Sign up

Radio Veronica One

Radio Veronica One – Interview to Giuseppe Craparotta

“Tutto molto radio”

Torino, July 11, 2018

By Manuel Giancale

Manuel: “In our second hour we’ll be talking about shopping. Above all we’ll be discussing how much technology has entered our stores. How it’s the technology that chooses where to place certain items, what color to choose or what to emphasize most. This is because we, like artificial intelligence, unfortunately are (or fortunately are, depending on the point of view) persuaded into buying a certain item rather than another. Thanks to Evo Pricing we’ll find out what is behind all this because we will have an Evo Pricing expert on the phone.

Let’s start our second hour, we are now half way through and the same with our patience when women take us – I speak for the men – when women take us shopping and we get into a shop to discover that all the clothes have been arranged in a certain way, not just due to the storekeepers’ taste but with the added help of technology. In short, behind everything that has to do with sales, there’s an artificial intelligence: it’s something that I’m personally a bit afraid of but today we’ll learn to understand it better with Evo Pricing. Soon we’ll be hearing from their Senior Data Scientist Giuseppe Craparotta who’s going to explain what is behind the organization of a store, especially chain stores.

On Radio Veronica One’s “Tutto molto radio” program we talked about marriage in our first hour, a marriage also implicates having to go shopping with your partner. This has certainly already happened to you, you’ve been in a shop and in a chain store, the bigger shops, like, I don’t know… no, I can’t make names, alright, because here at the radio they’ll get mad. In any case, chain stores are usually super organized, to the point that store managers are given a picture of how the store windows and even the shelves need to be displayed for specific reasons, and behind all of this you have artificial intelligence. So, I ask myself: “How is it possible that artificial intelligence can understand exactly where to put things, what position to put them in, how high they should be, what colors to put near each other, etcetera… all thanks to a, let’s say it, in quotes “a simple algorithm”, simple, so to speak and I’ll ask Senior Data Scientist Giuseppe Craparotta from Evo,, to find it all out. How can artificial intelligence make us buy those exact things? Welcome to Radio Veronica One, Giuseppe.”

Giuseppe: “Thank you, thank you. Greetings to all the listeners and thank you for having us. How is all this possible? First of all, we need to have structured data, if you’ve heard of big data, you can imagine the amount of data that these stores can have with all the stores, items and amount of consumers who go often into their stores. So, then what does artificial intelligence do? In reality, it doesn’t do anything other than extract information from the data and give suggestions.

For example, you can calculate the probability of a certain item that will be sold in a certain store and, for example, the number of pieces that will be sold. This certainly depend on many factors like, price, the characteristic of an article or the weather. What we at Evo Pricing do is put all this data together and sometimes even with help from the store managers, who can give us their opinion on the local tastes and trends. At this point, when we have the sales probability for each item, in each store and put it all together. Then we can find the best location, where the items should be put and the ideal price, obtaining a good balance.”

Manuel: “I’m sorry, Giuseppe, just to be sure I understand: we could find swimming suits even in the winter and winter ware in summer, because artificial intelligence understands that even in the summer people tend to think about winter and vice versa in winter.”

Giuseppe: “Exactly. In this case, artificial intelligence tries to extrapolate and understand what the emerging trend is. If there’s a need and there is an availability of that certain stock piece, for example at the central warehouse, the artificial intelligence will know where to allocate the pieces and which stores to send them to.”

Manuel: “So nothing left to chance, is the concept?”

Giuseppe: “Nothing is left to chance.”

Manuel: “But in the store managers taste there is still a bit of the human touch?”

Giuseppe: “We love defining ourselves as a company that uses artificial intelligence but we do remember to add the human heart and human intuition to the base of each of our algorithms. So, we really do get help from who, in the end, is the industry expert. Who’s more of an expert than the person who has to do with customers every day?”

Manuel: “Right, rightly so. You deal with large companies, but of large chain stores because the smaller shores have less data to be able to create an algorithm and therefore an artificial intelligence ad hoc to learn, to position, to define prices, etcetera etcetera…”

Giuseppe: “Yes, we primarily, deal with large distribution because that’s where we can get greater data complexity. Of course, we can service the small shops, we can suggest that they follow our method, which is to try and exploit as much of the data they have and use a bit of their intuition, merchant’s intuition.”

Manuel: To ask the customer in layman’s terms we would say “What did you like, what would you like to see up there or down there, I guess, that’s somewhat an idea of what artificial intelligence is.”

Giuseppe: “Yes, true, because in the end artificial intelligence is nothing more than the listening to data, trying to read it and then understanding it, but for sure the first step is reading and then at times even asking the customers what they want.”

Manuel: “There’s also a new app that will automatically tell the shopkeeper where to place and at what price to sell certain items of clothing, an app that you are making, right?”

Giuseppe: “Exactly. This is a project that we are about to finish and will probably make every merchant, that deals with planning, dream come true because we will simply be telling them to just “Take a picture of the item you are interested in putting in your offer and we’ll tell you how many pieces it will sell, what the potential pieces of that item are and what it would be, what could be an ideal selling price. Clearly this begins with the recognition of different parts of an image from a photo.”

Manuel: “Can I say that at this point, I’m a bit afraid of technology? That is, sure in a certain sense you always have the human factor because it’s an alliance between artificial intelligence and humans and you need it all to do business, that’s fine. But what scares me a bit, that is a shopkeeper taking orders from the technology: no, put it higher because down there it won’t sell or take that color rather than another…”

Giuseppe: “In the end our role is to give recommendations.”

Manuel: “Sure.”

Giuseppe: “It’s true that best recommendations for those doing the implementing, is to have common and a critical sense. In fact, when you asked me what type of customer does it make sense to do this for? It makes more sense for the bigger customer with a large number of stores because managing this kind of a processes by hand is more difficult.”

Manuel: “Of course.”

Giuseppe: “So from this point of view, technology is of help to us.”

Manuel: “Absolutely yes. Thank you so much Giuseppe Craparotta for your testimony, for showing that even behind artificial intelligence, in any case, there is still a human, thank goodness.”

Giuseppe: “Exactly.”

Manuel: “Because if not, they will begin making machines amongst themselves and we’d no longer exist on Earth. Thank you so much Giuseppe, have a good day and work!”

Giuseppe: “Goodbye.”

Manuel: “Bye, thank you.”

Transcript:    English   /   Italiano


For press-related queries please contact

Hey! Was this page helpful?

We’re always looking to make our docs better, please let us know if you have any
suggestions or advice about what’s working and what’s not!

Send Feedback