These days, everyone is talking about artificial intelligence. The possibilities are exciting and seemingly unlimited, but many brands are struggling to figure out how to put “unlimited possibilities” to work in specific, meaningful ways.

Here are three ways brands can use AI today to improve online and in-store retail experiences for their customers.

Personalize product recommendations.

Anyone with an Amazon account is familiar with the concept of customer recommendations. This AI application allows retailers to serve customers tailored product recommendations based on past purchases, browsing history, and popular trends.

More and more brands are harnessing the power of AI to implement this capability across their digital channels, from websites to email campaigns. And they are seeing results. It’s estimated that 35% of Amazon’s sales come from the recommended products feature.

Here are a few different ways to do product recommendations using AI:

• Cart abandonment. Jewelry retailer Blue Nile does an excellent job of re-engaging stalled buyers. An abandoned cart email “sandwich” reminds the shopper of the original product they left in their cart and makes recommendations for similar products.

• Second sell. Retail brands can use the checkout process as an opportunity to recommend complementary products to buyers. For example, a fashion retailer could prompt a buyer purchasing shoes to add leather conditioner or a pair of dress socks to their purchase as they finalize their order.

• “Recommended for you” products. Many brands, such as Etsy, feature personalized recommendations on customer account pages or send them through email. While making accurate recommendations requires amassing data over time, it is a strong component of any retention marketing strategy, as it encourages and rewards customer loyalty.

• Similar products. Suggesting similar products makes it easy for shoppers to compare features and prices.

Use data-driven in-store experiences.

Many retail brands are beginning to take online tactics, like personalized recommendations, into their physical stores. For example, if a customer has viewed a product online and then walks into a store, brands can serve them an offer for that product or something similar.

Brands can use in-store monitoring systems to collect the same sort of valuable data that digital traffic provides. These systems can determine how customers physically travel through your store: where they spend the most time, what parts of the store they skip, even how they react to products and displays. Brands can use this data to build seamless, customer-centric, data-driven in-store experiences the same way they would for their digital experiences.

The epitome of this, of course, is the Amazon Go store concept, which uses machine learning to create an automated retail experience and has generated a great deal of buzz. In the Amazon Go store, AI technology monitors each customer’s cart while they shop, and bills them automatically when they leave, which negates the need for waiting in a checkout line, and provides Amazon precise customer data.

There are many ways brands can integrate online and in-store experiences without going full out like Amazon, though. Here are a couple options:

• Ship to store. More brands are taking advantage of the opportunity to combine in-store and online retail experiences. Customers enjoy the free shipping, and brands get a primed opportunity for a second purchase. Allowing customers to pick up online orders at a physical store also encourages a brand’s digital and in-store teams to share data and technology and to hone the way they work together, leading to better experiences for employees and customers alike.

• Easy access to product information and reviews. Allowing shoppers to immediately access product information via a QR code, product code, or even image recognition capitalizes on the quick decisions they make every day. Google has coined the term micro-moments to describe these instances.

Image recognition search.

While still considered an emerging technology, brands can already use image recognition to understand a customer’s individual tastes and recommend products tailored to them. Pinterest, for example, uses this capability to look at an image and suggest similar content.

In a retail situation, this allows brands to offer each of their customers access to their own digital “personal shopper.”

Nearly 60% of shoppers use their mobile phone to research products while shopping in a store. Despite having no brick-and-mortar stores, Amazon recognized this as an opportunity. By incorporating image recognition right into the search function of its existing mobile app, Amazon makes it easy for shoppers to compare products.

Darrell Etherington of TechCrunch noted, “Like a cowbird laying its eggs in the nests of wrens and sparrows, Amazon is using the retail floors of its competitors to demo products, while landing the final sale through competitive pricing.”

As another example, IKEA uses image recognition to allow shoppers to snap photos of images in the company’s catalog to pull up information about that product, swap colors and patterns, and even see how the product would look in their own home.

Retailers that recognize opportunities in what others deem challenges or barriers will be the ones who achieve the full potential AI has been promising for years.

Learn how Salesforce Einstein can help your brand harness the power of AI.