Individual pricing, or there and back again: What UK businesses can learn from Delta’s AI pricing U-turn

Imagine booking a flight that’s tailor-made just for you: the nearest airport, the flight time, the seat. Even the drink selection is perfect for you.

In most cases, this might seem like a distant dream, especially when you’re crammed into a packed plane during peak summer travel (seriously, have they brought the seats even closer together?), but you might soon find that one aspect of that flight is tailored to you: the price tag. 

This concept is called individualised or personalised pricing; the next step on from dynamic pricing. The idea leapt from the theoretical into the headlines thanks to Delta Air Lines’ recent foray into artificial intelligence-powered revenue management. 

In Delta’s Q2 2025 Earnings Call,  Delta announced that it is optimising revenue by leveraging AI-enhanced pricing solutions. Delta’s initial goal involved deploying AI-based revenue management technology across 20% of its US domestic network by the end of 2025. In the heart of the renewed pricing structure would have been technology that was described as capable of predicting the amount people are willing to pay for premium products.

However, following sharp criticism from the public, including some US senators and multiple consumer advocates, Delta clarified that it has never used, tested, or planned to use AI to set prices based on personal data. Instead, it emphasised that its pricing remains rooted in aggregated data and traditional dynamic pricing methods. 

A glimpse into the future or a warning for businesses?

The Competition Markets Authority (CMA) describes personalised pricing as pricing strategy where a good is sold by the same seller to different buyers at different prices. Sometimes the information is based on aggregated information, but it has the potential to be very specific and based on personal data. 

In the UK, the furore surrounding Oasis ticket sales drew the attention of the CMA to dynamic pricing. Dynamic pricing is where businesses adjust the prices of their products or services based on current market demands, competitor pricing, and other external factors.

This approach often involves raising prices during peak demand periods and lowering them during times of low demand. Although there is no legal definition of dynamic pricing, the CMA defines it as “situations where firms adjust prices rapidly and frequently in response to changing demand conditions”.  

The aim of the AI-based pricing technology is to provide the right price at the right time to the right customer. This means effectively tailoring ticket prices to one’s perceived willingness to pay. As individual pricing could lead to maximisation of unit price, from business perspective, personalised pricing makes a lot of sense.

The perceived value of products and services depends on the person buying it, and gaining the maximum value per unit is essential for profitable business. Using individual’s personal data as a base point seems like a logical place to optimise.

There are examples of aggressive dynamic pricing situations that can be interpreted as bordering on individual pricing, and the CMA is certainly alive to the issues having published top tips for businesses using dynamic pricing this summer. While the guide is specific to dynamic pricing, it is a good starting point for personalised pricing as well. 

Legal implications in the UK

For UK businesses watching this development with interest – or even considering similar strategies – it’s essential to understand not only the reputational impact this decision might have but also the legal landscape before diving in headfirst. It is not only the reputation of the business that might be tested. 

Personalised pricing is not inherently unlawful in the UK. However, its implementation must be carefully managed to avoid breaching data protection, competition and consumer protection laws. Under the new consumer protection laws, the CMA now has the power to seek direct enforcement against, and potentially fine companies engaging in unfair consumer practices without needed to go to court.

Under the UK GDPR and the new Data (Use and Access) Act 2025, businesses must have a lawful basis for processing personal data used to determine pricing. If pricing decisions are based on profiling or automated decision-making, additional safeguards apply.

This includes the right for individuals to obtain human intervention, express their views, and contest decisions. Transparency is key: consumers must be clearly informed about how their data is being used and for what purpose. Also, one of the automated decision-making grounds must apply to the suggested situation. 

The CMA has noted that failing to provide accurate information, or any information at all, about the pricing model a business uses (e.g. dynamic or personalised pricing) could be a misleading action or omission if the average consumer needs that information to take an informed transactional decision.

Practical guidance for businesses: Proceed with caution

Before embracing a personalised pricing approach, check you will be able to:

Be transparent

Ensure consumers understand how prices are determined and what data is used. Make sure that the information is provided in a timely manner and in a way that is truly visible for the consumer. 

Assess fairness and target demography

Avoid targeting vulnerable groups, such as children or using sensitive personal data without robust justification.

Be clear to consumers what pricing model is used and that prices can change

Consider what consumers need to know to make decisions and ensure that prices do not change within the customer journey.

Review your data practices

Ensure compliance with applicable data protection legislation, especially around profiling and automated decision-making and appropriate marketing practices.

Investigate competition law risks

Be alert to potential abuse of market power or anti-competitive effects. The CMA has published reviews on this topic and flagged the potential for personalisation to cause price discrimination.  

Consider unforeseen consequences 

Consider whether your proposed use case could trigger unforeseen consequences such as claims that it is discriminatory. Think about reputational risk, too. Even if legally compliant, personalised pricing may be perceived as manipulative or unfair. And of course, seek legal advice.

Conclusion: Navigating risk of personalised pricing

Personalised pricing may offer commercial advantages, but it’s not without legal and ethical complexity. This is also an area of law we expect to evolve quickly over the next few years.

It’s really important to keep abreast of what the courts, regulators and other standard setters are saying about issues such as transparency and fairness. If you're exploring this space, we’re happy to help you navigate the risks and opportunities with confidence.