Leveraging AI in Retail Pricing: Dmitry Ustinov’s Strategies

AI pricing uses artificial intelligence to set the best prices for your products. It looks at large amounts of data, such as sales history, competitor prices, market demand, and customer behavior, and sets optimal prices that increase profits and boost sales, according to the pricing software platform Symson.

At the forefront of leveraging artificial intelligence (AI) to achieve effective pricing solutions is Dmitry Ustinov, an associate partner at a leading management consulting firm. To date, he has successfully run over 20 projects globally that leverage AI to optimize pricing, including his elasticity-based localized pricing and personalization solutions, ultimately driving significant top-line and bottom-line growth for his clients.

Now, he shares his strategies, including down-to-earth examples of how AI is changing retail pricing and increasing retail efficiencies. While we couldn’t talk to Dmitry directly due to his confidentiality restrictions, his recent publications in Forbes and other major media, as well as testimony from his clients, allow us to shed some light on the recent innovative usage of AI in pricing.

Dmitry’s journey in the development of effective pricing solutions did not come as a surprise. He started using AI to optimize pricing solutions due to his solid background in analytics and consulting. After finishing his studies, majoring in applied mathematics at the Moscow Institute of Physics and Technology in Russia, he came to work at several renowned places, such as Yandex, IBM, and Boston Consulting Group (BCG), and a series of analytical startups. These positions essentially laid the technical groundwork for his expertise in machine learning and advanced analytics, which he now exploits in the commercial sector. These positions and unique work experience across different sectors allowed Dmitry to build a unique expertise and capabilities in the field of applying AI to solve growth tasks for B2C companies.

Another factor contributing to Dmitry’s success is the global footprint of his impact. Dmitry has worked across Eastern Europe, Central Asia, the Middle East, the United States, and Latin America, mostly focusing on retail, telecom, and other B2C industries. This is where his innovations and expertise belong, making a significant impact on these sectors.

And of course, the doer’s mindset contributes to Dmitry’s success. You do not see many management consultants who can roll up their sleeves and do coding. But Dmitry is one of those. He constantly tests the most innovative ML frameworks and applies those to practice. For example, he won a silver medal in the Kaggle Microsoft Malware prediction challenge, being placed in the top 4% of competitors, among the top-performing data science teams and AI researchers across the globe. 

One of Dmitry’s most outstanding works remains the development of elasticity-based localized pricing approaches. While the concept of elasticity has been known for decades, implementing it in a real-life environment for a retailer, tech, or telecom company is extremely challenging. This difficulty arises because it is hard to estimate real elasticity, which depends on multiple factors and is often affected by local events, seasonality, and other variables.

So, the core idea is simple: adjusting prices can optimize sales and profits. Many companies attempt to pass on costs indiscriminately to customers, which can be dangerous. Sudden price increases can reduce sales and erode customer trust.

The real value lies in smart price decreases. In the current unstable macroeconomic environment and constant inflation, it’s worth asking how much we can decrease prices to attract more customers. Identifying elastic items, where a price drop significantly boosts volume, is crucial. AI-based approaches help in making these precise adjustments, leading to increased purchases. Dmitry was the architect behind those AI-based approaches, piloting and scaling them across the globe.

This strategy has three key effects: first, direct increased sales of the discounted item; second, additional sales of other items as customers buy more during their visit; and third, strengthening the trust bond between customers and the company. Customers trust companies that offer fair prices, fostering a win-win relationship. This approach allows companies to thrive, customers to buy more and improve their well-being, and overall economic growth by boosting consumption.

This approach was already implemented at a number of retailers and quick service restaurants of massively different scales, from major European and U.S. players with 20,000 stores to small local players in Latin America with 50 restaurants. The impact was nothing short of exceptional, leading to over 5% increase in earnings before interest, taxes, depreciation, and amortization (EBITDA) and increased customer satisfaction.

In his latest series of articles on leveraging AI and machine learning for retail, Dmitry highlights that a real win-win can be achieved through personalization. The idea, in a nutshell, is to use AI and machine learning algorithms to understand what customers really need in order to provide the offers that would interest them most and where each individual customer can be the most elastic. This requires utilizing the latest advancements in AI, and it has been an extremely hot area for the last 10-20 years, with major companies like Netflix and Google working on their own recommendation systems. Now, each retailer can leverage these technologies through open-source libraries. But the real question is how to implement those technologies in the real-life setting of a brick-and-mortar retailer or a traditional telco company and ensure it brings incremental dollars.

However, what’s also critical, as Dmitry mentions in his articles, is that on top of the recommendation engine, another economic layer should be applied, either through a Next Best Action (NBA) model or a Next Product to Buy (NPTB) model. This layer should determine the total economic impact for the company and the client, prioritizing opportunities accordingly. This approach can provide an additional layer of win-win because it ensures the right deals are offered to the right segments of customers. Implementation of this methodology at scale back in the 2010s was the first of its kind, expanding the horizons for retail and telecom companies, and Dmitry was the mastermind behind this.

The most significant impact of this methodology comes not from squeezing margins from some segments but from providing extremely good value, leading customers to buy massively more. This is a game of very low margins where every additional percent of discount is a business-critical decision and can only be optimized through AI and ML models. These approaches were successfully implemented across a number of retail and telecom companies globally, each getting 5-10% incremental EBITDA. Total financial impact already exceeds $500 million.

In his recent article in Forbes, Dmitry also talks about the AI path going forward, focusing on GenAI implementation. “While this is definitely a revolution, many companies are still unclear about its implementation. This is the next big frontier,” he says. “In several years to come, every company will leverage generative AI, and the question is how to make it in the most efficient way.” Dmitry goes beyond GenAI hype and focuses on the real challenges that companies face and ways to overcome those challenges through technical means (e.g. new approaches to machine learning operations (MLOps) as well as business factors (e.g. structure suppliers’ contracts to ensure shared incentives). The way forward is not just AI advancement or innovative management practices, but a properly calibrated mixture of both, he adds.

Dmitry is not done yet. Despite these achievements, he plans on developing more advanced pricing mechanisms that will meet the needs of companies in the low-income sector. One of the ways through which he intends to support the development of these businesses is through the implementation of custom strategies to address the specific challenges they face with the hope that these companies will be able to achieve sustainable growth and profitability.

All in all, Dmitry Ustinov’s use of AI in pricing has opened the door to unlimited possibilities in the retail sector, bringing to it effective and transformative changes and pointing new directions in the industry. His work is a clear demonstration of the power of technology to enhance both productivity and profit, and his ongoing efforts promise to further revolutionize how retailers approach pricing in the years to come. As the retail sector continues to evolve, his contributions will undoubtedly remain at the forefront of pricing innovation, shaping the future of commerce in profound ways. “AI is more than a tool for us; it is a power to create an environment that redefines the way businesses function,” he concludes. “Our mission is to expand the limits of what is possible in pricing and to show clients the value we can deliver in ways they hadn’t even dreamed of.”