No More Empty Shelves: How AI Turns Retail’s Whack-a-Mole Into Precision Profit
Play Audio Version
My first professional job after college was working for Siemens-Nixdorf, the German technology company that specialized in retail systems. My focus was the Point-of-Sales (POS) systems, often known as the check-out because it’s the system you see an operator using when you are paying for whatever you are buying.
Although the POS was my main design and coding focus (proprietary versions of Basic and Cobol), I also became involved in a secret project where the British supermarket chain Tesco was exploring how to replace all their shelf-edge labels with LCD displays. This would facilitate dynamic pricing and the ability for the system to plan how much of each product should be displayed.
This was a long time ago so I’m not sharing any secrets here, but in the end we did a lot of research and testing and it never went ahead. We didn’t quite have the flexible computing needed to start planning dynamic shelf allocation, but more importantly it was felt that LCD shelves would just get damaged too often. Imagine investing in this system and then still needing to use paper price tickets… the funny thing is that this is now a very current story. Many supermarkets are now exploring the technology needed for dynamic pricing.
During my student days, I also worked in retail, but during this time I was at the sharp end. I worked in a Safeway supermarket in the UK. My work was centered around the produce section, so I was setting up fruit and vegetable displays and ensuring they were always full.
This was one of the most important tasks. We started work early and ensured the display was filled and looking great. During the day customers would take products from the shelves and eventually some sections would be empty. We had to work constantly to refill these empty shelves.
As you can imagine, this was like playing whack-a-mole.
We cruised the aisles, scanning for products that were already gone, or starting to be depleted. Then we scribbled notes, estimating how much is needed, and raced back to the warehouse to fetch products. On busy Saturdays this would be a constant race against time – during holiday periods it was almost impossible.
Imagine if a camera could just watch the aisle and scan for any developing empty spaces on the shelves. A computer system that understands the products on display could update the team, giving them predictions of what refills will be needed soon. If stock is running low then the shelf plan could be redrawn in real-time to accommodate alternative products.
This isn’t just a dream for retail professionals – like me during my student days – it is now a reality. The Goods Checker is a tool designed and built by IBA Group. It’s managed as a SaaS tool so it doesn’t need any sophisticated onsite hardware.
You can follow the links to click through and read the more detailed specifications, but what I found interesting – as someone that has previously had to do this work manually – is that it not only advises on when stock is needed, but it has a range of analytics too.
It can show how full the display was at opening and closing time. It can diagnose problems that may have caused lost sales. It can also help to predict times when the restocking process will need to be scaled up because more customers are expected.
The bottom line with a system like this is that there is a direct line from the purpose of the technology to improved revenue and profit – it’s not just a tech solution looking for a problem. The IBA team found that store displays were 25% more full during the busiest times. This directly translates into an increase in sales as customers enjoy their shopping experience more and are also less likely to ever find an empty shelf – in fact this is one of the most important functions of the system. It will always alert staff to an empty display section so they can immediately refill or change the display.
Jobs like restocking retail shelves are often positioned as simple and requiring few skills, but the process of ensuring that stock in on display in the right place and at the right time is critically important. We know how it feels to walk into a supermarket and see empty shelves – these days people will often post photos on social media asking ‘what is wrong here?’
Retail brands that want to ensure they always have the right products on display – without gaps – should certainly be exploring systems like this. I know that in my earlier days, stacking produce in a Safeway supermarket, it would have saved us a large amount of time if we knew what was needed – and when – rather than having to constantly guess this manually.