In these questions and answers, Martin Smethgurst (pictured below), MD EMEA, Trax, discusses with RetailTechNews the work that Trax does with retailers, and why industry adopts computer vision technology, and what retailers need to make the most of this technology.
RetailTechNews: Can you explain what Trax is doing and how technology works?
Martin Smith Hearst: Trax Is a world leader in computer vision solutions for retail. We work with major CPG companies such as Coca-Cola, AB InBev and Heineken, as well as major retailers, to draw and analyze what is happening on the physical shelves.
Our computer vision technology uses a combination of artificial intelligence, precise image recognition, and automated learning engines to mimic the human visual system. Recognizes images and then puts them in context to convert store images into actionable visions. For example, we can identify the exact products on the shelf and provide statistics on how this product performs compared to its competitor on the shelf below, or even the larger version of itself.
At the core of our technology, neural networks, is a variable innovation in the field of vision in the computer, allowing us to excel in the discovery of things inside the shelves images, and categorize these images with unparalleled accuracy. Our deep learning architecture is trained on billions of shelf images, so that nuances can be recognized, such as a large and small version of the same product, and can overcome challenges such as light, reflection, background chaos, and poor partial hindrance.
Since the images are taken from multiple perspectives, we designed a sewing engine that details the shelves and displays them on a single frame of reference. This visual reconstruction produces an ideal 3D architecture that solves the problem of refined or removed products, which in turn represents the actual shelf in the store digitally.
What are the benefits of retailers to computer vision technology?
Quite simply, it allows retailers to digitize their actual store and compete against the rise of e-commerce. The computer vision allows retailers to take advantage of data statistics, which were very important in the emergence of online shopping, to get a full view of the "health" of each physical shelf.
In retail today, it is common for CPG companies to send products to retail stores every day of the week, sometimes even several times a day, but often have no idea of the shape of the shelf. This is very important, because the way the shelf looks is already determining how much the company will sell. By taking advantage of technological innovation, such as computer vision, this is where a retailer can give an organization such as Coca-Cola or Nestle a picture of what is happening in stores around the world, so that these companies can reduce stocks or achieve the most profitable offers.
We work with customers to discover the ideal amount of shelf space for the payment brand. Some products are very flexible, so the more space you give to a particular product the more sales you have. Then there are other "inflexible" products. No matter how much space you give them, they will still sell exactly the same amount. In fact, if you reduce the amount of shelf space, they may still be selling the same amount.
Using our technology and using some truly intelligent data science, we are able To start To identify flexible products and increase their share of this shelf; meanwhile, we can also identify the inflexible products which can reduce shelf space so that they can save money without reducing profits.
Do retailers now benefit from computer vision technology? What are the challenges that hinder adoption and how can they be overcome?
Some leading companies have embraced this technology and have seen tremendous benefits, but we want to see this is the norm in actual retailing, where traditional stores are battling against large online retailers.
In terms of challenges, the industry has been stagnant for decades, with retailers and CPG sticking to traditional means, such as manual checks, to see how products look on shelves, how they compare with their peers, and whether items out of stock are being replenished. The main thing is to show companies how this process is not only a waste of time and old, it is also prone to error, which can be greatly reduced by computer vision – our technology boasts 96% +.
Moreover, it is about highlighting the benefits of this technology, such as allowing CPGs to make enforceable decisions. Not only will it tell you that your share of the shelf is not as good as it should be, but it provides statistics on how to improve it. Using computer vision and deep learning methods, solution can identify products and provide digital shelving data in full. Companies can give a real insight into the store so they can figure out where and how they can improve it.
Once retailers understand how their products are on the shelf, how can they improve them against these data?
Trax gives companies complete digital shelving and performance degradation. Once they are optimized, data can allow companies to identify the best product lineup for a store, understand the best product site and the best screens, as well as pricing and promotion strategies. Are likely to succeed.
Once all these things are realized, retailers and CPGs can make the necessary changes and, ultimately, improve profitability. It can reduce the hours spent in checking, reducing errors, increasing the productivity of salespeople, and thus increasing sales.
What is the next step in Trax?
Having now worked with some of the largest names in the retail business, Trax is now expanding the platform to include additional geographic areas, covering more time during the day, adding more frequencies and data accuracy, and moving to additional forms of trade.
This will establish Trax as "Bloomberg of retail" and "one-stop-shop" for rack data, and CPGs will be able to access data in any store in any category.
Where we can take a step further is to monitor and anticipate trends very quickly, for example the flavors of drinks that dominate the market. This can help retailers and CPGs understand their target markets better and give retailers the opportunity to move forward in these trends and market accordingly.