Offering consumers different prices is as old as commerce itself. However, in the age of data, the situation has changed: now, businesses have the tools to determine the maximum price we can accept to pay and charge us exactly that. Our personal data used for such predictions can be extremely broad in scope, including insights from our e-mail history, search history, shopping history, where we live and what we do in life; it can also be based on outdated or wrong information. When more businesses start employing such practices, the consumer will no longer be able to establish the objective value of products and services, as the price shown to each consumer will be different. With the associated risks of welfare loss and of discrimination, as well as the added harm to privacy and autonomy in decision making, algorithmic price personalisation based on information held about the individual consumer is cause for serious concern.

Behavioural price personalisation techniques analyse consumers' personal data to quote them different prices for the same products, typically set at the maximum (predicted) amount an individual can agree to pay. Despite a weak information requirement in the Consumer Rights Directive, such techniques are still highly problematic in terms of reducing consumer welfare, questionable transparency, as well as the risk of unfair discrimination; they are also not accepted by the majority of consumers.

Price personalisation is not compatible with EU laws that protect consumers and their personal data. This includes the General Data Protection Regulation (consent requirements, purpose limitation), the Consumer Rights Directive (pre-contractual information), as well as unfair contract terms and unfair commercial practices law. It also creates a risk of discriminatory practices, where it can be used as a tool for locking 'undesired' consumers out of a service, simply by rendering it too expensive for them to afford.

Against this background, BEUC calls for a general prohibition of pricing techniques using personal data to adjust the price based on behavioural predictions made about the individual, such as:
•    Assessing the consumers' individual willingness to pay,
•    Using profiling to predict the likelihood of switching to a different provider,
•    Filtering the customer base by giving 'undesired' consumers over-inflated prices.

The prohibition should leave out individual and group discounts which are not based on profiling, as well as price differentiation techniques which are:
•    Fully transparent, including about the data and assessments involved, and limited to such data and type of assessment that are strictly necessary and directly relevant to performing the given service, such as insurance risk assessments.

In all cases, the burden of proving compliance must be on the trader.