9 AI in retail examples to watch out for

1. Recommended pricing markdowns

More retailers are turning to AI for help with markdown optimization. 

Markdown software can bridge the gap between sale prices and what consumers are willing to pay. It looks at a product’s price elasticity against forecasted and base demand to calculate the predicted demand for any new price.

For retailers, optimizing prices can help them decide on the best markdown price for each SKU and create campaigns around markdowns to maximize relevant KPIs.

Recently, a leading UK retailer used this type of AI to optimize 15% of its stock file. Working in partnership with Peak, this retailer used AI to combine data from website analytics with other sources like sales and ERP systems to get an accurate view of demand per SKU. When plugged into machine learning algorithms, the retailer found the “perfect price range” on an individual product level and increased sales.

The results? 

The changes identified an opportunity to add $3 million in additional margin — the equivalent of 1% of the retailer’s overall turnover. 

2. Stock replenishment

Not only can AI predict what stock you need, but it can also automate the restocking process. Using real-time inventory monitoring, AI automatically detects when an individual item is running low on stock or about to hit your minimum safety stock level.

For retailers, this eliminates the guesswork behind what stock to order or remembering to place a replenishment order with a wholesaler.

Let’s say a retailer sells a particular brand of shoes. It stocks 46 different lines, some more popular than others. Using AI, the retailer can track individual SKUs and set custom safety stock levels while also monitoring things like seasonality and ongoing promotions.

When an individual SKU is running low, an AI platform like Peak can now automatically alert and prompt a retailer to send a recommended replenishment order.

No more stockouts. Just happy customers 😊

3. Smart shelves and labels

Smart shelves are changing the way retail outlets like supermarkets manage customer-facing inventory. These shelves can monitor inventory levels, track when stock is running low and even detect when customers pick up stock but then put it back onto the shelves.

This tech is relatively new, but The Brainy Insights estimates the smart shelves market will reach $30 billion within the next decade.

While many supermarkets are still considering using the tech, some are now moving to electronic barcodes to reduce stockouts and optimize inventory. Global supermarket chain Lidl recently announced it would introduce electronic shelf labels nationwide, which it claims will save 206 tons of carbon each year by eliminating paper labels.

Following suit is US retail giant Walmart. After trialing digital shelf labels, 2300 stores across America will now use them to optimize stock replenishment and improve order picking and fulfillment.

4. AI-powered recommendations

Imagine walking into a store where the salesperson knows your style, your favorite colors and even your shopping budget.

That’s AI at work. It uses algorithms to sift through customer data like previous purchases, browsing history and even social media activity to recommend products you’ll love. E-commerce powerhouses like Amazon and eBay are probably the most recognizable retailers using AI-powered recommendations to drive more sales at the checkout.

With Amazon Personalize, retailers can use generative AI to create batch recommendations of related items for their customers.

Image source: AWS 

As generative AI learns from continuous inputs, the recommendations will improve over time and learn what products customers like.

5. AI-powered self checkouts

Self-checkouts have been around for years, but AI-driven systems are now filling the gaps.

Up until now, a self-checkout still required you to manually input information about certain items, like hand-picked fruit. Now, algorithms inside a checkout combine 3D vision with AI to instantly recognize an item’s shape, texture, size and color to correctly identify it and add it to your shopping basket.

AI can also help with other transactions, like alcohol, that used to require a human to verify someone’s age manually. Checkout terminal manufacturer Diebold Nixdorf recently released an AI-powered checkout that estimates the age of customers buying alcohol. If it thinks a customer is under the age of 27, it calls for an employee.

🎥 See it in action here. 

The new checkout also uses computer vision to monitor if a customer doesn’t scan an item or only scans one of several items to minimize losses.

6. Inventory management

All retailers know that juggling inventory levels and maximizing margin — all while avoiding stockouts — isn’t easy.

AI can take all of your existing inventory data and use it to optimize safety stock and minimize storage costs while always making sure products are available for customers. By analyzing data like customer orders, product, location, historical sales data and even seasonal trends, AI can ensure in-demand items are always in the right place at the right time.

As AI already has the data points it needs to make complex decisions, it can also:

• Forecast against demand volatility and create optimal safety stock levels
• Leverage forecasts against stock levels and backorders to decide what reorder points products should have

Think of it as a (really quick and smart) inventory management assistant 👌

7. Virtual fitting rooms

Shoppers can now try on clothes in their homes before they buy them, all thanks to AI. This is great news for the 55% of people who were disappointed when an item they purchased online looked different than they expected when it arrived in person!

Google’s recently-released virtual try-on (VTO) for apparel tool drives this tech, which uses a new diffusion-based AI model to get a realistic idea of what a worn item will look like in reality. The model uses diffusion to add extra pixels to an image to create “noise” and then removes them to rebuild the image in crystal clear quality.

Thanks to this granular detail, the AI model only needs one image of an item to accurately reflect how it will fit a customer.

8. Visual search identity

Imagine seeing a pair of sneakers you absolutely love — but you don’t know what brand they are or where you can buy them.

Using AI-powered visual search, you can now search for the availability of the sneakers anywhere by simply uploading a picture. Next, image recognition software scans the uploaded photo for things like color, shape and texture. The image will then be run through a database to find the exact product match.

This technology isn’t just cool, but super practical. It allows customers to find the exact product they love without searching for descriptions or specific branding.

9. Supply chain optimization

Finally, AI can optimize a retailer’s supply chain by tracking everything from delivery routes to supplier efficiency.

Think about all the moving parts in logistics and delivery. A retailer must think about warehouse storage, optimal delivery routes and overall costs. Using AI, retailers can analyze traffic data and delivery schedules in real time to find the best routes while calculating wholesale costs from suppliers based on raw materials.

Retailers must be ready for AI

AI is redefining what is needed to succeed in retail not just in 2024, but in the years to come; the tech is no longer a “nice to have” for retailers, it’s imperative to their long-term futures.

From personalized shopping and smart inventory management to virtual changing rooms and 3D checkouts, it’s a tool that can help in every corner of a retail business. More importantly, AI is trending towards becoming an expectation from customers — and it’s up to retailers to embrace this technology to improve their experience rather than disappoint them. Long story short, it’s time to jump into the AI pool with two feet.