Information goods such as ChatGPT or Spotify are emerging nowadays. Pricing them represents a key strategic challenge for one main reason. In contrast with traditional goods, they have a near to zero marginal cost and different perceived values according to customers (individually/segment). There are multiple pricing strategies for information goods such as versioning, bundling or group pricing but they are limited because mainly static and they do not take into consideration the individual willingness to pay. But how is the use of AI going to impact pricing strategies of information goods?
To illustrate this question, let’s talk about the Amazon case. Amazon uses algorithms that adjust prices in real time settings. Around 2.5 million prices are modified every day, which means an average price that changes every 10 minutes (Mattes, 2023). Pricing algorithms consider multiple factors such as competitors’ prices, supply and demand through estimated demand and inventory levels or timing through time of the day/year or special events. Since many sellers on Amazon use those algorithmic pricing tools, some interactions can lead to tacit algorithmic collusions such as a collective price increase. Even if this system increased profitability, it also comes with drawbacks. Indeed, even if it tends to decrease when customers get used to it, frequent fluctuations can undermine trust, and this is amplified when prices are perceived to be unfair or arbitrary. It also presents regulatory challenges to determine how those algorithms must be regulated. Moreover, the main challenge it faces is this strategic dilemma of how far personalisation should go without alienating users.
This leaves us room for reflecting: are pricing algorithms ethical and where should we draw the line? And how should platforms balance personalisation and fairness when pricing information goods?
Mattes, Thomas (2023). Algorithmic price administration: How Amazon hijacks competition through automation. Berkeley Technology Law Journal, 37(4), 1179–1240. https://btlj.org/wp-content/uploads/2023/08/0008-37-4-Mattes.pdf
