Dynamic pricing’s AI revolution is here, but can ethics exist alongside profits?

Last year, the media reported a huge amount of backlash from fans over Oasis concert ticket sales as volatile price fluctuations left many feeling outraged and unable to afford to see the band.

This led the UK’s Competition and Markets Authority (CMA) to launch an investigation into Ticketmaster’s ‘dynamic pricing’ practices, which were used to calculate and put in place the varying ticket prices. Since this incident, despite being an industry staple for years, the term ‘dynamic pricing’ has mostly been spoken about with a negative sentiment.

And yet, dynamic pricing has been used by companies for years as a legitimate tactic to price based on demand. Unfortunately, now that AI is involved, the situation is getting trickier for channel partners and those they work with to navigate, creating new capabilities but also coming with an existing narrative of mistrust. Dynamic pricing is a normal business strategy, but what is needed is clarification between the definitions of  ethical and unethical pricing strategies, such as the difference between dynamic pricing and fake discounts, among others. 

Many are asking themselves: Is it possible for organizations to earn profit from AI-driven dynamic pricing strategies while remaining ethical? To answer this question, we need to look at the intricacies of consumer data and predictive analytics. 

Defining ‘dynamic pricing’

Dynamic pricing is the strategy used to adjust prices according to external factors such as classic supply-demand fluctuations, special events, weather forecasts, or competitor actions.

If you were to ask the public, they would say that dynamic pricing is largely associated with surge pricing, when businesses suddenly increase prices to match demand. A clear example of surge pricing is in the mobility sector, when providers such as Uber and Bolt increase prices several times during peak hours, bad weather conditions, and if there’s a lack of drivers. Despite surge pricing being only one of the dynamic pricing tactics, it’s the one that makes the headlines and so remains in the minds of consumers. 

The aim of making frequent price changes is to offer the most attractive price to the consumer and hit the optimal price point for them to buy. As a result, more often than not, we find dynamic pricing drives prices lower, not higher, to enhance retailers’ competitiveness. However, consumers may not be as aware of this, because it is not as obvious when a price lowers, but it’ll stick in mind when it rises.

A great example of this is how e-commerce outlets can optimize their digital shelves with dynamic pricing strategies. Demand forecasting can help to reduce overstocking, which positively affects both business costs and sustainability.

Prices can be adjusted depending on sell-by dates as well, which means the consumers feel a positive effect. Companies that manage to implement inventory management based on demand forecasting find that their inventory levels improve by as much as 35%, according to research by McKinsey.

Despite this, to increase sales without losing profit margins, it’s important that businesses carefully balance pricing up and down, ensuring that they stay somewhere slightly below the product’s average market price. Hitting the optimal price point is a difficult task, so it comes as no surprise that businesses have welcomed and onboarded AI assistance with enthusiasm. 

AI is changing sales as we know it

Many believe that Amazon changes product prices as much as 2.5 million times a day (about once every 10 minutes). It goes without saying that doing this manually would be impossible, as the level of scalability and precision is massive for an individual, but light work for AI. 

In the context of dynamic pricing, AI systems help businesses to analyze vast datasets to predict and understand customer motivations, supply-demand dynamics, and calculate the optimal product value at speed.

AI can examine thousands or even millions of data points in real time. Therefore, if it’s properly integrated with upstream and downstream data systems, AI tools can automatically set and fluctuate the prices, providing businesses with unmatched price elasticity. This can benefit the customer too, for example, if there are signals that a customer is willing to abandon their cart, the system might re-engage them by offering a one-off discount code. 

Although leaving the algorithm alone to iterate through billions of potential scenarios and counting the increased profitability might sound like a dream come true for many organizations, Ticketmaster’s story is a tale of caution. This isn’t a tool that comes without potential negatives if an organization doesn’t put the proper guardrails on.

AI limitations and their consequences

There are many examples of unethical dynamic pricing tactics — from skyrocketing prices for toilet paper, face masks, and sanitizers during the COVID-19 pandemic to irrational fees for a taxi ride during extreme weather conditions.

An AI model that’s trained to function on a set of primitive rules based on supply-demand fluctuations won’t be able to apply regulatory compliance or ethics to its decision-making. Unfortunately, it will simply maximize its function, increasing sales and profit margins, despite the negative social consequences. 

In order to ensure that this new technology properly navigates legal and ethical questions, a sophisticated AI system is required, which is costly to develop. First, it needs a sophisticated algorithmic architecture. Second, the developer has to consider what data is being plugged in. 

Although machine learning models that power pricing systems can be trained on internal company data, such as historical sales, a lot of training data is also collected from public internet repositories using web data collection solutions. This can range from market and competitors’ data to public consumer reviews and even social media trends. Limited data availability, or limited usefulness of the available data, is an ongoing issue for many businesses. A lack of multifaceted or accurate data can lead to broken decision-making or incorrect insights.

However, more data isn’t necessarily the answer, as it can sometimes bring even more trouble. Issues may arise if an AI system is using very personal data for granular segmentation, as it is balancing on the edge of ethics and risking violating data privacy laws or simply leading to biased pricing for different genders (think the ‘pink tax’) and social groups. 

With regulation posing a threat on one side and global competition pressing in from another, the pressure is on for businesses to adopt new practices that could help navigate different ethical issues, while still increasing profitability and operational efficiency with the help of emerging data technologies and AI. 

Keeping ethics at the forefront of pricing elasticity

Before businesses utilize AI-driven dynamic pricing solutions, they must carefully consider the following: 

• Data biases that could result in unfair pricing for some customer groups
• The quality and accuracy of the data that the AI system is trained on
• The nature of products or services they are selling - could increased pricing lead to negative social consequences?
• The transparency of both communication and marketing to consumers about the use of this method.

It is also essential to make sure that AI operates within legal limits, while still maximizing profits. The AI itself isn’t expected to interpret complex regulatory environments surrounding retail and pricing. However, it must be bound by clear rules, such as caps on specific price increases, especially when it comes to essential products. This will help to avoid the controversies we saw with the skyrocketing price of face masks and hand sanitizer.

Finally, businesses should aim for transparency, especially in the case of extreme price fluctuations. If prices change frequently and without a clear reason or notification, it could result in distrust and negative perceptions from previously loyal customers. Although manipulative pricing strategies could bring large profitability in the short term, in the long run, they may usually lead to significant, painful reputational and financial costs. 

Where next?

Dynamic pricing is rooted in creating the perfect incentive to buy by offering the most attractive price to the consumer. Because of this, dynamic pricing triggers healthy competition, as the most attractive price is usually the lowest one. When paired with AI systems, advanced web intelligence technologies enable businesses to develop creative pricing strategies that were implausible in the past. 

However, because of opportunities opened by AI, dynamic pricing has started favoring very granular user segmentation and behavioral techniques, bringing to the fore complicated legal, regulatory, and ethical challenges. Businesses abandoning demand-based pricing strategies is not likely, but moving towards more governmental regulation is looking unavoidable. 

While this is playing out, it is important to remember that the overall health of the economic and legal environment plays a significant role in the effect of dynamic pricing.

If the competition is lacking and the environment is expressing oligopolistic or monopolistic tendencies, then dynamic pricing strategies will only enhance the dominant positions of big industry players. Despite its bad reputation currently, it is worth remembering that in a free competitive market, dynamic pricing will tend to push prices down and strengthen the overall position of the consumer.