Data-Driven Decision Making: Use AI to Improve Pricing Decisions

Pricing can look smart in the data and still feel wrong to the customer. 

In December 2025, Reuters reported that Instacart ended AI-driven price tests after criticism that shoppers were seeing different prices for the same grocery items. Some shoppers saw prices up to 23% higher than others, according to the report. 

For a CFO, that is the caution. A pricing recommendation may optimize for demand, capacity, or margin and still create a problem for the customer relationship. AI can frame the decision, but people need to own the judgment behind it. 

For growing brick-and-mortar businesses, the opportunity is more practical than personalized pricing. A business may have list prices, past transaction data, occupancy or utilization data, seasonality, and local demand signals. Those inputs often sit in different places, which makes it harder for leaders to see where price is lagging demand or where capacity may justify a different approach.

As a TechCXO fractional CFO, I see pricing as one of the clearest places where data-driven decision making can improve the economics of a multi-location business. The goal is a stronger starting point for the pricing conversation, not a black-box answer. 

Use Demand Data to Frame the Pricing Conversation 

Pricing across a multi-unit business depends on more than a list price. Local demand, available capacity, seasonality, and the prices customers have accepted all shape what a location can realistically charge. As the business grows, leaders need a more consistent view of those conditions.

I recently built an AI-supported tool that brings those inputs together. It looks at the list price, the last price we actually achieved, how busy a location is, and the time of year. The output gives the team a stronger starting point for deciding what terms may make sense for a customer and for the business. It helps frame the pricing conversation, but it doesn’t replace the judgment behind the final decision. 

Pricing decision support has been available to large real estate organizations for years, but AI now makes a lighter version more accessible to smaller multi-unit businesses. It’s not a question of losing a once critical job. It’s enabling something at the brick-and-mortar level that was never done before.

“Leaders can use AI to see where pricing is lagging demand, where capacity may justify a different approach, and where a location’s results deserve a closer look.”

Pricing still requires leadership guardrails. A business has to consider customer trust, brand position, local competition, fairness, contract commitments, and the risk of changing prices faster than customers can understand. Recent scrutiny of retail price experiments shows how quickly trust can erode when changes feel opaque or inconsistent. AI can support the pricing decision, but the workflow needs defined business rules, human review, and a defensible reason for the recommendation. 

The measures should reflect both sides of the decision. Track realized price alongside utilization, occupancy, and location-level revenue. A higher price that leaves capacity unused or weakens retention is not improving the broader economics of the location. 

Keep Pricing Decisions Defensible 

A pricing workflow should make the business more disciplined. It should help leaders see what has changed, compare locations more consistently, and understand which recommendation deserves a closer look. 

The FTC’s surveillance pricing study raises a broader concern about companies using personal data, behavior, location, timing, or shopping channels to set individualized prices or discounts. Most smaller multi-unit businesses are not building anything that complex. The caution still applies. Pricing can damage trust when customers cannot understand why they are being charged differently. 

For me, the value of AI-supported pricing is a better starting point. A good workflow brings demand, capacity, seasonality, and pricing history into view so leaders can make a more informed decision. The final call still belongs with the person who understands the customer, the market, and the economics of the location.

Ready to make AI-supported pricing more useful and defensible? TechCXO’s AI services help companies define practical use cases, prepare the data, and build workflows that turn analysis into better business decisions.