Hikvision Dual-lens DeepinView People Counting cameras installed at Dutch retail shopping centre

The retail sector has felt the disruptive impact of e-commerce more than most. The inevitable costs of staffing, rent, maintenance and daily operations has left traditional brick-and-mortar stores struggling to stay dynamic and compete effectively with their more agile online counterparts.

Not only that, but the one competitive advantage of physical retail outlets - being able to provide a more personal customer experience - has also taken a hit in recent years, thanks to a lack of in-store and operational efficiency. Customers find it hard to locate the products they are shopping for, are faced with longer queues for payment, and the most valuable ‘VIP’ customers are often overlooked. At a time when retailers should be striving to excel, this is bad news, as a 2018 study from KPMG reinforces:

“The customer experience is more important than ever as retailers are striving to differentiate themselves in a challenging and crowded market. Consequently, experience per square foot will be the new retail metric to measure success.”

Technology for better customer experience

One challenge that retailers often face is not having a way to collect accurate data and visualise it intelligently in order to make smarter business decisions about the shop floor. For many, reorganising store layout and installing additional equipment is not a realistic option - either because there is not the available space, or because of budgetary constraints.

In these circumstances, an effective solution could be artificial intelligence (AI) and deep learning. Together, these cutting-edge technologies create a video surveillance offering that can give retailers valuable insight into real-time operating conditions, to boost customer service and in-store competitive advantage.

Enhancing retail experience

Intelligent video technologies can be applied to retail in a multitude of ways. This includes:

• People-counting, to track customer flow and numbers.
• VIP recognition, so retailers can tailor service to key individuals.
• Heat mapping, to reveal the popularity of specific goods in the store.
• Queue detection, to monitor the queuing situation in real-time.
• Point of Sale (POS) integration, to retrieve precisely when dispute occurs.

People counting technology

Evaluate the conversion rate, which depends on the number of purchases divided by the number of people in the store

Understanding the flow of customers is the key to optimising a store’s sales performance. In addition, by using an accurate people-counting camera, stores can understand daily/weekly/monthly peak times, so that store managers can arrange staffing accordingly.

What’s more, by layering this data over days, weeks and months, retailers can start to see customer flow trends. The data can be benchmarked against the store’s sales figures and its overheads, to inform future operational strategy, and help managers to optimise store profitability. Furthermore, people counting can also help retail managers gain insights into the customer experience.

One example would be to evaluate the conversion rate, which depends on the actual number of purchases divided by the number of people entering the store. Retailers can calculate this rate easily using people counting technology. Even better, people counting can also help stores determine the results of their marketing activities. Each marketing campaign can be quantified by people counting to measure return on investment. This is just one of several technologies that provide insight into consumer behaviour and optimal marketing practices.

VIP recognition

Loyal high-end or VIP customers are enormously valuable to retailers, but service staff cannot always identify them immediately. This means that business operators miss valuable opportunities to extend a special service.

Intelligent facial recognition cameras, however, can provide unique opportunities that were previously unavailable. When customers opt to participate in VIP marketing promotions or other incentives, facial recognition technology will give retailers accurate tools to identify VIPs so staff can provide the right kind of service at the right time.

There would be no more need for customers to show membership cards to receive special offers, for example. The whole shopping experience becomes more convenient and efficient.

Heat mapping

A single, compact camera can capture a panoramic high-definition image over an impressively wide area

Retailers usually want to know which products customers are most interested in, and which promotional products can attract customers to stop. With heat-mapping technology, retailers can see a colour-coded map of how much time shoppers spend in specific store areas. This can reveal hot spots and dead zones, as well as how many people actually shop for a specific product, versus those who just decide to pick it up as they casually walk by.

By layering this data over time, retailers can get insight into how to best place items in store, and how best to design the overall store layout. One of the most useful heat-mapping tools available to the retail market is the fisheye camera.

A single, compact camera can capture a panoramic high-definition image over an impressively wide area, while its heat-mapping function provides the colour-coded map. Because fewer cameras are needed than with most heat-mapping solutions, installation is cost-effective, making this the ideal solution for a location such as an independent shopping mall.

Queue detection

Waiting time can have a devastating effect on the in-store customer experience. If a customer sees a long queue in a store, they may stop shopping and go elsewhere. Worse still, if there is always a long queue every time they visit the store, they may avoid that store completely in future.

Queue detection solutions can help retailers manage their checkout lines, and the queue detection cameras are designed to monitor queues in real-time. The cameras count the number of people in each queue, and if there are too many people in each checkout line and customers still continue to flock to the checkout area it triggers an alarm to store management, who can open a new checkout.

Point of Sale (POS) integration

The centre management team installed Hikvision Dual-lens DeepinView People Counting camerasTransaction records and surveillance videos can provide valuable information to reduce disputes at checkout. In the past, these are two completely independent data systems. In order to find the right video footage, the surveillance video must be retrieved manually. Therefore, collecting evidence around a transaction normally can take several hours, which is costly for the business.

Point of Sale (POS) integration, also known as POS Text Overlay, is uploading of transaction data to the video management system through the POS and encoding it with the video stream. This combination of retail transaction data with surveillance videos makes it easy to identify certain transactions to be reviewed for customer complaints, mainly used for post-retrieval when a transaction disputes occurs.

Smart retail solution for Dutch shopping centres

People-counting is very important to shopping centres. Retailers always want to know what footfall they are likely to expect, and larger chains use this information to decide whether or not to move into the shopping centre at all.

At the Schalwijk Shopping Centre in Haarlem, the Netherlands, centre management was counting footfall by using the number of connections to the centre’s Wi-Fi. However, this was delivering inaccurate and limited data. People were still using the centre Wi-Fi from outside the mall, while many shoppers inside the mall were not using it at all.

To address this, the shopping centre decided to work with Hikvision, the world’s supplier of innovative video surveillance products and solutions. The centre management team installed Hikvision Dual-lens DeepinView People Counting cameras (iDS-2CD6820F/C) at all the entrances to the shopping centre. The cameras feature deep learning technology, binocular stereo vision, 3D people detection and height filtering - all contributing to their amazing 97% accuracy rate.

Transforming in-store experience

As a result, the Schalwijk Shopping Centre now has highly accurate visitor numbers and flow information, which can maximise both customer experience and layout effectiveness.

The intelligent, self-learning video solutions outlined above are designed to empower retailers with much of the insight they need to bring brick-and-mortar stores right up to date. Those that are willing to adopt these solutions will very rapidly see their value, and be able to reap the rewards of increased competitive edge.

Through accurate people-counting, quick identification of VIPs, optimal layout of goods and reduced waiting times, retailers can transform the shopping experience and enhance their sales strategy very quickly indeed. Moreover, thanks to the valuable customer insight that is generated on an ongoing basis, they will have the tools to create an in-store shopping experience that should only get better and better.