NashTech Insights

Let AI get your Shrinkage under control

Vikas Hazrati
Vikas Hazrati
Table of Contents
artificial intelligence, network, web-5866644.jpg

According to the National Retail Federation, shrinkage cost retailers over $61.7 billion in 2019 alone.
Retail Shrinkage

What is Shrinkage?

Simply put, shrinkage is the loss of inventory thus causing retailers to lose money. Yes, it is scary when your hard-earned money walks out of the store without ringing the register. Some of the common types of shrinkage are

1. Shoplifting or theft

This is by large the biggest contributor. Shoplifters could steal any product and storm out of the store without paying. As per the data available from 2023, this has grown by 13% with people coming back to stores after COVID. Organized retail crime (ORC) is another huge factor in this segment.

2. Sticker SWAP/Return fraud

With many stores allowing self-checkout, it is easy for a person with dubious intentions to paste the bar code of an inexpensive product on that of an expensive product and walk out of the checkout lanes.

Return fraud on the other hand takes on many different forms including returning stolen merchandise or returning merchandise that has been used or purchased with counterfeit money or receipts. 

3. Administrative error

Human mistakes can happen all the time. These could range from mislabelling merchandise for a lower price or refunding merchandise for more than it is worth.

4. Employee theft 

Employee theft is another significant contributor. This includes stealing merchandise, ringing up fake returns, improperly using an employee discount and the more frivolous ones who might end up stealing cash from the POS.

The biggest concern when it comes to retail shrinkage is that the loss of inventory cannot be recovered. This directly impacts the retailer’s bottom line.

Overall, retail shrink is trending sharply upward. The results of a 2020 NRF survey show that losses from theft, fraud and other retail shrink factors rose nearly 22% from 2018 to 2019. Interestingly, the NRF also found that ORC did not slow down during the pandemic. In fact, it cost retailers an average of $719,548 per $1 billion in sales in 2020 — a slight increase over 2019.

Retail by numbers

How can AI help?

Fortunately, modern technology and the infusion of precision retail can put new tools in the hands of retailers, allowing them to leverage diverse data that has been captured over years to predict and prevent retail crime. This data is not limited to videos from CCTV but also customer data which can be linked back through their loyalty cards, payment methods etc.

Some of the tooling that can help,

Video monitoring and machine learning

Many existing surveillance cameras and sensors can be equipped with analytics technology capable of serving a wide range of purposes without necessarily complicating shoppers’ experiences. Using AI-powered video surveillance to monitor stores 24/7 and detect suspicious behaviour in real time is a strong possibility with modern technology. AI can analyze customer behaviour and alert store personnel when it detects something unusual. AI would understand the distinction between usual and unusual by learning through patterns of the past. The system can also provide footage of incidents for evidence and investigation.

For example, a retailer might geo-map ORC patterns to identify frequently targeted stores, departments or products. If they can predict shoplifting patterns by analyzing the data, retailers will be able to train and deploy security personnel more effectively across all locations.

Turn Data and Insights into action

Strategically placed AI at checkouts can capture the weight and size of an item and detect whether a lower-priced item is scanned instead. Retailers can also use enhanced video surveillance cameras, which combine AI with data analytics to monitor the behaviour and in-store movements.

The video monitoring and analysis of the videos would help in employee monitoring as well by detecting anomalies in employee behaviour, such as accessing unauthorized areas, unusual transactions, and unusual working hours. By analyzing employee behaviour, AI can help retailers identify and prevent internal theft.

Predictive and Preventive Analytics

By identifying retail crime patterns, adding new security layers to known problem areas like self-checkouts and monitoring for known signs of retail crime, retailers can stay one step ahead of shoplifters.

Since many stores have loyalty cards and can track returning customers through the use of their phones or credit cards, it can detect inventory discrepancies, unusual returns, and suspicious transactions and flag scrupulous customers who need to be kept an eye on. The sensors can track these customers and warn the store.

Other solutions

Then there are other solutions that can help like

  • Cash automation technology. Using smart safes and cash recyclers with individualized employee PIN access, retailers can monitor the movement of cash in near real-time.
  • RFID systems. With embedded sensors, radio frequency identification tags can alert retailers to theft. RFID-enabled smart tags attached to expensive products communicate with an electronic reader to track products. These devices can be removed at the checkout; if they’re not removed, a security alarm is triggered when the customer tries to exit the store.
    • At self-checkout counters, the RFID tags would trigger an assistant call alert to help with the checkout through which the store would be able to counter sticker swap as well.
  • In-store heat maps. Retailers are testing new heat map systems powered by thermography, using thermal technology to map emotions onto the body with colours. For instance, red indicates anger or anxiety. Retailers can also see the most heavily trafficked areas of the store to count customers, identify popular products and track patterns.

How can NashTech help?

NashTech has been working with retailers to make a significant dent in their shrinkage through the use of technology. Our Edge AI capabilities coupled with video monitoring and building machine learning models on already existing data have helped major retailers lower their shrinkage by 25%.

From shelf space management to in-store health monitoring, from image optimization to analyzing consumer behaviour AI can improve consumers’ shopping experience and revenues. The flexibility and efficiency of our technology solutions deliver multiple impactful solutions to the retail industry on one platform. We would love to connect and work together to make your business successful.

Vikas Hazrati

Vikas Hazrati

Group CTO @NashTech

1 thought on “Let AI get your Shrinkage under control”

  1. Pingback: AI as a service: The Next Big Thing

Leave a Comment

Your email address will not be published. Required fields are marked *

Suggested Article

%d bloggers like this: