Babir Sultan, owner of five Fav Trip stores in the Kansas City, Missouri, area, recently had a publication call his stores a “playground for AI.” In addition to running convenience stores, Sultan is the co-owner of a development company working on AI features relevant to the retail industry.

When approaching AI, his goal is always to ensure that it’s a solution to a problem, something that makes “the shopping experience and the store owner’s life easier, better and allows them to make better decisions by using AI,” he said.

AI experts and vendors emphasize that AI should support your business strategy, not the other way around. “Eighty percent of AI projects fail, and only 30% [of those that succeed] hit their ROI objectives,” said Greg Buzek, founder, president and principal analyst of IHL Group, a global research and advisory firm specializing in technologies for the retail and hospitality industries. “Start with the business strategy first and see where AI might fit into it. Don’t do AI just for AI’s sake.”

But they also emphasized that retailers need to embrace AI sooner rather than later—in fact, soon not using AI will make their job harder, said Tav Tepfer, chief revenue officer at Invent.ai, which specializes in retail inventory optimization solutions. “Your competitors are going to do it,” she said.

While operators need to implement AI to align with their specific business objectives and goals, there are a handful of practical ways any retailer can start using AI today.

Personal Tasks

You have to learn to crawl before you can walk. Sources agreed that the first place you should implement AI is for your own personal or professional use.

“If you’re not comfortable using AI yourself yet, then you’re not ready to talk to a vendor about using AI for your business models—and I say that as an AI vendor,” said Emily Nave, head of product at InStore.ai, a company that captures and analyzes store audio and delivers insights for the retail industry. “There are a lot of everyday practical uses that people don’t even realize they can do. You can get instant value from it.”

She suggested using free tools like ChatGPT or Claude, for example, to help brainstorm, work through decision-making processes by asking it questions, use as a search engine to get enhanced results or answers to questions, or to help jumpstart tasks like drafting emails or job descriptions. “It’s great at starting things up for you,” she said.

“Before a retailer even thinks about optimizing their business model with AI, start by applying it to some of the simple tasks you do every day,” added Wesley Bean, president and chief operating officer at Engage3, which uses AI modeling for retail price optimization. “We spend so much of our time in meetings. How are you applying AI to help your decision-making process, whether it’s something simple like note taking, summarization or action item planning? Those are a great entry point to find efficiency.”

Buzek also said that for independent or smaller operators, AI can help with “getting knowledge about the business out of the head of the owner so it can be accessed by employees at any time.” He specifically cited using AI to create employee handbooks or document operating procedures. “Google’s NotebookLM is perfect for that. You load all those documents into it, and it becomes instantly queryable. Employees can just ask, ‘How do I deal with X.’”

Pricing Strategy

Optimizing pricing in a thin margin business is always a key focus for retailers, and it will continue to become more important as tariffs shake up the supply chain and the costs of goods and packaging.

“Right now everyone needs to be more nimble in terms of being competitive on price for both fuel and inside the store. So speed of pricing becomes of heightened importance,” said Bean.

He said AI helps by normalizing pricing quicker and faster than can be done manually, as well as by modeling scenarios as to how price changes can impact other areas of the business, from sales and profits across categories to topline numbers.

“You have all these different inputs across your whole portfolio and 10,000 different decisions that need to be optimized together. Management of that complexity requires more sophisticated decision-based tooling embedded in the science of using AI to evaluate things,” Bean said. “It can look at individual categories and tell you where you might be able to be more elastic on pricing because the consumer is less sensitive, or where you can’t because the customer is wildly sensitive.”

Tepfer said AI can help by determining the right choice of three common scenarios for pricing that help retailers optimize their strategy—“dynamic pricing, which can tell you to increase or lower your price; promotional pricing, or promoting a product to get that bump in revenue you want; and markdowns to clear out inventory so you don’t lose too much money on it. But all of them need to have price elasticity, and you need to understand how much inventory you have to be really optimal,” she explained.

The same techniques can be applied to fuel, which Bean said is continuing to become more dynamic due to tariffs, oil prices and the global economy. “It’s putting retailers in a position where if you’re off by even a few hours, you’re getting the price wrong and might lose out to competition in your local market. We’re seeing multiple price points move throughout the day and retailers need to capture those to control margin and provide competitive prices for their customers. You just can’t afford to be lagging in terms of being priced right, but that’s where the AI technology can help and help retailers optimize against all the data points they have.”

Inventory Management

For Tepfer, inventory management is an obvious use case for AI. “Retailers are constantly making decisions about what inventory to carry and how much, and getting that right every time can be hard. Inventory managers have thousands of different products to track, which they’re probably logging in spreadsheets, and millions of decisions to make, which they might be doing based on gut instinct or general rules,” she said. “It’s a difficult job and forecasting models alone don’t always make things better.”

What AI can do, she said, is take all of those decisions and dynamics into account and make a decision about what retailers should carry based on trends, seasonality, lead times, potential profit and other factors. “It takes actual data into account instead of relying on instinct for manual decisions. You want the data to tell you how to decrease your assortment or increase your assortment, for example, especially as tariffs come into play and there’s not as much money to go around,” she said.

She also noted that AI can help mitigate “phantom inventory.” Phantom inventory can happen for several reasons, she said, but it problematically reduces sales. “AI can send the store manager a report that says, ‘Hey, these are all things that are high on phantom inventory. Go check,’ or other alerts that force the problem to get fixed.”

As part of inventory management, supply chain optimization with AI can also help retailers manage costs, said Buzek. “Most of the trucks that are on the road right now are empty and aren’t picking things up along optimized routes—the ability to control that lowers cost and increases efficiency. Walmart has been dramatically successful in fixing that.”

Data and Customer Personalization

Tracking first-party customer data in a brick-and-mortar environment is notoriously difficult. “It’s one of the biggest limitations that convenience stores have,” said Buzek. But as more opportunities arise that require customer data to pull off—retail media networks, loyalty programs or digital coupons are all examples—AI can help turn physical foot traffic into real customer data retailers can use. “It’s really low-hanging fruit for AI,” added Buzek.

Sultan said he has used AI to recognize and track vehicles or capture customer demographic information through store cameras. He said he recently reached out to a company about a partnership, and it wanted to know how much foot traffic his stores had. Prior to using AI he would have had to ballpark manually or offer other metrics such as transaction counts, “but I just sent them a screenshot of what we’ve been tracking—how many customers come inside the store compared to how many cars are outside, how many customers we have, age groups and demographic info. Capturing this data makes us better partners and more attractive to vendors.”

It’s critical for marketing and retail media as well. “Imagine running a banner ad and being able to see exactly how many people actually looked at the ad, when they saw it and have information about them,” Sultan said. He is also looking at AI solutions for the future that will recognize license plates and connect cars to transactions, allowing him to know what specific customers purchase and how to target promotions or specific menu items to them. “We are looking at these kind of futuristic projects that will help us connect all the dots of our business,” he said.

Loss Prevention

With loss prevention and theft top of mind for retailers, Sultan recently started using AI to mitigate shoplifting. He’s currently testing a “person of interest” tool that helps stores identify banned customers. “We have humans watching 24/7, but a step up from that is utilizing AI to detect people. For example, if someone shoplifts in the morning and we ban them from the store, that shift worker is only around for so long. The shoplifter ends up coming back later in the day to repeat the behavior with new employees.”

If the AI system recognizes a banned customer has entered the store, it will alert the employees.

“That’s pretty exciting for us, because humans obviously make mistakes or miss things. Even our live security goes through different shifts where there are different people watching. They’re not going to be able to tell who’s who, but the AI doesn’t miss a beat,” Sultan said.

Employee Engagement and Retention

The best use case for AI in Nave’s opinion? “100% it is getting more out of your best employees.”

Retailers can use AI to increase retention, reduce turnover and foster employee engagement in a handful of ways.

First, reducing someone’s time spent on mundane or routine tasks essentially allows employers to scale them by freeing up their time for more important or quality activities. “I have employees where I joke, ‘Oh, I wish I had more of you.’ Now I kind of do because it takes them so much less time to do certain things,” said Nave. “Find those employees who have true skill and knowledge of the business and automate time-consuming parts of their job to free them up to do what really needs to be done.”

Nave also noted that giving employees, including frontline workers and store-level employees, training for and access to AI will be “a huge career opportunity for them. It’s a way for them to level up their resume, increase their career trajectory and get experience in tech when they might not have otherwise. That’s a benefit to them and in terms of attrition, this type of experience will get the right people to stay for longer and grow within your company.”

At Fav Trip, Babir uses AI to help evaluate the level of customer service his employees are offering. “We were able to have AI identify whether an employee was ‘present’ when customers came in, and we can see if its due to absenteeism, understaffing or someone doing multiple tasks,” he said. “We have been able to address it and reduce those incidents, which helps with customer satisfaction as well.”

Using AI to recognize frontline employees and reward them for a job well done will also foster a positive company culture. “You can always find someone who is doing something wrong. But it’s hard to know about those little moments where an employee did something great if you’re not in the store all the time,” said Nave.

She said she has seen clients use InStore.ai’s technology to reward cashiers who did something outstanding for a customer when no one was watching—like a cashier who got a call from an elderly women who had run out of gas near the store, so he walked with a canister of gas to top her off and even drove her back to the station to fill up.

“That’s such a helpful and amazing thing the manager might not have ever heard about if it didn’t show up in the logs,” she said. “Being able to say, ‘Hey, look at what we heard you did, that was really great and the level of customer service we want,’ and then recognizing that employee—it makes them feel great about their job.”