Facial recognition is technology that recognizes human faces and matches them to images of faces stored in a database. This technology has been around for years—in your smartphone camera, for example—but commercial establishments are also beginning to use it for retail applications. Brands can use facial recognition technology to improve customer service, increase operations efficiency for store staff, and bolster security for both businesses and consumers.
This article will explore facial recognition use cases and examples for retail, benefits of the technology, how different types of facial recognition software work, and alternatives.
How Does Facial Recognition Work?
Facial recognition technology works by first identifying a human face within its field of view and isolating that face from the surrounding environmental features or background. You’ve probably already seen this in action with digital cameras that recognize faces and draw boxes around them.
Advanced facial recognition software takes this a step further by localizing specific facial features and measuring factors like size, distance between features, and shape. The system then uses these measurements to narrow down the identity of the face based on information stored in its database. Eventually, the system pinpoints the exact identity of the person and reports this information to the user.
Types of Facial Recognition Technology
A few different types of facial recognition technology exist, with varying purposes, methods, and levels of reliability:
- 2D: Works by simply examining a two-dimensional picture of a person’s face (such as a photograph or digital image) and comparing this picture to other images of faces stored in its database.
- 3D: Offers more precision and requires a more elaborate setup; can detect and measure information about the geometry of a face and correct for variations in lighting, expression, and orientation of the face.
- Thermal: Detects temperature variations in a person’s face, which forms a digital image of the shape of their head, allowing the software to localize and measure various facial features, while ignoring irrelevant or distracting information such as clothing and makeup.
- Emotion-based: Aims to correctly recognize how a person is feeling based on their facial expression, with potential applications in marketing and advertising; personalization of products and services; security and crime prevention; optimization of education and training; and healthcare.
- Artificial Intelligence: AI software can continuously gather data, learn from previous experiences and errors, and collect and analyze various types of information from digital images. This technology is far from perfect (and also raises questions of legality and ethicality), it has the potential to gradually become a highly efficient and accurate form of facial recognition.
Facial Recognition in Retail: Use Cases & Examples
Facial recognition technology has numerous applications for retail stores, from providing a smoother experience for both customers and staff to security measures. Below are some of the most exciting potential uses for this technology in retail.
1. Customer Identification, Personalization & Analytics
Once installed, facial recognition in retail stores can collect various types of customer data, such as demographics, foot traffic, frequency of visits, and even customer moods.
With enough data collected, the system can then begin to do things such as make personalized recommendations for familiar customers on screens set up throughout the store. The technology would also make it easier to record and leverage customer preferences and purchase history. Facial recognition technology could also identify a returning customer in real time as they enter the store, based on a previously stored image of the customer’s face in the software database. The software could alert staff, who could then implement a personalized greeting and experience for that customer.
On the back end, facial recognition can also assist with customer analytics. Customer analytics is essentially the process of making business decisions based on predictions about customer behavior, as well as how buyers can be categorized into different groups that require different marketing strategies. Facial recognition can help with the research necessary for customer analytics, as it makes the process of identifying and sorting regular customers faster.
In action: Restaurant chain Pizza Hut launched a campaign in India for pizza recommendations based on facial cues and expressions, matching their offerings to customers’ moods.
2. Loss Prevention & Security
Better loss prevention and improved security are perhaps the most common current applications of facial recognition for retail. These require only basic facial recognition software and some security cameras; thus, the benefits are accessible even to small and medium-sized businesses that aren’t in a position to use more advanced tools like artificial intelligence (AI).
For preventing loss (whether this loss occurs via intentional malicious action or otherwise), facial recognition technology could help not only with detecting unauthorized access or actions by employees but also with identifying the individuals involved.
As for store security, a retail brand could build up a database of known shoplifters or troublemakers, and then configure their stores’ facial recognition software to alert staff or security team members if one of these individuals is detected entering a store.
In action:
- A retail company (that wished to remain anonymous) was able to stop 90% of repeat offenders at one of their stores due to facial recognition technology.
- In 2020, drugstore chain Rite Aid implemented facial recognition technology in 200 of its branches across the US to deter theft and improve safety for customers and store staff. The idea was for the technology to identify whether people entering the stores had a history of engaging in criminal activity. If the system determined that this was the case, it would send an alert to security agents. However, the implementation of the technology was controversial due to data privacy issues.
3. Biometric Payments
Retail stores can use facial recognition technology to increase convenience for customers when they pay for their purchases at the checkout counter. Customers will first allow the software to capture an image or scan of their face, as well as their credit card or other payment details. On subsequent visits, the buyer need only show their face to a camera and allow the software to verify their identity, at which point the purchase will be authenticated and their card will be charged.
In action:
- In 2022, Mastercard launched a program that enabled retailers in Brazil to let customers pay with biometric methods, including facial recognition.
- In 2020, restaurant CaliBurger allowed guests to register for a payment option based on facial recognition. The vice president of Cali Group predicted that the technology could eventually be used for loyalty programs as well as payments.
Pros & Cons of Facial Recognition in Retail
Pros | Cons |
---|---|
Improved security | High price |
Increased sales | Can be difficult to acquire or access |
Enhanced customer satisfaction & loyalty | Privacy, reliability, trust, and ethical concerns |
Facial recognition can do a lot for retail brands, but the technology is not without controversy. Here are some pros and cons of using facial recognition technology:
Pros
- Improved security: Combined with other security tools like cameras and databases, facial recognition software can improve security by detecting known shoplifters and troublemakers who attempt to reenter a store. The system would then alert employees or security staff.
- Increased sales: Once the software can identify the face of a regular customer, it can combine this with other customer data such as purchase history, preferences, traffic patterns, and spending habits. These data points give the business a basis for making key decisions with regard to inventory, marketing strategies, and more. Retail brands can enhance the potential benefits by combining facial recognition technology with point-of-sale (POS) data analysis.
- Enhanced customer satisfaction & loyalty: Facial recognition technology can detect when the same customer visits a store repeatedly. With this in mind, store employees can create a personalized experience that encourages that particular customer to buy more. Once the customer returns to the store and the technology recognizes them, it can notify store staff who will implement the personalized experience. Over time, the store can craft personalized experiences for all of its regular customers.
Cons
- High price point: Even the most basic facial recognition technology that would be useful in a retail store can cost thousands of dollars—up to tens or hundreds of thousands—for more advanced systems or in larger stores. This price range will likely put advanced facial recognition technology out of reach for most small and medium-sized businesses—at least for now.
- Difficult to acquire or access: As the technology is not yet ubiquitous, companies that insist on implementing facial recognition for their retail stores may need to do a lot of shopping around to find exactly what they want. Different vendors and developers will also quote different prices, so it may be a while before a business can find the best deal.
- Privacy, reliability, trust, and ethical concerns: Technology that can learn, memorize, and identify people’s faces will naturally create a great deal of controversy in terms of customer privacy and trust, as well as the general ethicality of the technology. You cannot count on people wanting or being willing to share biometric data—all the more so when they might not be aware that this data is being collected. Businesses need to be clear about how they are collecting and using this data, the general security practices of the store, and customers’ rights. In the worst-case scenarios, customers may even accuse a business of racial prejudice or other forms of discrimination.
Frequently Asked Questions (FAQs)
Click through the following sections to learn more about facial recognition in retail stores.
It’s difficult to say. Some brands may adopt the technology in secret, others may begin using it and then stop later on, and still others have made commitments not to use it. As of 2021, Ace Hardware, Apple, Lowe’s, and Macy’s were reported to have been among the retail stores using facial recognition technology.
In theory, facial recognition can improve security by storing information about the identity of known shoplifters and criminals, and detecting their presence should they attempt to enter the store again in the future. However, the use of facial recognition technology is a complex issue with considerations of privacy, ethicality, and discrimination.
For using facial recognition as a payment method, you will probably be able to refuse and stick with standard methods such as regular credit card payments. If a retail brand decides to use facial recognition technology for purposes like measuring foot traffic, applying customer analytics, and bolstering security, you might not have a say in the matter. In the case of hidden cameras, you might not even realize that your face is being scanned.
Bottom Line
Facial recognition technology is currently far from being widespread, but it’s gaining traction as retail brands see the potential for improved customer service, security, and loss prevention. Even if this technology is something you’re not interested in, or is currently out of reach, knowing what it can do and how large retail businesses leverage it will give you a leg up in finding alternatives and keeping your small business competitive.