Know consumer preferences, anticipate customer desires and offer them the most personalized shopping experience possible: thanks to artificial intelligence (AI) , retail is experiencing a real revolution.
This concept, which scientists have been working on for sixty years, is inspired by human thought, which combined with data, improves the experiences offered by brands to their consumers .
Data: the essential fuel for AI algorithms
Thanks to artificial intelligence, brands can now get to know their customers in depth. Consumer behavior can even be predicted through unstructured data, such as social media posts , which are transformed into understandable data via cognitive systems. Indeed, without data, artificial intelligence is nothing. With the explosion of algorithms and technologies, brands have the opportunity to build a unique customer repository (RCU), allowing them to go deeper into customer knowledge .
For example, by relying on data from their consumers' loyalty cards, combined with data relating to their ages, sexes, identities, etc., to the most frequently used purchasing categories (online or in stores), transaction, navigation, behavior and other data; brands can, thanks to machine learning and deep learning algorithms, personalize their offers . They are also able to detect future trends and predict the behavior of their customers. Artificial intelligence also allows retailers to automate many tasks, such as inventory management. Data thus becomes omnipresent, since it makes it possible to decipher consumer profiles, their appetite for products, their future behavior and even their way of thinking. The marriage of these many data with new technologies therefore offers the opportunity for retail professionals to act in a predictive manner .
Artificial intelligence to predict customer expectations
Thus, artificial intelligence really demonstrates the extent of its talents when it comes to predicting customer expectations. For a little over a year, AI has changed the game when it comes to marketing strategies. The algorithms now offer the possibility for retailers to establish purchasing attitudes for each of their consumers, and thus to predict their future desires so as to suggest to each the product most suited to their current expectations.
For example, during the 2017 edition of Black Friday, a day of promotions inspired by the United States, recommendations for products boosted by AI generated 30% of online sales revenue, and 24% during Cyber Monday (according to a Salesforce study, conducted in 30 countries).
Another real opportunity for artificial intelligence: the ability it offers to retailers to predict customer behavior in real time . Predictive analysis in real time allows merchants to engage the customer by offering them a new suitable service, but also to retain them and optimize the various channels allowing them to process their requests efficiently. As customer behavior data is the key indicator of purchase intention, when coupled with a customer's purchase history, it allows retailers to know which product will fully satisfy a customer and which are those for which a consumer is likely to be tempted.
Artificial intelligence to manage stocks
However, the right products must be available in stock. Again, artificial intelligence demonstrates the extent of its effectiveness vis-à-vis customers , accustomed to the Amazon model, who are really demanding in terms of time and price.
Several algorithms thus allow retailers to forecast their stock and automatically trigger a re-purchase if necessary . These algorithms take into account several hundred variables such as social networks, the weather or the press - which are exogenous data -, coupled with endogenous data such as promotions and last sales. These algorithms allow retailers to understand how exogenous data influences endogenous data. They are therefore able to refine the accuracy of sales forecasts and thus better manage their stocks to satisfy their customers.
AI and retail: some concrete examples
Purchases made via their smartphones, recommendations based on purchase history, targeted promotions disseminated on the right social networks, etc., consumer expectations and their uses are changing. Artificial intelligence allows retailers to subscribe to these new ways of consuming, in an ever more proactive way, thanks to innovations from historic players as well as startups.
Connected shelves from Kroger
The Kroger grocery chain (United States) now has connected shelves , developed in collaboration with Microsoft and other partners. The name of this device: EDGE (Ehanced Display for Grocery Environment). The latter allows the display, instead of the paper labels that we know well, of a wide video tape which can indicate the prices, but also display nutritional information and any other type of content.
The retailer can thus modify prices in real time , offer flash promotions and even personalized discounts addressed to customers who have shown interest in a particular product.
Kroger also aims to use its connected shelves to make it easier for consumers to identify the product they are looking for or the one that will perfectly meet their dietary constraints (allergens, specific diets, etc.).
Lowe's and its autonomous robot
The specialist in construction and gardening equipment, Lowe's (United States) has acquired a newcomer to its team: an autonomous robot of 1.50m! The mission of this somewhat special employee: to assist his human colleagues.
This robot, the LoweBot, is equipped with a voice and visual recognition system , cameras and sensors. He is able to scan the shelves to ensure the inventory of stocks, but also to advise customers who can address him orally or via his touch pad. The LoweBot is fully capable of supporting customers on the store shelves thanks to its sensors, similar to those of autonomous vehicles. He can also display current promotions on his back.
Lowe's teams can therefore focus on other clients with higher added value.
A virtual reality headset to allow customers to project themselves
Microsoft and its 360dgrees.com solution allow consumers to project themselves in a sales context. By putting on this virtual reality headset , customers can, for example, view products that are too large to be displayed in the store, look at different colors, try on the ones they want, or even observe products in a specific environment.
This system can also allow sales teams to train, in a training context. The possibilities are endless!
AI to detect emotions in real time
As early as 2017, Microsoft offered a technology based on facial recognition, called Realtime Crowd Insights. This integrates an artificial intelligence capable of detecting the faces of customers who have already come to a store. The AI can also perform statistical analyzes related to attendance, in real time, such as age, gender and even the emotions of each client. Realtime Crowd Insights is thus able to distinguish seven emotions by scanning the face of a client. Sellers who have access to this data are thus able to predict a client's state of mind and adapt their behavior accordingly .
