Three AI marketing trends for brick-and-mortar retailers

Artificial intelligence (AI) is clearly making its mark on retail marketing, and it’s having the biggest impact in the digital world due to the inherently trackable and quantifiable nature of activity on websites, in digital purchasing and within user accounts. The volume and cleanliness of the e-commerce data make this a perfect environment for leveraging AI.

That’s why it’s no surprise that Amazon is one of the paramount AI powerhouses in the world. The volume of data generated by logged-in user accounts on their website — such as information on purchases, add to cart actions, clicks and searches — is huge and allows them to tailor and customize their offerings and grow the business.

Brick-and-mortar businesses don’t have the same advantages. The data available is not so easy to access, record and analyze. While machine learning is already, or nearly, viable in some aspects of the retail industry, current AI-driven marketing solutions for brick-and-mortar retail businesses are often extremely expensive and have a poor return on investment in their present form.

However, that does not mean that physical retailers will be completely out of the AI game forever. It may take a few more years, but physical retail will begin adopting more of the same strategies that the e-commerce players are able to leverage now in the digital space. It will do this by moving that same trackability and quantifiability into the physical space.

In this article, I am going to cover three separate ways brick-and-mortar retailers will soon be able to leverage some of these cutting-edge AI marketing techniques.

In-store recommendations

In the future, more stores will be able to have relationships with customers by encouraging them to create some kind of user account. There are many benefits of having a user account, as you can see with Amazon. When you have an account with Amazon, you get better product recommendations. You also get one-click purchasing. Every time you are on Amazon.com, or you search for products on Google and you land on Amazon, you get this personalized, super-smooth experience.

The retailer encourages users to register an account so they can access all these benefits and save time. But Amazon also benefits, of course, because a user account is a rich source of information. They know who you are, and they can track your behavior and purchasing activity over time. They use this information to put the right products in front of you, which helps generate more sales. They optimize each transaction for customer lifetime value.

The benefits of user accounts

Physical stores will eventually do the same thing, and some are already doing it by offering store pickups or local deliveries. Take a grocery store, for example. If you are a regular shopper at a particular grocery store, and you have a user account, the store will have access to some of the same information that Amazon gets from a user account.

Instead of logging in to a website, though, you will probably log into an app that store sensors can detect. By your smartphone, the store AI will then know when you come into the store and can track the frequency of your visits and the products you buy. From that information, the store can improve your in-store customer experience in a number of ways.

The store could use artificial intelligence to analyze all the data about you and then provide special offers, coupons, discounts and package deals for products you normally buy. Rather than receiving a bundle of coupons with your receipt at checkout, you could receive coupons as soon as your app indicates that you are on premises. They may offer you discounts and specials for the specific products you buy or offer other brands to consider for your usual product categories.

Anticipating your customers’ needs

The store may also offer you some kind of recurring billing program. They could bill you every week for a bundle of groceries that is ready for you when you show up, with the product list generated by AI and approved by the user in advance. Alternatively, the store may alert a staff member to start gathering the products you usually buy when you enter the store and bring them to the counter for you. If the store app has some type of merchant account setup, you may even be able to do one-click payment as well, using an enrolled credit or debit card if you choose. You are in and out of the grocery store in a matter of minutes.

How does the grocery store benefit from all these special accommodations to regular customers? Just as with Amazon, these user experience extras translate to more sales and higher profits. As any retailer knows, buyers are more likely to be repeat customers when they have a positive user experience. If you make it easy and convenient for them to buy, they are also more likely to buy more. If they feel special when they come into the store, they are more likely to recommend the establishment to others.

Customizing the user experience would also equip the store to do some upselling by offering customers relevant products with higher profit margins in lieu of their regular brands. For example, a customer who typically buys a generic brand of eggs could be offered sustainably farmed eggs at a small discount. The eggs would still be more expensive than the generic ones, but the discount may potentially ease someone into purchasing them anyway, which means slightly higher profit margins. And, assuming the product is worth it, that customer may choose the more expensive product next time.

Alternatively, the grocery store could move certain people to purchase more perishable or overstocked products by making price adjustments based on their buying behavior. For example, if they know you are in the habit of buying salad ingredients, the store might offer you big discounts on red peppers and cucumbers.

Out-of-store recommendations and advertising

AI marketing strategies can also help retailers target customers when they’re outside the store. Still using user account information, an AI-enabled marketing program can determine the best way to get people with certain purchasing behaviors and patterns back into the store and maximize their cart value. This is essentially fielding patterns of purchasing and patterns of promotion to buyers with similar profiles — like lookalike audiences online — to make recommendations for any given buyer.

Driving customers back to the store or delivering to them

Armed with these recommendation systems, retailers can deliver relevant marketing material aimed at getting the customer to go to the store. For example, the store can call, send a text message or deliver an email announcing new products, special deals or unique discounts just for you. If you normally go to the store every Wednesday, and you do not show up one week, they might prompt you with a specific discount or special that appeals to your interest. If you buy a lot of chocolates, they might offer you a two-for-one deal just for that day. This will encourage more consistent purchasing.

Another strategy is to offer some products that you can purchase online. Some larger supermarkets already offer this service in-house. One example is Stop and Shop, which has a service called Peapod, and Walmart also offers this type of service. For smaller retailers, though, it makes more sense to leave the delivery logistics to a third party.

Companies such as Instacart and Amazon Fresh provide an option for local businesses to offer delivery of certain products to their regular customers. It may be possible for grocery stores to go into partnership with these delivery services so people do not even have to go to the store. The store can offer to deliver the goods to them.

Split testing

Split testing is another area where online businesses have a significant advantage when it comes to employing AI in marketing. Also known as A/B testing, split testing is simply trying two things with similar customers and figuring out which one gives better results.

While that is simple enough to do for online retail, it is a bit more challenging for offline stores. For example, if Amazon wants to test a version of its home page, it can present 500 versions of the home page against 500 segments of Amazon.com visitors. It can then test tiny adjustments — to the checkout page, to the buttons, to the text, to the images, to whatever they think might help — in real time to see what happens.

With a physical store, split testing is much more challenging. For one thing, it is a major production to change a physical store’s layout in any significant way. But with the help of artificial intelligence, it may, in fact, be possible to determine the type of person influenced and affected by deliberate changes in the physical environment. You may be able to vary a store’s layout in smaller ways to optimize sales based on the results of in-store split testing.

Taking the grocery store scenario once again, suppose you have a number of people walking into the store that fit into different categories based on their user account information. You could use AI to look at this information and consider changing the layout of certain sections in the store, such as dairy or bakery products.

You may try layout 1 for this week and layout 2 the following week. You can toggle between these two layouts in alternating weeks for a month or two and observe how the changes affect customer behavior.

You will be able to answer questions such as:

  • Did customers buy more products with layout 1 or layout 2?
  • Did customers buy higher-priced products with layout 1 or layout 2?
  • If we display our high-priced products more prominently, do people walk out with more of these products per person than before?
  • Which kinds of users are buying more of these products?
  • Are we able to convert people who normally do not buy high-end stuff to purchase them with this new setup?

You can also experiment with small changes in your presentation such, as varying the wording of your specials and sales to see which versions get people to buy more of a product. For example, you can switch between “50% off” and “Buy One Get One Free” (which is essentially the same thing) to see which presentation works better for certain products.

Another way you will be able to use AI in marketing in the future is through machine vision. This technology uses cameras to identify customers in-store and tie them with their account profiles to detect patterns. With the benefit of lots of data and machine learning, you make predictions about the types of marketing strategies that will work for certain customers and automate the process. Should you be concerned about privacy issues — and many of your customers probably are — you could do this without using facial recognition, provided you have a user account program in place.

These are three ways brick-and-mortar retailers will likely employ AI for marketing in the future. A few more may emerge in the next few years as artificial intelligence makes it way from the digital into the physical space.

These concepts will definitely be much harder to apply in the physical than in the digital world, but when AI does mature in the half-decade ahead, it has the potential to deliver a high return on investment. That’s because a majority of retail sales today still happen in physical stores rather than via e-commerce.  Your best move right now is to wait for that wave of AI innovation to come within reach and be ready to take full advantage.


Opinions expressed in this article are those of the guest author and not necessarily MarTech Today. Staff authors are listed here.


About The Author

Daniel Faggella is an email marketing and marketing automation expert with a focus on the intersection of marketing and artificial intelligence. He runs TechEmergence, a San Francisco-based market research and media platform for artificial intelligence and machine learning applications in business.

<!–
–>
https://martechtoday.com/feed