Artificial intelligence applied to pricing in an uncertain customs environment

To preserve their margins, companies must demonstrate agility and rely on reliable data to understand the impact of increases in customs duties on profitability, and to determine whether costs should be passed on or absorbed. In this context, a strategic approach based on Artificial Intelligence makes it possible to model consumer behaviour in response to price variations and to optimise pricing decisions across different scenarios.

I - a data driven pricing strategy

In the face of customs duty fluctuations, a pricing strategy must be built on fully data driven foundations. Four key objectives form the pillars of an effective response :

• Passing from intuition to analytics : moving away from instinct based decisions or general rules towards rigorous analysis of historical sales data. This transition helps reduce cognitive bias, improve the reliability of trade offs and bring objectivity to pricing decisions.
• Leveraging elasticity with precision : rather than applying a uniform pricing policy, strategies should be differentiated by product, brand or geographic market. By analysing past consumer behaviour, AI enables accurate estimation of price sensitivity and supports more tailored decision making.
• Building agile response capabilities : the ability to instantly assess the impact of different pricing scenarios as soon as customs changes are announced, enabling faster and more optimised decisions.
• Optimising margins and volumes : maximising margins while maintaining control over volumes by identifying consumer tolerance thresholds and critical price points.

AI at the core of price elasticity analysis

Thanks to machine learning algorithms, it is now possible to leverage large volumes of historical sales data to accurately map how consumers respond to price changes. This approach significantly outperforms basic pricing analyses. It inherently integrates a wide range of contextual variables, such as seasonality, promotions, and even perceived value associated with a product or brand, providing a granular understanding of purchasing behaviour that is far closer to market reality.

II - Turning data into decisions : insights and action levers

An AI driven pricing model enables clear answers to the fundamental strategic questions faced by any company dealing with customs duty fluctuations. These challenges can now be addressed through quantified insights, providing essential guidance for optimal decision making.

This approach makes it possible to :

• Understand sensitivity by brand or product line : accurately identify which brands or product lines are most responsive to price variations. This often highlights significant differences within the same category. Being able to capture these variations allows for a differentiated customs pricing strategy : absorbing cost increases on highly sensitive segments to protect volumes, while passing them on, even partially, to more resilient segments.
• Identify high elasticity segments over time : detect segments where a small price increase leads to a disproportionate drop in volumes. These insights are critical to determine which products should undergo price increases, based on their pricing positioning, and to define the appropriate new price. Passing on costs to highly elastic segments may prove counterproductive in terms of revenue and market share, whereas accurately identifying them helps avoid costly mistakes.
• Detect critical psychological thresholds : identify pricing breakpoints where elasticity shifts sharply. A precise understanding of these thresholds enables optimisation of price positioning. Companies can avoid these high resistance zones, or choose to cross them only if a cost benefit analysis clearly demonstrates an advantage, even when factoring in potential volume decline.
• Analyse differentiated effects by channel or geographic market : AI models consistently reveal significant variations in price sensitivity across distribution channels (online or offline) and geographic markets. This highlights that the same customs increase will not be perceived in the same way, nor generate identical customer reactions everywhere. These insights enable precise adjustment of pricing strategies, whether passing on or absorbing costs, based on local specificities and competitive dynamics, rather than applying a uniform approach that would inevitably be suboptimal.

‍

III - use case : pricing strategy in the luxury sector

Optimising pricing strategy in the luxury sector through AI

Eleven supported a leading luxury group in the design of an analytical tool aimed at optimising its pricing strategy. The objective was to accurately assess the impact of price increases on demand in order to maximise revenue. To achieve this, Eleven developed a decision support tool based on Artificial Intelligence, designed to assist pricing teams in their strategic trade offs.

Leveraging a granular analysis of historical sales data, enriched with machine learning algorithms, Eleven built a model capable of simulating various price increase scenarios. The tool enables the anticipation of potential impacts on demand, taking into account factors such as seasonality, market trends and the broader macroeconomic environment.

A specific challenge in the luxury sector

In the luxury sector, modelling price elasticity is particularly complex, as demand can shift rapidly depending on trends, perceived brand positioning or the economic context. It is therefore essential to accurately anticipate the impact of pricing adjustments in order to preserve both competitiveness and profitability. A thorough analysis of past pricing strategies, combined with the identification of critical psychological thresholds, is key to informing future decisions.

A robust analytical approach for informed decision making

To address these challenges, Eleven implemented a rigorous analytical approach aimed at isolating the specific effects of price increases. This methodology made it possible to neutralise the impact of exogenous factors such as product mix effects, new product launches or broader market dynamics. The model developed not only anticipates the impact of price increases on overall revenue, but also forecasts demand transfers across geographic regions, based on regional price differences.

The tool designed now enables teams to steer their pricing strategy on a product by product basis, relying on reliable and contextualised quantitative simulations. It represents a powerful optimisation lever in an increasingly competitive and fast evolving environment.

‍

Conclusion

Elasticity modelling is a key lever for building an agile and resilient pricing strategy. In a context where pricing and customs uncertainties are becoming more frequent, it enables companies to anticipate demand reactions, make informed decisions and protect margins. Price positioning and the ability to anticipate market responses are becoming critical differentiating factors. Integrating this analytical dimension means securing a sustainable competitive advantage.

To learn more about these topics and identify the impacts and opportunities relevant to your organisation, contact our expert : Morand Studer (Managing Partner and AI Expert)

‍