Out of Stock No More: Ensuring Effective Stocking for Retail

A study conducted in 2020 on North American retail highlighted that out-of-stock (OOS) products on the store shelf accumulated a loss of $71.4 billion ^[1] for retailers, with poor inventory management, unpredictable seasonal demand, demand from omnichannel online shopping services, and lack of communication from marketing functions being some of the factors contributing to a shortfall in revenue ^[2], while also causing customer dissatisfaction and churn. According to research, poor customer experiences have resulted in approximately $62 billion ^[3] in abandoned and lost sales every year in the United States, with out of stocks exacerbating this challenge further. From bath tissue and pet food to detergent and soap, the retail industry saw out of stocks occurring across multiple product lines, with supply chain disruptions, excessive demand, and shortages prevalent across the globe ^[4].

As OOS products continue to pose a significant challenge for retailers, let’s take a look at a few measures that can help businesses manage their inventory better, to ensure that in-demand products do not stock out or oversell:

Automating inventory management through shelf scanning

Inventory verification, fulfilment, and restocking are now being increasingly automated in retail, with some stores experimenting with IoT and robotic systems to scan planograms and capture inventory positions, to enable better ordering and restocking. For example, a grocery store chain used mobile robots to help with inventory scanning and real-time data, providing images of products available on store shelves and notifications on products that were out of stock ^[5]. This shelf-scanning robot helped reduced out-of-stock products by 20 to 30%, being able to detect OOS inventory 14 times better than manual scanning and inventory tracking. Such robots, which are powered by algorithms that use sensors and system requirements to select and move orders, can help streamline inventory management significantly, reducing the risk of inadequate stocking and customer dissatisfaction.

Analyzing market trends with data mining

The latest market trends, innovations in personalization, and seasonal changes, among others, can be tracked, analyzed, and explored through data mining techniques. With the use of advanced algorithms, retailers can plan inventory based on the trends in the market. For instance, data mining can help detect patterns among trendsetters in a particular area and alert the retailer to stock up on the products endorsed by them. It can also help detect shopping patterns closer to holiday seasons and alert the retailer to stock up on popular products. For instance, with electronic goods being among the most in-demand and OOS product during the holiday season ^[6], data mining can be used to detect trends in buying patterns based on social media posts, product reviews, online searches for electronics, and more. Data-driven strategies therefore make it easier to anticipate future customer requirements and stock products accordingly ^[7], ensuring that retailers are future-ready.

Building predictive capabilities with ML

A worldwide survey conducted in 2020 among retailers revealed that merely 8% of retail stores implemented real-time automated processes while planning and forecasting their future supply and demand ^[8]. Real-time automated processes provide data that facilitates demand forecasting and prompt decision-making, enabling businesses to stay agile and responsive to changing market conditions. Retailers are now increasingly leaning towards incorporating ML-based demand forecasting to eliminate inefficiencies resulting from misalignments between supply and demand. In large data sets, ML-algorithms can spot patterns and detect demand fluctuation. For instance, 2020 saw an upsurge in the adoption of pets early into the pandemic ^[9]. ML algorithms can be leveraged to detect increases in pet adoption in advance and predict the increased demand for pet food and supplies. Reporting and analytics tools can further be used to determine the re-order point, current order quantity, and stock in hand. Using these technologies, omnichannel retailers can integrate data from all their sales channels and create a monthly recurring process that analyzes historical forecasts, comparing them with current market demands to determine the efficacy of forecasting and stocking.

Streamlining inventory management with Robotic Process Automation (RPA)

A combination of IoT (Internet of Things) and robotic applications have recently enabled retailers to acquire better visibility to backroom inventory. No-touch store ordering and replenishing systems have been implemented by several businesses, requiring minimum or no intervention from store managers ^[10]. In this regard, robots equipped with sensor technology, detecting out-of-stock items in real-time by identifying differences in temperature ^[11], weight, and shelf display, are being experimented with. RPA can further streamline inventory management by replacing manual data logging and promoting efficiency, eliminating the possibility of stock miscalculation due to human error. In fact, 74% of retailers opine that using robots in their stores would increase inventory accuracy ^[12]. This has led to a steady rise in the robot automation market revenues, and it has been estimated to increase from $2.9 billion in 2020 to more than $10 billion by 2023 ^[13], with applications for retail ranging from store planning and shelf optimization to stock replenishment.

Tracking stock with bar codes and RFID

Bar code technology was used for the first time in 1974 to sell a pack of chewing gum in a retail store ^[14] and it has since gone on to replace the manual tracking of stock and streamline inventory management, improving speed and security through real-time information. 2D barcodes are currently being used by most businesses to keep tabs on inventory. For instance, managers can view up to four bar codes at a time and get information on serial numbers, part numbers, date, and lot. Combined with data-driven approaches, ubiquitous technology such as bar codes can allow improved monitoring of stock by store associates, driving better bottom-line results and reducing costs for retailers.

Radio Frequency Identification (RFID) is proving to be another effective way of increasing SKU visibility, making it possible for businesses to track every product in the store. By scanning RFID tags, real-time data about the quantity of the item in stock can be obtained. In fact, researchers at the University of Arkansas have found that RFID technology helped a multinational retail store reduce stock outs by 16% ^[15],  highlighting the impact of data-driven stock tracking. Over the years, RFID technology has evolved as well – in addition to becoming more efficient and portable, it can even be used to tag liquid and metals, enabling a wide range of applications for retail.

AI in inventory management has revolutionized stocking and storing for industries. Following the implementation of advanced AI&ML, several businesses have been able to automate inventory management effectively, with improvements in productivity and reductions in errors being the most immediate outcomes.

To make the most of these advanced technologies, retailers should focus on use of smart devices, computer vision, AI&ML, augmented reality, and data capturing software with multiscanning to regulate inventory and avoid stock outs.

To make the most of these advanced technologies, retailers can focus on

• Developing accurate demand forecasting solutions, which can adapt to evolving customer preferences and changing market scenarios
• Streamlining inventory management, to understand on-hand stock, both on shelf and for backroom inventory, by utilizing the power of IoT, robotic automation, and RFID technology.
• With advanced technologies being leveraged to strategize on stock out prevention measures and ensure optimal stocking in retail stores, higher sales, increased customer satisfaction, and business growth are now within reach for retailers.

Bibliography

1. Ozbun, T. “Challenges of Online Grocery Shopping: Out-of-Stock Items U.S. 2020.” Statista, January 27, 2022. https://www.statista.com/statistics/1127446/challenges-of-online-grocery-shopping-out-of-stock-items-us/.

2. Ozbun, T. “Challenges of Online Grocery Shopping: Out-of-Stock Items U.S. 2020.” Statista, January 27, 2022. https://www.statista.com/statistics/1127446/challenges-of-online-grocery-shopping-out-of-stock-items-us/.

3. Hyken, Shep. “Bad Customer Service Costs Businesses Billions of Dollars.” Forbes, August 29, 2016. https://www.forbes.com/sites/shephyken/2016/08/27/bad-customer-service-costs-businesses-billions-of-dollars/?sh=60c64b735152.

4. Russell Redman 1 | Mar 24. “Nielseniq: Out-of-Stocks during Pandemic Lead to Billions in Missed CPG Sales.” Supermarket News, March 24, 2021. https://www.supermarketnews.com/consumer-trends/nielseniq-out-stocks-during-pandemic-lead-billions-missed-cpg-sales.

5. Heater, Brian. “Simbe’s Robots Will Be Deployed across Midwestern Grocery Chain, Schnucks.” TechCrunch, August 26, 2021. https://techcrunch.com/2021/08/26/simbes-robots-will-be-deployed-across-midwestern-grocery-chain-schnucks/.

6. Case, Brendan. “Holiday Web Shoppers Face an Increase in Out-of-Stock Goods.” BQ Prime, November 9, 2021. https://www.bloombergquint.com/business/holiday-web-shoppers-thwarted-by-increase-in-out-of-stock-goods.

7. Markets, Research and. “Global Predictive Analytics Market (2020 to 2025) – Advent of Machine Learning and Artificial Intelligence Is Driving Growth.” PR Newswire: press release distribution, targeting, monitoring and marketing, November 3, 2020. https://www.prnewswire.com/news-releases/global-predictive-analytics-market-2020-to-2025—advent-of-machine-learning-and-artificial-intelligence-is-driving-growth-301165726.html.

8. Case, Brendan. “Holiday Web Shoppers Face an Increase in Out-of-Stock Goods.” BQ Prime, November 9, 2021. https://www.bloombergquint.com/business/holiday-web-shoppers-thwarted-by-increase-in-out-of-stock-goods.

9. Ho, Jeffery, Sabir Hussain, and Olivier Sparagano. “Did the COVID-19 Pandemic Spark a Public Interest in Pet Adoption?” Frontiers, April 9, 2021. https://www.frontiersin.org/articles/10.3389/fvets.2021.647308/full.

10. Hoberg, Kai, and Sandria Weisshuhn. “Designing Smart Replenishment Systems: Internet-of-Things Technology for Vendor-Managed Inventory at End Consumers.” European Journal of Operational Research, December 16, 2021. https://www.sciencedirect.com/science/article/pii/S037722172100268X. 10.

11. Torres-Sánchez, Roque, María Teresa Martínez-Zafra, Noelia Castillejo, Antonio Guillamón-Frutos, and Francisco Artés-Hernández. “Real-Time Monitoring System for Shelf Life Estimation of Fruit and Vegetables.” MDPI, March 27, 2020. https://www.mdpi.com/1424-8220/20/7/1860/htm.

12. Leonard, Matt. “A Majority of Retailers Struggle with Inventory Forecasting.” Supply Chain Dive, April 9, 2019. https://www.supplychaindive.com/news/retailers-struggle-inventory-forecasting/552164/.

13. Taylor, Petroc. “Global Robotic Process Automation Market 2023.” Statista, January 2020. https://www.statista.com/statistics/740440/worldwide-robotic-process-automation-market-size/.

14. Harford, Tim. “How the Barcode Changed Retailing and Manufacturing.” BBC News, January 23, 2017. https://www.bbc.com/news/business-38498700.

15. Roberti, Mark. “EPC Reduces Out-of-Stocks at Wal-Mart.” RFID JOURNAL, April 7, 2020. https://www.rfidjournal.com/epc-reduces-out-of-stocks-at-wal-mart.