Department stores differ from their smaller retail counterparts in how they approach people-counting and conversion measurements. While the overall proposition is the same — to see how many people came in and how much they spent — they operate on a much larger scale. With multiple floors and diverse retail categories within a single location, department stores aim to go beyond basic key performance indicators (KPIs) to gain deeper insights into how their space is used. While traditional traffic counting technology can provide some of this information, Re-ID offers a comprehensive solution — and much more.

Re-ID, or Re-Identification technology, uses sophisticated artificial intelligence (AI) and deep learning algorithms to identify anonymous individual shoppers by analyzing their unique characteristics, such as clothing patterns, colors, accessories, and logos as they move throughout a retail space. This allows an individual to be recognized across multiple sensors, without using facial recognition or personally identifiable information (PII) of any kind.

The first step in any people-counting solution is being able to count perimeter traffic, or the number of people coming through each entrance. The step beyond that, however, is understanding where customers are going once they’re in the store. Are they heading to apparel, housewares, or home appliances? Which floor do they go to, and how many floors do they visit? How long are they spending in each of these areas?

Traditional people-counting counting technology can partially achieve these insights. Cameras placed at the entrances of each floor can track the number of visitors to specific areas, while Wi-Fi technology can complement this by providing data on dwell times. However, this approach has its limitations. If a person visits a floor more than once, they will be counted more than once. Wi-Fi can provide an estimate of dwell times but requires multiple tags to cover large areas and still only offers sample-level data.

Re-ID technology goes beyond traditional people-counting technology by identifying and recognizing unique visitors across multiple devices. This ensures that if a visitor temporarily disappears from view — such as entering a fitting room — they won’t be re-counted or identified as a new visitor when they reappear.

With Re-ID technology, a shopper could visit multiple floors multiple times and be counted only once. It can also show which areas of a floor people visit without the need for excessive devices, making it especially beneficial for stores-within-a-store, leased spaces, restaurants, and cafes. This data can highlight the value each area brings to the department store and reveal the percentage of overall shoppers who engage with them. Additionally, since Re-ID can track when a visitor enters and leaves, it can provide census level dwell time at zone, floor, and store level.

One of the most powerful benefits of Re-ID technology is being able to understand the customer journey. By monitoring visitors throughout a department store, Re-ID can see where customers go and which paths they take. For example, Re-ID can track the percentage of shoppers who first visited homewares, then move to shoes, and finally to beauty. It can identify whether specific zones, brands, or concessions are driving significant traffic, highlight the most popular categories among customers, and reveal cross-visitation patterns between different categories.

Re-ID can also help department stores identify whether visitors are just passers-through rather than genuine shoppers. This is particularly relevant for stores attached to large parking lots, where some individuals may simply pass through the store on their way out to the lot. Since Re-ID can identify unique visitors and track their in-store dwell time, visit duration parameters can be established. Visitors whose dwell time falls below a set threshold can be classified as passers-through, providing the department store with a more accurate representation of their actual traffic number.

Given the size of a department store compared to a typical retail store, it’s no surprise that there will be significantly higher numbers of employees, who may work across multiple floors. Having these additional people in-store, however, can impact overall visitor numbers — and therefore artificially decrease conversion.

Re-ID addresses this issue by excluding staff without the need for special wearables, such as badges or lanyards, unlike other staff exclusion technologies. Re-ID’s artificial intelligence excludes employees based on factors such as the duration of their time in the store (as shoppers are unlikely to remain in a store from open to close) and the appearance of their uniforms. This negates any issues with staff compliance and wearables not aligning with the store’s brand guidelines.

Re-ID can be used to measure the success of activations and in-store events. If a particular area of the store is hosting a promotion or event — such as a “Breakfast with Santa”, a sale, or a new pop-up — Re-ID can analyze whether there were significant changes in customer movement patterns and dwell times. It can help answer critical questions like: Did people spend more time in this area of the store? Did shoppers only visit the event space, or did they visit other areas of the store? Re-ID helps evaluate the impact of these activities by providing a clear understanding of shopper behavior.

Re-ID technology has revolutionized people counting, offering a more accurate way to measure visitor numbers by excluding passers-through and employees. It also provides valuable insights into where shoppers are spending their time, helping stores better understand foot traffic patterns and shopper journeys. By optimizing how department stores track and analyze movement throughout their spaces, Re-ID delivers actionable insights that drive smarter strategies and improve overall performance.

Excited to hear more about how Re-ID can deliver dramatic improvements in traffic accuracy, customer-journey analytics, and more? Contact Sensormatic today to start the conversation.