Thought Leaders in Artificial Intelligence: Antuit Co-CEO Sivakumar Lakshmanan (Part 2)

Sramana Mitra: Is this the use case that you started your company with?

Sivakumar Lakshmanan: There are two primary use cases. One is around forecasting and the other is a pricing use case. The pricing use case determines the price at the retail shelf and how to mark down the product. 

Sramana Mitra: Why don’t we do the pricing use case?

Sivakumar Lakshmanan: Let’s take fashion retailers. It could be apparel or Michael Kors. These are companies that operate on season. They have four seasons in a year. The products stay for a season and then they’re off the shelf. At the end of a 16-week life of a product, the inventory is not useful. You need to markdown and clear those products.

What is the right time and right level to mark down the product? You leverage data and intelligence. You don’t want to go deep too early. You also don’t want to have leftover inventory. Your timing and the depth become very important. In the last two years, that use case has manifested itself into an omnichannel fulfillment use case.

COVID activated e-commerce. I don’t remember when I went to a store to buy a shirt. We buy online. That trend means that I need to make a more holistic decision around how I deploy my inventory in my store and how I fulfill orders fully knowing that I still need to mark down the product. I’m looking for a purple shirt and I’m buying online. I live in Texas. There is a store nearby that has one medium-sized. There is a store near Dallas, which is 30 miles from where I am, that has 12 of the same shirt that I want.

Now I need to make a decision on how to fulfill demand. Do I go from the Dallas store which is a dollar more expensive in transportation cost? If you look at it more holistically, you will be looking at your inventory and you’re saying, “If I’m not moving this Dallas store product, I run the risk of marking down this product. Why don’t I incur the $1 extra?” Play this out across thousands of stores with everyone buying online every minute. 

Sramana Mitra: Interesting. It sounds like the retailers you’re working with are fulfilling their online orders from stores and not from a dedicated e-commerce warehouse.

Sivakumar Lakshmanan: It depends on the retailer, but we increasingly see buying online and picking up or delivering from store. As a customer, we need a large assortment next to us. We want the same experience as walking into the store. It’s increasingly hard. 

Sramana Mitra: What data signals are driving these use cases?

Sivakumar Lakshmanan: You need an understanding of demand. Demand underpins most of the decisions. 

Sramana Mitra: So historical sales data is what’s driving most of it?

Sivakumar Lakshmanan: Not necessarily. You need to understand the logistics cost. You need to understand your markdown liability. You need to understand the individual trends of each store and the assortment. If you’re a cosmetic company, one would expect the lipsticks to fly off the shelf when people will take off their masks. There’s a lot more that goes into it than just historical data. There’s a lot of external data. There’s also customer loyalty information. 

This segment is part 2 in the series : Thought Leaders in Artificial Intelligence: Antuit Co-CEO Sivakumar Lakshmanan
1 2 3