologies:

Heat maps

Heat maps use the temperature analogy to create a highly intuitive way of displaying metrics such as distribution of the number of visits throughout the week, the distribution of the duration of visits throughout the day, and the time of visit and number of visits to a particular section of the store.

Using heat maps, store managers can see the “cold” areas of the store that don’t perform as well as others in terms of visitor traffic or ROI, and they can better schedule staff shifts or the number of checkouts to respond to fluctuations in traffic during the day.

Category correlation matrices

Category correlation matrices are another powerful visual aid that can be produced with the NTT DATA solution. They show, in matrix form, the relative probability of shoppers visiting different categories depending on the first category they visited. 

A/B testing

Retailers can also use NTT DATA’s In-Store Tracking to perform A/B testing of different layouts and to execute real-time market campaigns with digital displays in the store or shop window, whose content changes based on the demographic make-up of the visitors currently in the store.

How can we help you?

Get in touch