Blog: Pricing & AI—Retail Analytics and Strategies
The newspaper of the pricing
The reference price is the lowest price charged in the 30 days prior to a price reduction.
Any advertisement for a discount must display the original price, and following a September 2024 ruling by the Court of Justice of the European Union, the discount percentage must be calculated based on that price.
A temporary price drop therefore becomes the benchmark for one month. Successive price reductions and perishable products are subject to specific rules. Applying the rule requires a dated price history, organized by product ( product by product ) and by channel.
Price compliance refers to the legal rules that a price and its display must follow.
In France, four rules apply: the reference price for discount advertisements, the prohibition on reselling at a loss, the EGalim promotional caps, and price disclosure requirements.
These rules are simple to state but difficult to apply to thousands of products and multiple channels. Most discrepancies stem from incomplete data at the time the price is set, not from any intention to mislead.
Selecting and implementing a pricing tool is a four-step process: defining the need, choosing the solution, demonstrating its value in a real-world setting, and then implementing it with the data and teams.
Projects that fall short rarely choose the wrong software. Instead, they define the problem poorly from the start, overgeneralize without restraint, or underestimate the data and the teams’ buy-in down the line.
A pricing analysis is an audit of a company’s pricing policy based on its actual data: sales, margins, pricing schedules, promotions, and competitors’ prices. It examines the consistency of prices, market positioning, price image, and promotions.
It identifies which discrepancies warrant action, along with the expected return, and organizes them into a detailed roadmap. An average “ Price Index ” is often misleading: when weighted by sales, it shows that just a few dozen benchmark products determine the price image.
A pricing proof of concept (POC) does not require perfect data, but rather data that is appropriate for what it needs to demonstrate, with known limitations. The product catalog, sales data, prices and promotions, purchase costs, and inventory levels form the foundation; the granularity and depth of this data depend on the test’s objective.
Before calculating a profit, the recalculated revenue, volumes, and margin must match those in management accounting. A known and documented error does not skew the comparison; a hidden error does.
A pricing POC is a limited test, conducted using a retailer’s actual data, that compares a solution’s pricing decisions to a baseline scenario measured in the same way. It proves nothing if the scope, baseline, metrics, duration, and resources are not established beforehand.
Three categories of metrics should not be confused: model performance, economic gains, and team usage. And the gain measured within a selected scope is a ceiling, not an average: it should be extrapolated with caution, based on documented assumptions.
A SaaS pricing management platform centralizes a retailer’s prices, rules, competitor prices, and forecasts in a browser-based tool. In the retail sector, it delivers responsiveness, measurable margin gains, native compliance with French regulations, predictable costs, and scalability.
Its limitations are just as clear: it does not correct poorly matched data, outdated rules, or an organization lacking leadership. Booper, a French software company founded in 2014, manages more than 4,000 retail locations and aims to increase margins by 0.5 to 3 percentage points within a few months.
A franchisor cannot impose a minimum resale price: doing so is punishable by law; the franchisor may only recommend a price or set a maximum. A franchisee’s margin is therefore governed by a documented target, defined by network segment and applied uniformly across each segment.
The best indicator of this governance is not adherence to the suggested retail price, but the rate of compliance with it by segment. A suggested retail price that franchisees in a given area do not follow often signals a poorly set price, not a lack of discipline.

A reversibility plan describes what a customer recovers when leaving their service provider, in what format, within what timeframe, and at what cost. For a pricing solution, four assets are involved: raw data, rules, arbitrage history, and the product matching table. Generally, only the first of these is covered, while the other three must be reconstructed manually, a process that takes months.
Reversibility is negotiated before signing, based on three principles: specifying the assets to be returned, requiring a format that is readable without the tool, and quantifying the timeline, cost, and parallel operation. The goal is not to leave, but to have the option to do so.

Pricing decisions account for only a small portion of a pricing team’s week. The rest is spent on preparation, coordination, formatting, justification, and corrections. This work is real, and it takes time away from what creates value.
The order of correction is: eliminate, then automate, then delegate. A tool first reduces the amount of preparation required; coordination and justification only improve if decision-making processes also change.

The model is the easy part of a pricing tool. What takes years is the product catalog, product matching with competitors, rules, explainability, and compliance—some of which have expiration dates. Across all tools, 71% of in-house developments end up being abandoned, according to Exclaimer (2025).
The hidden cost is that of continuity: an internal tool relies on two or three people, and it is its success that creates dependency. The build process remains meaningful within a narrow and stable scope, as a complement to a platform. The deciding factor is not the ability to build, but the ability to maintain it for ten years.

The data describes what happened; the context explains what is not done here. This context is implicit: price benchmarks, supplier constraints, unwritten rules, and category-specific guidelines. No search engine can guess it, and no general-purpose AI can retain it from one session to the next.
The least expensive way to formalize it is through rejection: every recommendation that is rejected, along with the reason given, reveals a piece of it. Formalizing it has lasting value, because the context outlives the models and remains a company asset even when the tool changes.

The accumulation of rules is a cumulative process, not an isolated incident: it grows with every legitimate decision made in isolation. Five forms predominate: the exception that has become permanent, the seasonal rule that is never suspended, contradictory rules, the orphaned rule, and the redundant rule.
Before it costs you profit margin, it costs you trust: a team that no longer understands a price stops following the lead. The solution is a short cycle rather than a major overhaul, and setting an end date as soon as each rule is created.

A business case for a pricing solution combines two types of return on investment. Friction ROI refers to the hours saved: easy to measure, but capped by the size of the team. Decision ROI refers to the margin points gained: slower to quantify, but with no upper limit.
Margin gains are measured against a control group, never by comparing before-and-after figures. Before the results are clear, the percentage of recommendations that are actually implemented indicates whether those gains will materialize. The test every business case must pass: without the time savings, is the project still profitable?
Optimix merged with Maxxing in July 2025, and PricingHUB was acquired by Akeneo in June 2026. For their customers, the question of the independence of the pricing roadmap has become a legitimate concern.
What sets a query apart in France isn't the processing speed: it's the native support for French rules, pricing by store or by zone, the validation process, and testing with your own data.