Pricing by Objectives: How AI Learns to Set the Right Price
September 29, 2026
Goal-based pricing starts with the target outcome (margin, revenue, category competitiveness) and lets AI calculate the prices needed to achieve it. A goal is only meaningful when defined along with its constraints—for example, achieving a certain margin without damaging the brand’s price image.
When the target is out of reach, the engine flags it and quantifies the trade-off: this is often the most useful response for management. The pricing manager retains the final decision.
For years, setting a price meant applying a series of rules product by product and then looking back to see if the margin had held up. With inflation and customers who can compare prices in a matter of seconds, this method has reached its limits.
AI now makes it possible to start with the desired outcome and work backward to determine the prices needed to achieve it. Here’s what this shift changes, based on twelve years of pricing projects conducted at Booper.

Goal-Based Pricing: What Is It?
Goal-based pricing is a pricing method that starts with the desired outcome (margin, revenue, competitiveness) and uses AI to calculate the prices needed to achieve it, rather than relying on a series of rules product by product.
This is the AI-driven version of target-based pricing. The difference is significant: target-based pricing sets a target level to be maintained, while goal-based pricing calculates the path to reaching that level, product by product.
In the past, we would set rules and then assess their impact on the margin at the end of the period. Today, we can approach the problem from the other direction and determine, before setting a price, what is achievable.
A lot of data, few effective pricing decisions
Contrary to popular belief, data is no longer the problem. Receipts, sales history, inventory, purchase prices, competitors’ prices, weather—retailers have plenty of it.
What's missing is a long-term pricing strategy—and often a dedicated pricing function. The result: prices set category by category, depending on what needs to be addressed immediately.
AI makes it possible to move from reactive pricing to predictive pricing: anticipating sales, running simulations, and making decisions before implementing them. We discussed this in a public webinar co-hosted with Booper.
Setting the Right Price with AI: Start with the Goal
| Dimension | Rule-Based Pricing | Pricing by Objectives |
|---|---|---|
| Starting Point | One rule per product or product family. | A specific business target to be achieved. |
| View | The rearview mirror: you measure after the fact. | Forecasting: We run simulations before making a decision. |
| Scope | Each price is optimized individually. | The performance of a department, a store, or the network. |
| If the goal is out of reach | We find this out at the end of the period. | The best compromise is identified early on. |
Based on the objective, the engine simulates thousands of scenarios. It takes into account business constraints (minimum margin, psychological pricing thresholds, consistency across formats), store-specific purchasing behaviors, and the retailer’s specific rules.
If it finds paths that reach the goal, it suggests them. If it doesn't find any, it identifies the best possible compromise and shows what's holding it back.
profit margin increase in just a few months for retailers supported by Booper, according to results observed in its projects.
(Booper figures presented in *Informations Entreprise*, September 15, 2026)
A goal is never without its challenges
This is the first point we emphasize to management: a single objective is a trap. If you ask a system to maximize profit margins, it will raise the prices of less price-sensitive products to the point of damaging the brand’s price image.
A useful objective always comes with a limit: some leeway, but without negatively impacting the benchmark index (KVI). Volume, but without falling below a minimum threshold. It is this constraint that transforms a calculation into a strategy.
The case of the Barbotteau Group, a retail company in the French Caribbean, illustrates this point well. The group used a single " Markup " for the entire store.
From now on, each category is assigned a role (traffic, revenue, or margin), and the engine calculates three optimal prices—one for each objective.
When tested on a selection of products, the approach resulted in a 1-point increase in the " Markup " score for the products tested over the course of a quarter.
These results were presented by the group's director of pricing during the webinar mentioned above.
Customer complaints about prices being too high have disappeared, and competitors have followed suit by raising their prices. The margin target was met without compromising the price image, because the two were addressed together.
When the Goal Is Out of Reach
A counterintuitive insight from the field: the most useful response from a search engine isn't always "here are the prices." It's often "this goal isn't achievable, and here's why."
The team sets a +1-point margin on the shelf, without negatively impacting the price index of benchmark products compared to the main competitor.
The engine responds: under these constraints, the maximum achievable is +0.6 points. A full point can only be achieved by accepting a slightly less favorable rating on about ten highly visible products.
The nature of the discussion is changing. It’s no longer a matter of “the engine can’t handle it”; it’s an explicit trade-off between profit margin and price image— one that only management can decide.
This second approach avoids promising the executive committee a result that the market will not allow. For the simulation process itself, see our method for testing a price before launching it.
Anticipate cannibalization rather than optimizing a single metric
A price cut can boost sales of one product while hurting two similar items: sales of that product rise, but the department's profit margin declines.
That is why our models incorporate intra-category cannibalization, as measured by cross-price elasticity. We optimize an entire department or store, not a single price.
In practice, that's what sets apart a tool that boosts sales from one that improves profitability.
AI and Price Execution: What's Changing for Teams
Pricing teams spent a large portion of their time creating simulations and then analyzing them manually.
Once the simulation and analysis have been implemented on an industrial scale, retailers working with Booper see a 70% to 90% reduction in the time required to set prices.
The role is shifting toward analysis and decision-making. Decisions are made collaboratively by teams from pricing, procurement, and category management, with clear approval processes: this is the collective governance of pricing decisions.
We call this step “price execution”: moving from analysis to action more quickly, in a streamlined, systematic way.
The logical next step—agents adjusting prices within certain constraints—is described in our article on agent-based pricing.
How this plays out in practice
Booper was founded in 2014 with a simple belief: pricing deserves the same level of analytical rigor as supply chain management or marketing.
Twelve years later, the team consists of about 30 retail and pricing specialists, including four PhDs in AI and data science. More than 4,000 retail locations and €44 billion in revenue are managed using its tools, both in France and internationally.
In BOOPER MPS, you set the objective (margin, revenue, competitiveness by category) and its constraints. The engine simulates scenarios, factors in cannibalization, and presents possible trajectories along with their rationale.
No prices are applied without validation through your approval processes. Details are available on the modular pricing platform.
To see how these principles fit into a comprehensive approach—from analysis to in-store implementation—check out our guide to retail pricing optimization.
See also in this feature
- The Booper Solution: Pricing Strategy Consulting
- Strategic Pricing: Definition and a Comprehensive 7-Step Method
- Pricing Analysis: From Price Audits to a Costed Roadmap
- Revenue Growth Management (RGM): Definition and Drivers for the Retail Industry
- Pricing KPIs: Essential Metrics for Building Your Dashboard
Frequently Asked Questions
What is goal-based pricing?
It involves starting with the desired outcome (margin, revenue, competitiveness within a category) rather than rules product by product.
An AI engine simulates pricing scenarios within the retailer's constraints and suggests those that meet the objective or offer the best compromise.
How does AI help determine the right price?
It assesses how each product reacts to price changes, including the effect on related products, and then identifies the pricing combination that best meets the established objective.
The right price is the one that meets the objective without violating the constraints—neither the lowest nor the highest.
What happens if the goal set is out of reach?
The model indicates the maximum achievable under the given constraints and what is holding things back. The manager then decides whether to revise the objective or relax a constraint, knowing the cost of each option.
Are goal-based pricing and predictive pricing the same thing?
They go hand in hand. Predictive pricing anticipates the effect of a price before applying it. Goal-based pricing uses these predictions to work backward from the goal to the prices.
How does AI account for cannibalization?
The models estimate how a price change affects sales of comparable products. The engine then optimizes the entire category, which prevents gaining volume on one product at the expense of the department's margin.
How much does an AI pricing solution cost for a retailer?
The cost depends primarily on the scope of the project: the number of SKUs and stores, the modules selected, and data integration. It is evaluated against the expected return, which Booper estimates at an increase of 0.5 to 3 percentage points in margin within a few months for its clients.
Will AI replace pricing teams?
No. It automates the generation of simulations and a large part of the analysis. The pricing manager sets the objectives and constraints, decides between scenarios, and retains the final approval.
Sources: Booper key figures presented in “When AI Learns to Set the Right Price,” *Informations Entreprise*, September 15, 2026; Booper × Groupe Barbotteau public webinar, Pricing Club, 2026. Updated: October 2026.
AI handles the volume at every stage of the data journey: collection, product matching, verification, forecasting, and simulation. The expert selects competitors, resolves ambiguous cases, provides context, formulates the strategy, and makes decisions.
At Booper, Booper Mind makes suggestions, MPS monitors, and the user decides. Knowing how to interpret elasticity and question a forecast is more important than knowing how to code.
Dynamic pricing is permitted in France. Personalized pricing based on automated decisions is allowed, but customers must be notified of it. Algorithm-based price-fixing remains prohibited, even in the absence of contact between competitors.
The strongest safeguard for a retailer: being able to say, for each price, which rule generated it, what data it was based on, and who approved it.
An AI agent does not simply look at a listed price: it compares the actual price—which includes applicable promotions, shipping, availability, and unit price—for the products the customer requests.
Competitiveness is now measured rather than perceived. The accuracy and consistency of pricing data across channels have become the brand's new calling card.
