How AI is helping retail companies

This week, we flew to Dusseldorf to attend EuroCis, the leading trade show for retail technology to see for ourselves what are the most pressing problems that technology, and AI in particular, are solving for retail companies globally.?

During Covid times, users got used to ordering online and the online business grew.? Retailers now face the challenge of merge offline and online businesses and seamlessly connect digital and in-store channels. In online retail, everything is measured and tracked, and technology is helping stores to collect data so managers can measure their stores performance, detect problems and take measures to optimize their businesses and profits. Customer is king, and retail stores need to engage their clients with personalised experiences, in order to display the right product, at the right time, to the right person. And, once the client has made the decision to purchase, retailers need to offer a seamless cash-out experience avoiding un pleasant queues with cashless, and even contactless payments.

On the other side, the supply chain is suffering a global mess so optimizaing the supply chain is crucial for keeping sustainable profit levels for stores. Connected stores allow to have better user insights, which allows to make better demand predictions so retailers can optimize their purchasing processes, inventory levels, warehouses and distribution centers.

Ai is helping retail companies in many ways, from smart replenishments, predicting demand & supply, optimizing supply chain and inventory stocks, providing in-store analytics to measure stores operations and performance, to personalized recommendations to customers walking in stores. These are some simple AI implementation examples that every retailer can implement today.

In-store analytics. Using cameras (in can be the cameras which are already installed in the stores, or some simple webcams), we can count people and know which aras are most utilized. Counting people is the most important, and simple, step to measure the store performance. By knowing how many items are sold, and how many people visited the shop you can do an easy performance ratio. As simple as this may sound, most stores still are not using this simple measure. On top of that, we can have heatmaps on how clients behave to understand store space utilization. AI can also determine age, gender, mood, how much time it spends on a certain spot which becomse very useful for retailers to measure their promotions and calculate if their marketing efforts are having an acceptable ROI.

Shelf monitoring. Not having products on display is obviously, a huge problem for stores. In fast moving goods and drinks in particular, shelves need to be replenished few times a day, and an empty shelf for a few minutes can mean a loss for the retail store. Also, detecting products in the wrong shelf, or competitor brands next to each other can be a problem. A simple solution is using cameras that can detect empty shelf, and wrong product position.?

Dynamic pricing. Adjusting prices dynamically based on demand changes can increase sales, profits and reduce inventory stress. Ai can help with sophisticated models that predict demand taking into account many factors such as competition, seasonability, user behaviours and add business rules depending on the pricing strategy the store wants to push. Do we want to push specific products to specific audience? Do we want to reduce some specific inventory? Do we want to beat som specific competitor on a specific product(s)? Ai actually allows to something even more sophisticated, which is finding patterns in the data and basically retailers can be recommended what to sell at what price.

Stock & Demand prediction. Simply by having access to the ERP system data, AI can build time series models which make better demand predictions. There are a big amount of factors which can influence the consumption across retails stores. The same product has different consumption patterns in different locations, and if add in the mix weather and seasonability, traditional statistical models can become too big to handle efficiently. AI can learn those patterns from the data, and make better predictions which can be used by retailers to do a better forecasting. Retailers can negotiate better deals with their vendors and secure delivery with more accurate order planning during the whole year. For example, a retail store can place an order for the next 6 months with an exact delivery plan for the incoming months.

Supply chain optimization. AI can’t help unblocking a container ship blocked in Hong Kong, but definitely AI can help in many ways throughout the supply chain, which is a big issue globally nowadays. Optimizing truckloads, delivery schedules, pick optimization, predictive ETAs (estimated time of arrivals) for deliveries, staff planning, dynamic replenishment, probabilistic inventory, category management… AI helps optimzing all these important steps of the supply chain to maximise profits. The challenge of course is the source of truth (the data), and while connecting to the ERP is the easiest way to get started, there are also other ways to overcome this problem, for example to connect AI models directly to Excel sheets.

At Etnetera, we develop smart solutions for our clients so they can get results with AI in weeks, not years. If you are interested in finding how you can improve your retail business with AI, reach out and we will be in touch in a timely manner.

https://www.etnetera.cz/en/ai