Sign up for our newsletter and get the latest big data news and analysis.

Optimizing Fuel Pricing in a Convenience Retail Environment with AI and Machine Learning

In this special guest feature, Niels Skov, SVP, PDI Fuel Pricing Solutions, outlines how fuel pricing is a complex business for convenience retailers. Using advanced digital capabilities like AI and machine learning to get fuel pricing right can have a significant business impact far beyond an operator’s raw margins on gasoline or diesel. Niels is responsible for the strategic vision for the fuel pricing solution set and working cross-functionally to deliver PDI’s solutions to the market. He has a strong track record of business transformation, technology leadership, and commercial success in data-driven markets and brings vast global experience to the team with a career that has taken him all over the world. Before coming to PDI, Niels was with Genpact, where he was responsible for building new solutions and service offerings for enterprise customers of the Contract Management product line.

In the U.S., convenience retailers sell an estimated 80 percent of all fuel, representing on average 69.9 percent of a c-store’s total revenue. With so much hinging on fuel sales margins, it’s critical for operators to get their fuel pricing right. Fortunately, automation, AI and machine learning can help convenience retailers optimize this core aspect of their business.

A Historical Perspective  

Retail fuel prices have traditionally been set based on the best available data. Decades ago, that meant considering the wholesale cost of fuel and the rates a convenience retailer’s immediate competitors were offering. Establishing a desired profit margin and determining what consumers were willing to pay have also been key considerations. These data points are still at the heart of fuel pricing strategies, but three things have changed significantly: there are now many more data points to consider, information is available in real or near-real time, and very little of this process needs to be done manually.

Today’s convenience retailers don’t have to manage their fuel pricing based on static margins or by driving around checking on competitors’ posted per-gallon prices. Instead, automated pricing tools (some more sophisticated than others) are available, allowing c-store operators to modify the price at which they’re offering a gallon of fuel as many as 20 times a day, dynamically and without the need for manual entries.

The Data Advantage

Even convenience retailers who don’t adjust their advertised price regularly can benefit from an automated, data-driven approach to fuel pricing. A pricing strategy and system that accounts for more than just the essential supply and demand data will produce a more optimal price and profit. Comprehensive pricing systems incorporate historical trends, data from in-store (non-fuel) sales, loyalty program data, seasonal trends, special events, time of day, and even the performance of the convenience store network (if the c-store is part of a chain). All this information provides more precise and profitable fuel pricing, but also requires sophisticated technology to make sense of it all.

The Role of AI and Machine Learning

This is where AI comes in. Fuel pricing at this level is a very data intensive process, particularly for retailer networks conducting transactions in large volumes. The application of machine learning, AI and deep digital capabilities can be very effective in helping convenience retailers translate volumes of data into actionable pricing decisions, effectively optimizing the retailer’s fuel sales strategy. AI and machine learning also serve a predictive function, allowing the convenience retailer (or network of retailers) to better anticipate future pricing scenarios. With predictive analytics and advanced data management, machine learning and AI allow a fuel pricing system to more accurately achieve a desired margin or volume outcome.

When used in conjunction with connected data, AI and machine learning facilitate a level of granularity that can further benefit a convenience retailer’s fuel pricing strategy.  A platform that leverages data from across a convenience retailer’s operation could, for example, identify a particular customer that fills up between the hours of 3-5 p.m. every Friday, allowing the retailer to adjust pricing specific to that customer. Alternatively, a retailer could combine this data with its loyalty program data to manage specific incentives for a certain customer or segment of customers. With the right capabilities, the possibilities are endless.

Fuel pricing is a complex business for convenience retailers. Using advanced digital capabilities like AI and machine learning to get fuel pricing right can have a significant business impact far beyond an operator’s raw margins on gasoline or diesel. The right advertised price of a gallon of fuel can attract a customer who in turn spends on higher-margin, in-store products, as 45 percent of fuel customers will go into the store after fueling up. Conversely, the wrong price can send that customer to a competitor down the road, or, on the other end of the spectrum, undercut any profit margin a retailer might get from that sale. Understanding that intelligent fuel pricing impacts the entire convenience retail operation is the first step toward achieving more profitability across all facets of the business.

Sign up for the free insideBIGDATA newsletter.

Leave a Comment

*

Resource Links: