- Retailers such as Starbucks and Circle K are using Automatic License Plate Recognition (ALPR) to enhance drive-thru, curbside pickup and service experiences for customers.
- Drive-thru and curbside have also grown significantly during the pandemic and retailers are now looking for ways to differentiate these experiences as the use of these channels are expected to remain above pre-pandemic levels.
- Being able to recognize a customer’s vehicle offers retailers the opportunity to improve customer service, same store sales and improve employee productivity.
- Automatic License Plate Recognition (ALPR), a computer vision technology, can differentiate the customer experience alongside complementary technologies including mobile and RFID. License plates, with consumer opt-in, are matched to retailer customer and order databases, and are used to enhance the store experience. The cost of ALPR has reduced significantly and opt-in approaches meet privacy concerns.
- Drive-thru (in quick service & an expanding set of retail categories) can be enhanced with pay-by-plate, menu board personalization, and operations automation.
- Curbside pickup times, which consumers rate as a key purchase criteria, can be reduced with ALPR, and additional curbside capacity created.
- Service experiences can be personalized (e.g., recognize a car as it pulls into a service lane at an auto dealership) improving customer loyalty and staff productivity.
- In the next article we’ll cover deployment of ALPR with open source Kubernetes in a store environment.
We’ll provide a quick overview of Automatic License Plate Recognition (ALPR), also known as Automatic Number Plate Recognition (ANPR). We’ll then cover how retailers can use ALPR to enhance curbside pickup, drive thru and service experiences for retailer customers. In Part 2 we provide a technical tutorial on how to deploy ALPR in a quick service restaurant environment using Platform9 Managed Kubernetes as part of a software-driven store strategy.
What is Automatic License Plate Recognition (ALPR)?
ALPR is a branch of computer vision technology that uses optical recognition on images to read vehicle registration plates. It first reached broad adoption in transportation and law enforcement. The cost to deploy ALPR has dropped dramatically through low cost IP camera systems (cameras may be used to support multiple use cases including security) and open source ALPR software and there has been a growing set of commercial applications in retail, education, construction and other industries.
ALPR uses a series of image manipulation techniques to detect, normalize and enhance the image on a license plate. Optical character recognition (OCR) is then used to extract the numbers and letters on the license plate. ALPR systems are generally deployed in one of two basic approaches: on-site so that the entire process can be performed real-time, or by sending images to an external datacenter and performing the OCR process there with a bandwidth and latency cost. When done on-site, the information capture of the license plate can typically be completed in under 250 milliseconds. This information can be readily sent to a remote datacenter for further processing if necessary, or stored on-site for later use.
Retail and Restaurant Use Cases for ALPR
Key use cases for retail and restaurants include drive-thru, curbside pickup, and personalized service experiences:
Drive-thru sales are up by an estimated 25-35% during the pandemic, with Americans visiting drive-thrus 6 billion times a year. The pandemic accelerated retailers’ moves to provide more drive-thru options. Originally used by fast food chains, the drive thru model is being embraced by other retailers in part because it provides a socially distanced, contactless experience for the customer. Quick service restaurants and coffee chains such as Starbucks, Chipotle and Shake Shack have announced plans to increase the number of locations with drive-thrus. Convenience store Wawa is even launching its first drive-thru only stores. Pharmacy Walgreens also announced drive-thru options at 7,300 locations during the pandemic. According to the Wall Street Journal, banks are reopening unused drive-thru lanes and installing new ones.
Drive-Thru Now Generates 50% of Global Starbucks Sales, ALPR-supported service reaches 40% Adoption in Company’s 2nd Largest Market
Starbucks Korea (the company’s 2nd largest market by store count, and a technologically advanced country with the world’s highest smartphone penetration of 97%) rolled out its My DT Pass service in 2018 which links the car’s license plate to the coffee chain’s pre-paid cards, allowing for automatic payment of coffee purchases. When a pre-registered car enters the drive-through area, the system identifies the license plate automatically, while allowing the Starbucks barista to check the customer’s preferred name, order details, and coupon usage. Starbucks reported that tests test showed the My DT Pass system provided a reduction of 13 to 15 seconds per car in a drive-through line, shortening the total wait time by 10 percent. Adoption has grown and in 2020 the company reported that 40 percent of drive-thru customers were using the My DT Pass service; drive-thru use itself had also grown 32 percent. The company also reported 1.5M DT Pass users as of October 2020, 50% up from 1M 8 months prior, and representing about 6,000 registered users for each of its 268 drive-thru stores.
Enhancing Drive-Thru with ALPR
Multiple use cases are available to drive thru operators with particular benefits for wait times, a critical element of the quick service value proposition. “Every 10 second delay in the drive thru equates to a 1% loss in comp sales,” Sue Pittacora, former senior director of global business insights and analytics at McDonald’s told Nation’s Restaurant News. Turnaround time, customer order value, queue abandonment, retention, food costs and restaurant labor metrics can potentially be improved as ALPR supports three use cases:
- Menu boards personalization: license plate results, after customer opt-in, are used to generate recommended items on the digital menu board. These menus can accelerate the order process (e.g., show a usual order) or be used to increase ticket values through recommendations. Starbucks announced plans in 2019 to offer a customer opt-in system to personalize drive-thru menu recommendations based on order history and 400 other variables including weather and time of day. McDonalds introduced personalization and Forbes notes that “recommendation algorithms have already demonstrated an unspecified increase in order size and the chain plans to roll them out to all its drive-thrus in the U.S. by the end of the year”.
- Pay by plate: the payment process is automated by presenting the customer their usual payment method without needing to hand a credit card too an employee, providing a socially distanced and faster experience for the customer.
- Operations automation: data on the individual customers in the restaurant queue can also be used to send proactive signals to internal food preparation systems to be ready for preparation of menu items highly likely to be ordered, helping turnaround time and potentially reducing ingredient waste.
Tech startup 5Thru estimated that order time reductions from drive-thru technology could help restaurants process around an extra 30 cars a day. With the average quick service restaurant check size of $10.99 in 2020, this would represent a potential increase of $132,000 in sale store per year, a potential 5-10% increase given unit sales of $1.4-4.5M/year for top QSR chains.
US fast food retailers rolling out include burger chain White Castle, which has drive-thrus at 95% of its 362 restaurants. The company will use ALPR to personalize orders at the drive-thru, for customers who have opted in to through White Castle’s Craver Nation app. The software connects to their order history and the company’s new AI voice system and connected to a digital menu board with recommendations. “Bringing technological empowerment to the drive thru is the next chapter in innovation for our industry,” said Susan Carroll-Boser, vice president of technology at White Castle when the move was announced.
McDonalds is also testing ALPR to accelerate drive-thru experience, seeking to reduce wait times further form the nearly 10% reduction it achieved last year; the average time reduced from six minutes and 18 seconds in 2019 to five minutes and 49 seconds in 2020, according to market researcher SeeLevel HX.
Retailers have rapidly adopted curbside pickup services with 50.7% of the top 1000 retailers offered it in early 2021 according to a recent survey by Digital Commerce 360, up dramatically from only 6.1% in early 2020. Consumers have embraced it, with curbside pickup adoption has reached 58% of US consumers by the end of 2020, with over half using it for the first time that year. It is viewed favorably by consumers with 59% of people rating it as more convenient than in-store shopping and 69% people rating it as safer than in-store shopping. The four most common were groceries (85% of curbside pickup users), beauty & personal care (33%), pet supplies (32%) and home improvement (24%). 71% of users saying that the reduced social interaction was a benefit.
European Retailer Kingfisher See 226% Growth in Curbside Pickup
In home improvement, retailer Kingfisher has seen curbside as the fastest growing channel during Fy20/21 in which the group grew 6.8%. Kingfisher CEO Thierry Garnier commented that “Click & Collect [Kingfisher’s curbside pickup brand] sales have become the largest and fastest-growing fulfillment channel at group level, with 226% growth. Supported by our newly implemented group digital stack, our platform has scaled rapidly and is now supporting 500,000 Click & Collect orders per week”.
Image: Curbside Pickup at B&Q, one of European Retailer Kingfisher plc’s UK home improvement brands.
Target reported curbside sales growth of 734% in Q2 2020. Ulta Beauty has seen benefits with digital sales doubling in fiscal year 2020 when in-store sales were challenged; many stores still remain at only 50-75% of capacity. For grocery at least, curbside pickup is now considered table stakes with online sales projected to grow from 5.1% at the end of 2019 to 15% by 2030.
Image: Target app showing curbside pickup experience. The user chooses vehicle type and color, to assist vehicle recognition.
Enhancing Curbside Pickup with ALPR
The priority for retailers during the pandemic has been to set up a workable system and cope with the sudden order growth, and most retailers used mobile-enabled systems. Looking forward the challenge for retailers is to differentiate their curbside experiences from other retailers, with speed of pickup being a critical driver. Shoppers who waited under two minutes for their pickup order from retailers were four times more likely to make a future purchase, a study by Rakuten Ready found. The study also found that pickup times varied from 2 minutes to 9 minutes, suggesting opportunity for retailers to stand out.
Deploying ALPR can help streamline curbside pickup processes. Customers enter their license plates with their order. On arrival at the store retailers’ cameras scan the plate and notify the retailer. Retailers can welcome people with digital signage or customized voice messages as they arrive, letting them know their order will be brought to their car shortly. And if the order has not been found, it will alert the customer to contact a store employee. A license plate based approach can also overcome challenges with mobile-based approaches when customers can also forget or incorrectly use mobile check-in methods or park in incorrect parking spots. Optimally the retailer is also using geofencing/location-sharing to know that customers are heading for the store before their arrival in the carpark, providing further lead time for order preparation.
Auto Service Experiences using ALPR
In service environments (e.g., gas station, auto dealership service lane, carwash) license plate recognition can be used to improve customer loyalty, productivity and lifetime value. Key use cases include
- personalized greeting. Drivers arriving at the service lane in their cars are greeted with a personalized welcome on a video screen
- pay-by-plate. Frequent carwash users can pay automatically without needing to provide credit cards, similar to the quick service use case.
- employee productivity. Service managers are alerted that the customers have arrived along with relevant information about the customer, and can respond faster to customer needs.
- upsell opportunities. Cars going through the carwash can trigger alerts to dealer personnel about vehicle recalls or other opportunities to approach the customer.
Circle-K’s parent company announced in April 2021 that it is rolling out an ALPR-based solution at gas stations in multiple countries after successful trials, noting it would be first in market, and that the combination of ALPR and a mobile app would allow an easier fueling experience and enhance its forecourt offering. Customers no longer need to provide a card or PIN for each transaction. Customers now fill up with fuel and, through number plate recognition, pay for the fuel on the Circle K Easy Fuel app.
Image: Circle-K pay-by-plate.
Image: Personalized Video Screen above Vehicle in Service Lane at an Audi dealership – system by MDL Automation offered with ALPR & RFID.
Image: ALPR in use at car wash, with touchscreen allowing upsell to subscription plans without need for RFID device – system by Sonny’s The Carwash Factory.
Connecting ALPR with the Retail Customer Experience
Being able to recognize a customer’s vehicle offers retailers the opportunity to improve customer service, same store sales and improve employee productivity in drive-thru, curbside pickup, and service experience use cases. Developing customer journey maps that identify pain points and opportunities to improve the customer experience provide an ideal lens with which to consider how ALPR can be used to create retail service innovation. Below is an example completed by a Carnegie Mellon team showing how customer experience at Sheetz, a convenience store with 628 outlets known for made to order food, could be enhanced through a variety of technologies including ALPR.
What’s Next – Implementing Retail ALPR
In Part 2 we will cover deployment of a containerized ALPR application with Kubernetes using helm charts and show a drive-thru license plate recognition example.
- How To Set Up ALPR (Automatic License Plate Recognition) with Kubernetes to Improve Retail Drive-Thru Customer Experience (Part 2) - July 26, 2021
- Improving Retail Customer Experience with Automatic License Plate Recognition (ALPR) – Part 1 - July 26, 2021
- How to Set Up and Run Kafka on Kubernetes - September 16, 2020