How are algorithms for hotel room pricing developed

Dynamic Pricing - Who Will Benefit?

In September 2016, consumer researchers at the Heinrich Heine University in Düsseldorf came together for a forum of the consumer research network. The aim of this forum was to jointly develop an agenda for the implications of dynamic pricing for consumer research and consumer policy. In the time talk we document the lectures from the perspective of retail, consumer research, neuroscience, law and consumer policy.

Dynamic prices: a win for retailers and consumers

Stefan Genth

Dynamic prices are as old as trade itself. One should think of market trade, which has always lived in a special way from the dynamics of its prices, for example because price differentiation over time makes sense for fresh products. It is also known to everyone that regular customers may receive a particularly “good” price, as well as those who negotiate skillfully or the customer groups who benefit from coupons, promotions or free gifts. In contrast, “fixed” prices, even those that are agreed without negotiation between seller and buyer, are a relatively new phenomenon that we only observed increasingly in the 20th century in the context of new sales formats. Price differentiation plays an important role in connection with the price policy instruments used by retailers. Prices are differentiated spatially (e.g. in certain sales areas), temporally (e.g. in campaign periods) or individually (e.g. for individual customer groups). This gives companies the opportunity to react to changes in the market situation or to provide targeted purchase impulses.

In the course of the development of online trade, but increasingly also in stationary trade, by means of technical possibilities such as electronic price labels, prices are becoming increasingly dynamic. Today, companies have access to extensive information via appropriate systems and traditional market research. These make it possible to adapt sales prices to the price behavior of competitors or the ordering behavior of customers at short notice.

The advantages are apparent. Retailers can directly influence sales of certain products via prices that are adjusted over time. Everyone knows the advertising activities through which the retail trade offers parts of the range or individual articles for a limited time at a particularly low price. With such a temporal differentiation, the individual willingness to pay of the customers is not taken into account. Retailers are also lowering prices here for customers whose demand would have remained high at the original price. This is where retail strategies come in to differentiate individually, i.e. to offer individual customer groups identical products at different prices.

Information on the demand behavior of certain customer groups can form the basis for appropriate price-setting decisions. The demand of individual target groups can be stimulated precisely by means of individual price reductions. Customers with a low willingness to pay can be given an attractive offer in a targeted manner. On the other hand, customers whose purchasing behavior was not very price-sensitive in the past can be “excluded” from advertising for special offers.

Overall, retail today has a wide range of options to dynamize and individualize its prices. The use of the corresponding hardware and software will lead to an increased implementation of corresponding strategies of price differentiation not only in online retailing in the coming years.

Legal framework

Dynamic and individualized prices are to be assessed economically as positive. The entrepreneur's autonomy to set the price himself is also an elementary component of a free economic order. For this reason, the principle of freedom to set prices applies in the German legal system.1 It follows that courts are also not allowed to control the "fairness" of a requested price. As early as 1958, the Federal Court of Justice ruled with unrivaled clarity: "Within the framework of a market-oriented economic system, a company is fundamentally free ... to set its own prices." 2

At the same time, consumer advocates point out, not without reason, that even with the newly discovered forms of dynamic pricing that have been forced by digitization, the interests of consumers must not be lost Opportunities for companies to operate with dynamic prices protect consumers and at the same time present companies with, in some cases, not inconsiderable practical challenges.

Even with dynamic pricing, retailers must avoid misleading consumers:

  • Advertising: To do this, they must ensure that the prices offered in the advertising and the prices actually demanded in the shop do not differ to the detriment of the customer offered price is charged, the retailer not only has to reckon with the displeasure of his customers, but also violates the prohibition of unfair competition law to mislead (§ 5 Abs. 1 Ziff. 2 UWG), can therefore be warned by competitors or consumer associations and, if necessary, to surrender a cease and desist declaration.

The retailer must therefore rule out consumer deception about the price level, even with dynamic prices. Since this is not always easy for technical reasons, there are practical limits to price adjustments, e.g. if the retailer uses printed advertising and wants to set prices dynamically at the same time. De facto, price adjustments downwards are then only possible for the advertising period. If necessary, the retailer will have to use new and more flexible advertising methods with a time limit.

  • Price labeling on the shelf: Similar practical problems arise with dynamic pricing in connection with price labeling on the shelf. This is probably only practicable with the help of digital price tags and thus initially requires considerable investments on the part of the retailer. But even if these technical requirements are met, the retailer must ensure that the price does not change to the disadvantage of the customer on the way from the shelf to the checkout. This illustrates the practical limits of trading when using dynamic prices. A price change is currently only possible outside of business hours. However, it cannot be ruled out that digitization will open up new paths for retail in the future and enable legally compliant behavior.
  • The special case of online retail: online retail is not familiar with the practical problems of stationary retail. The online shop operator can therefore use the options of flexible pricing more intensively today. While dynamic pricing in the online shop can be implemented without any problems, the operators must, for example, ensure that the prices they set in price comparison portals are up to date.5 A challenge that online retailers can already cope with today, for the most part.

Entrepreneurial action can prevent misleading through dynamic pricing. However, this practically limits the possibilities of trade.

Dynamic pricing is not unfair per se either. In principle, it can be assumed that the average informed consumer will recognize that changes in supply and demand in the case of dynamic prices have an impact on the price level. Consumers are already familiar with this from petrol stations, where they refuel more cheaply on Monday evening than on the weekend or during the holiday season. In the future, consumers will also be able to recognize when making regular purchases that they can repeatedly do their purchases more cheaply in times of low frequency in the store; hectic and thoughtless hamster purchases are thus excluded, so that a misleading business activity according to § 5 para. 1 no. 2 UWG does not come into consideration. 6

Group prices

Price differentiations for certain groups of people are discussed particularly critically in public. This is irritating because since the fall of the Discount Act, price reductions have become common for certain people, such as regular customers, and consumers are happy to take advantage of them. It is therefore surprising to see the skepticism of individualized pricing, because it is apparently rated as “unfair” by some consumer advocates and consumers

The basic considerations are:

  • Basically no requirement of equal treatment for prices: It can be stated that the German legal system does not generally recognize a general requirement of equal treatment for retailers in connection with their pricing.8 Rather, freedom of price setting is in the foreground, as is also appropriate for a free market organization. In any case, the question arises whether problematic envy factors do not actually play a major role if consumers want the same prices for everyone, even if this excludes efficiency gains9 and the average price therefore tends to be higher for everyone. However, it cannot and must not be the task of the legislature to satisfy such irrational consumer needs if the objective consumer interests are also counteracted at the same time. A responsible policy must not succumb to the temptation to enact unnecessary and, with regard to the objective of consumer protection, even counterproductive regulations in order to satisfy certain attitudes that are economically disapproving.
  • Transparency requirement: The price setting in relation to the price to be paid by the individual customer must of course be transparent. However, reference prices do not need to be specified for this. The reference to an average price is not necessary in order to create transparency with regard to the specific price.10 He can obtain comprehensive and exhaustive information about the price level even without a reference price by comparing prices.
  • Freedom from discrimination with regard to the General Equal Treatment Act (AGG): As long as there is no violation of the AGG, e.g. because members of certain ethnic groups or because of gender are charged different prices, price differentiations are to be accepted as a result of a free economic system. As an indefinite, subjectively defined term, “fairness” is in any case not a suitable legal standard to justify an obligation to uniform pricing for all consumers.

Apparently different prices for men and women in bulk deals have attracted attention in the political discussion. Different prices are illegal if the price for the same product is different depending on the gender of the customer, although the purchase contracts are typically concluded in a large number of cases without regard to the person under comparable conditions, as is regularly the case in retail. Such pricing obviously constitutes inadmissible discrimination on the grounds of gender and would therefore violate Section 19 (1) of the AGG.

In practice, however, different facts actually determine the discussion. Different prices for functionally comparable but differently designed products are criticized. In these cases, the price is differentiated not depending on the customer, but differentiated depending on the product design. Even if the design specifically addresses a gender, the enlightened consumer is free in this case to opt for the comparable product in a - perhaps less appealing - design at a lower price. A violation of the requirements of the AGG can therefore not be seen in such marketing behavior by the companies. Here, too, entrepreneurial freedom should be given preference over excessive regulations in the wake of a mania for equality.

The applicable fair trading law combined with the provisions of the AGG thus already sets a framework for individualized pricing that effectively protects the customer from being misled and discriminated against. In addition, the seller's autonomy to set prices and differentiate from individual customers may not be restricted by law. Otherwise the principles of our free economic order would be called into question. In a nutshell, Bornkamm states: "There is ... no general ban on price differentiation for the same goods in the same shop." 11 The retailer may therefore deviate from the usual prices for individual customer groups and grant individual special prices. If, despite the already existing overregulation in the area of ​​consumer protection, this freedom of the entrepreneur were also restricted, only a rudiment of freedom of contract would remain. Such an approach would also be constitutionally questionable, because not least the freedom to set prices is protected by Article 2 of the Basic Law (general freedom of action) and Article 12 of the Basic Law (freedom of occupation) as an element of freedom of competition.

Individualized prices

In the case of individualized prices, the retailer must of course first observe the data protection regulations when calculating the price. He may therefore only use the customer's personal data for pricing within the statutory limits of the Federal Data Protection Act without the customer's consent.

The consumer must not be misled by an individualized price, because here too the prohibition of Section 5 (1) No. 2 of the Act against Unfair Competition (UWG) applies. However, this fact is usually not met in the case of personally tailored prices. The individualized and possibly higher price contains neither a statement about the usual average price nor the assertion that the requested product is offered to other consumers at the same price.12 The obligation of the retailer to disclose the reference price under consideration by the consumer protection ministerial conference 13 would be with associated with considerable additional work, would make dynamic pricing disproportionately complicated and de facto discourage traders from using this instrument. This would unnecessarily minimize the opportunities that are associated with individualized prices, not least for consumers with lower incomes. This, too, cannot be the goal of a consumer policy that is otherwise open to any redistribution for social reasons, but tends to ignore the social component of dynamic prices.

New information obligations for the use of individualized prices are also not necessary, because digitization in particular has created a previously unknown price transparency for consumers, which enables them to make an immediate price comparison both in shops and online. However, if the customer consciously refrains from taking advantage of these opportunities despite the existing options, there is no reason to compensate for this self-selected deficit with mandatory information.

Legislative need for action?

Despite the existing legal framework, it is sometimes considered necessary to limit the existing options for dynamic pricing. The 12th Conference of Consumer Protection Ministers on April 22nd, 2016 decided on the following measures to strengthen consumer protection:

  • Retailers should not only be obliged to inform the consumer about the reference price of the products offered, but retailers should also disclose the conditions for a deviation in a "traceable and transparent manner".
  • As part of a "voluntary commitment", online retailers should agree to leave prices unchanged for a minimum period and to graphically display the previous price trend for a certain period of time.

There is no need to further restrict entrepreneurial freedom through these measures and to regulate entrepreneurial activity, as has been shown. Consumers are adequately protected by the existing legal framework. The UWG protects against unfair behavior on the part of the retailer and, in particular, against misleading. The provisions of the AGG also effectively exclude discrimination against certain consumer groups. The data protection law guarantees that the customer's data is not used for the price calculation without authorization.

The planned voluntary commitment to provide information about reference prices and price trends is problematic, not least from an antitrust perspective, as the entrepreneur is obliged to disclose otherwise secret information.In this way the transparency of the market is artificially increased. This can facilitate coordinated competitive behavior.15 In any case, if this strategically important information is made transparent, competition incentives are reduced.16 This restricts competition in the supposed interest of consumers. This effect would be reinforced by the obligation to keep prices unchanged for a certain period of time. The anti-competitive behavior advocated by the consumer protection ministers would tend to lead to higher average consumer prices in the medium term.


The legislature should take its duty seriously to carefully weigh the individual interests of market participants against each other before considering further legal measures to enforce alleged consumer interests. Additional burdens on the entrepreneur in connection with possible new information obligations, practical difficulties in the legally secure presentation and in particular the constitutionally protected freedom to set prices must be contrasted with the intended increase in consumer protection, taking into account the existing possibilities for consumers to independently obtain information. The result will be that new information obligations, which are difficult to implement in practice and thus can undermine the freedom of entrepreneurs to set dynamic prices, must be dispensed with.

The legislature is therefore well advised to act cautiously in the dynamically developing environment of price setting and not to withhold the advantages of free (price) competition from the consumer. In practice, this means that more consumer protection is achieved than with any regulation.

  • 1 BGH of March 13, 2003, Az. I ZR 212/00, NJW 2003, p. 2096 f.
  • 2 BGH of April 18, 1958, Az. I ZR 158/56, NJW 1958, p. 1140.
  • 3 Minutes of the results of the 12th Conference of Consumer Protection Ministers on April 22nd, 2016 in Düsseldorf, p. 39 f., Https://
  • 4 H. Köhler, J. Bornkamm, J. Feddersen: Law against Unfair Competition, 34th edition, Munich 2016, Bornkamm § 5 marginal no. 7.17.
  • 5 BGH, judgment of March 11, 2010, Az. I ZR 123/08, MMR 2010, p. 745.
  • 6 Cf. F. Hofmann: The tailor-made price, in: Competition and Practice (WRP), H. 9, 2016, pp. 1074, 1078.
  • 7 Minutes of the results of the 12th Conference of Consumer Protection Ministers, loc. Harris Interactiv AG: Dynamic and individual pricing - opportunities for retailers, indecisive consumers, press release from June 9th, 2016, (December 5th, 2016).
  • 8 H. Köhler, J. Bornkamm, J. Feddersen, loc. Cit., Bornkamm § 5 marginal no. 7.14.
  • 9 Harris Interactive AG, loc. Cit.
  • 10 F. Hofmann, loc. Cit., Pp. 1074, 1080.
  • 11 H. Köhler, J. Bornkamm, J. Feddersen, loc. Cit., Bornkamm, § 5 marginal no. 7.15.
  • 12 F. Hofmann, loc. Cit., Pp. 1074, 1080.
  • 13 Results of the 12th Conference of Consumer Protection Ministers, loc.
  • 14 Ibid.
  • 15 Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal cooperation agreements, 2011 / C11 / 01, para. 65.
  • 16 Ibid, para. 86.

* The research project “The Influence of Dynamic Pricing on Consumer Confidence and the Choosing a Place to Buy in Stationary Food Retailing - An Empirical Analysis” was carried out by the Ministry for Innovation, Science and Research (MIWF) of the State of North Rhine-Westphalia as part of the Consumer Research Competence Center NRW (KVF NRW ) promoted. The KVF NRW is a cooperation project of the MIWF with the consumer center North Rhine-Westphalia e.V. and the Ministry for Climate Protection, Environment, Agriculture, Nature and Consumer Protection (MKULNV) of the state of North Rhine-Westphalia.

Dynamic and personalized: How is price setting developing in online retail?

Michael Schleusener

The subject of “dynamic pricing”, often also in connection with the terms “individual prices” and “personalized prices”, is enjoying increasing public awareness. Dynamic pricing goes hand in hand with the increasing availability and processing of more and more data that companies collect from consumers in many places and in the most varied of forms.

In online trading, lower menu costs (price change costs) give rise to new leeway for price adjustments. The number of price changes has increased while the amount of each price change has decreased - so there are many small price changes. In addition, competitive prices can be read out easily and automatically. Reactions to these competitive prices are possible more quickly and thus implemented more frequently. However, the skills and time of management are still limiting factors.1 From the consumer's point of view, the picture emerges of a large number of price changes, the causes of which are not immediately apparent and are therefore related, among other things, to their own behavior and thus perceived as personalized. Observable price changes over time can be traced back to deliberately different prices at different times or are the result of the price-setting activities of competitors and are therefore not demand-oriented at all, but rather competition-oriented.

Personal, individual price differentiation should be understood to mean that a customer is shown an individual price based on personal data that is not offered to any other customer even under similar situational circumstances (time, place, hardware used) (first degree price differentiation). Dynamic pricing, on the other hand, refers to prices that change over time, often oriented towards the competition and independent of the individual data of a user (price differentiation 2nd degree). Another reason for different prices over time can be different willingness to pay on the part of users, which should be skimmed off (3rd degree price differentiation). This form of price differentiation can only be used if arbitrage can be prevented.

Dynamic pricing

A temporal price differentiation according to the time of purchase is well known. Consumers have known this form of pricing in the aviation industry since 1978.2 Since then, this form of pricing has evolved in the context of revenue management. Consumers have understood the pricing mechanism and accept different prices based on the differences in transactions (different booking times). 3

It should be examined to what extent online trading is comparable to the service industries in which revenue management has been practiced up to now. Suppose an online retailer varies its prices over time, for example over the course of a day. A price reduction increases the total contribution margin if the additional contribution margin achieved with this price reduction more than compensates for the waiver of unit contribution margin that occurs at the same time. Due to the variable costs, the percentage increase in volume must be significantly greater than the percentage price reduction in order to achieve a positive contribution margin effect. Since a market expansion is hardly to be expected for most goods, this additional quantity will largely come about as a result of a time shift. This policy can only be economically worthwhile if the excess volume comes from competition. In the case of an oligopolistic supply structure, such changes in volume will be noticed by competition. This in turn will now lower prices in order to regain the lost quantity. In the end, all providers have lowered their prices, volumes and market shares have stayed the same, their returns have fallen and consumers are paying less.

These effects can occur much faster if the pricing and the determination of competitive prices are automated. The example was documented very early on that the price for a simple scientific paperback had increased to over US $ 7 million through automated price adjustments over several rounds, which makes the absurdity of such an unreflective approach clear

Rule-based dynamic pricing is more commonly used today. The online retailer can thus set up a differentiated set of rules according to which prices are adjusted in a competitive manner. Only prices from certain retailers can be matched, or a price difference to the competition can be maintained for certain brands. In terms of content, this is not a new approach; what is new is the automated collection of competitive prices. This technology-driven approach is available to all retailers without major investments, as it is now also offered as Software as a Service (SaaS). Marketplace sellers can even use such an integrated tool on Amazon's marketplace. However, these software solutions do not release the retailer from defining the basic pricing strategy, the value position of his offer and the brand premium for the brands offered

There is also the option of using more complex algorithms. In addition to the price, a number of other independent variables are identified that can influence the sales volume. This includes the competitive prices, the range of comparable products in a branch, the type of product presentation, customer reviews and also the time of day and day of the week. On the basis of historical data, an attempt is made to determine the respective significance of these influencing factors and then to make forecasts of the sales quantities at alternative prices. The decisive factor is which proportion of variance these models can explain and how structural breaks are dealt with. Naturally, the algorithms used are not disclosed by the providers, so that it is difficult to estimate the potential for success of these measures.

If a company changes the prices over time, customers will experience that the price for the recently purchased product from the same supplier has dropped for a short time. It is to be expected that at least some of the affected customers will get in touch and also ask for the lower prices. A strategy of dynamic price changes therefore also includes dealing with such customer inquiries. A conceivable solution approach is practiced by the American company BestBuy, which issues a time-limited price guarantee and matches both its own lower prices and those of selected online shops and stationary retailers.6 The customer will then be reimbursed the price difference on request. Such price guarantees are to be assessed depending on the contribution margin, pricing, competitive situation and redemption behavior of the customers. Alternatively, they could re-order the item at the lower price and send the item back at the higher price, which is an interesting option, especially for free returns. However, such a form of arbitrage opposes the implementation of price differentiation.

In addition, it is expected that the attention given to the award will continue to increase. If customers expect frequent price changes, it is becoming more and more attractive to take advantage of these price changes and to buy at the lowest price. New companies and offers are emerging that make it easier for customers to take advantage of the best time to buy. Providers like send a reminder when the desired price has been reached on and also show the historical price development. The Spottster company offers a similar service but is watching a large number of traders. In the hotel sector, DreamCheaper takes advantage of price fluctuations for hotel rooms by canceling the originally booked hotel room in the event of a positive price development and then re-booking it for less. On the website in the USA, the price development was predicted and the forecast of the cheapest time to buy was also backed up with its own price guarantee.7 Against this background, the question arises why a user should buy himself today when he has software According to your wishes ("at the best price and delivery within a week at the latest", for example) you can choose the best offer and order it yourself.

Personalized pricing

Personalized prices are often suspected but can rarely be proven. In an exploratory study in which providers in Germany from ten industries were examined with the help of different stimuli, personalized pricing could only be determined in the tourism sector.8 A study by the Chamber for Workers and Employees for Consumer Protection in Vienna came to similar results.9 Two other studies in Europe and the USA were also only able to determine personalized prices to a very limited extent in 2012 and 2014.10 Most of the reports on personalized prices could be based on other forms of price differentiation (e.g. location-based) or simply tests or inadequacies in the display of the Reduce prices.

Even if it were possible to use the user data to predict individual willingness to pay, the implementation of personalized prices still faces a number of obstacles. Consumers who visit an online shop via price search engines must receive the price displayed in the search engines. For them, only price differentiation via the sales channel comes into consideration. If the online provider still has a brick and mortar store, i.e. if he is active as a multichannel retailer, then individualized online prices lead to different prices online and offline. In a competitive comparison with other online retailers, the customer may well find other, lower prices than the price that corresponds to his willingness to pay.11 This is the case when the other online retailers also offer individual prices, but possibly at a different one Assessment of the customer's willingness to pay. It is more likely, however, that the companies do not even have the skills to determine such willingness to pay. In addition, they may have other corporate goals. With a volume-oriented strategy, the maximization of sales comes to the fore and a profit margin-maximizing skimming of willingness to pay is dispensed with. Another central point, which will not be discussed further in this article, is the consumer reaction against the background of the perceived fairness of prices, which can lead to negative consequences for the company

Consumers can also use software applications to react to personalized pricing and to have to pay the lowest possible prices. In the simplest case, the consumer would leave the order to a person who gets cheaper prices based on their profile. Or the transmitted data is manipulated by the consumer's software in such a way that the characteristics of a person with low willingness to pay are transmitted to the server of the online shop, so that correspondingly low prices can be realized.

Conclusion and theses on further development

  1. The current situation is only partially meaningful, as companies and service providers experiment a lot. Many important prerequisites, for example for a clear identification of customers across end devices, have not yet been met in the technical implementation.
  2. It can be stated that price differentiation will play a bigger role online. But first of all, the price differentiation across distribution channels is likely to gain in importance and enable consumers, who bear the corresponding search costs, access to low prices.
  3. Before the introduction of personalized prices, an individualization of the product range shown is to be expected.
  4. Individual prices will probably initially be implemented as individual discounts via vouchers.
  5. A complete absorption of the consumer surplus will not be possible as long as the providers pursue different goals in competition. New competition will often arise, especially on the Internet, with its very low market entry barriers via marketplaces.
  6. Consumers, for their part, will use the technology to obtain the most affordable prices they can get. When using data-based, personal prices, customers will be able to manipulate their profile data with the help of standard software and also transfer the purchase decision for a specific retailer to an agent (shop bot).
  7. The way to immunize its own customers against competitive offers is very promising, as Amazon is already successfully implementing with its Prime offer. Such an approach prepares the ground for later being able to play out individual prizes without being permanently exposed to competition.
  • 1 Cf. M. Bergen et al .: Shattering the Myth of Costless Price Changes, in: European Management Journal, 21st vol. (2003), no. 6, pp. 663-669.
  • 2 Cf. R. Klein, C. Steinhardt: Revenue Management, Berlin, Heidelberg 2008, pp. 2 ff.
  • 3 See M.Friesen, S. Reinecke: Perceived price fairness in revenue management in air traffic, in: Thexis 4/2007, pp. 34-39.
  • 4 See M. Eisen: Amazon’s $ 23,698,655.93 book about flies, (10/17/2016).
  • 5 Cf. D. Bounie et al .: Online Price Dispersion: An International Comparison, Interdisciplinary Institute for Innovation, Working Paper 12-TS-02, April 11, 2012.
  • 6 See A. Zimmerman: Best Buy Plays Web Hardball, WSJ, October 12, 2012, p. B1, (October 22, 2016).
  • 7 The offer from was developed on the basis of an investigation of freely available prices for flights, see on the scientific approach O. Etzioni et al .: To Buy or Not to Buy: Mining Airline Fare Data to Minimize Ticket Purchase Price, SIGKDD '03, August 24-27, 2003, Washington DC.
  • 8 Cf. M. Schleusener, S. Hosell: Personalized price differentiation in online trading, investigation and elaboration on behalf of the Expert Council for Consumer Questions at the Federal Minister of Justice and Consumer Protection, October 2015, http://www.svr-verbü wp-content / uploads / 2016/01 / Preisdifferenzierung-im-Onlinehandel_eWeb-Research-Center.pdf (22.10.2016).
  • 9 See Chamber for Workers and Salaried Employees in Vienna (publisher): Dynamic Pricing - the individualization of prices in e-commerce, November 2015, ( 10/22/2016).
  • 10 See A. Hannaket al .: Measuring Price Discrimination and Steering on E-commerce Web Sites, Proceedings of the 2014 Conference on Internet Measurement, pp. 305-318; J. Mikians et al .: Detecting Price and Search Discrimination on the Internet, Proceedings of the 11th ACM Workshop on Hot Topics in Networks, pp. 79-84.
  • 11 Cf. D. Ulph, N. Vulkan: Electronic commerce and competitive first-degree price discrimination, (22.10.2016).
  • 12 See O.V .: Varies Prices of Identical Items for Test, The Wall Street Journal, September 7, 2000, p. B19; L. Xia, K.B. Monroe, J.L. Cox: The Price Is Unfair! A Conceptual Framework of Price Fairness Perceptions, in: Journal of Marketing, 68th year (2004), pp. 1-15; S. Spiekermann: Individual Price Discriminaton - An Impossibility ?, in: A. Kobsa et al. (Ed.): International Conference for Human-Computer Interaction, Proceedings of the CHI2006 Workshop on Privacy-Enhanced Personalization, Montréal 2006, pp. 47-52.

The consumer science perspective: from customer confusion to price confusion?

Peter Kenning, Maximilian Pohst

In the course of digitization, operational marketing has also undergone considerable changes in recent years. In addition to the digitization of communication and production, this development is now also increasingly affecting company pricing policy.1 Newer approaches such as “Name-Your-Own-Price-Strategies”, “Real Time Bidding” and “Targeted Pricing” are found in operational reality Precipitation and form the subject of scientific work.2 Many of the mostly empirical studies that are based in this area, however, focus more or less one-sidedly on the opportunities and risks that arise for companies located on the supply side of the market. In addition, the subject of this article is to consider the phenomenon of dynamic pricing from a consumer science perspective and, based on this, to derive theses that relate to its future significance and further development.

The neoclassical concept falls short

One possibility of integrating the perspective of the consumer or the customer into the discussion about dynamic pricing is based on the model of the responsible consumer, to conceptualize his economic role as the customer of a good in a neoclassical way.3 In fact, many economic studies on this The subject of price is primarily understood as a neoclassical concept. Accordingly, the model of the price-sales function is mostly used and the possible advantage of price differentiation in the simplest model of a price-sales function that usually falls linearly is pointed out.

From a consumer science perspective, however, this theoretical basis is hardly up-to-date, because many current research work shows that the price not only has an information-economic significance that goes beyond the neoclassical functions, but also that other factors can influence the function of the price for consumer behavior .

In addition, it has been proven that price is only one, albeit an important, determinant of consumer behavior of interest. And finally, consumer policy, which acts as the addressee for consumer science, increasingly pursues a differentiated consumer model, which often contradicts the assumption of the responsible consumer.4 Accordingly, from a modern consumer science perspective, it appears expedient to introduce at least information economics and behavioral economics into the discussion to allow a more complete consideration and assessment of dynamic pricing.

The information economic perspective

Information economics is based on the assumption that people (want to) behave rationally but cannot, e.g. due to information asymmetries (IAS), which can lead to market failure. A well-known example of this is the “Market for Lemons” described by George Akerlof.5 Two approaches are available for dismantling the IAS: “Signaling” and “Screening” .6 In both approaches, the price can represent information for the consumer by signaling a certain quality, for example. Both approaches are based on the fundamental assumption that the functioning of markets is based on the principle that more information is better than a little. In fact, this is not the case because consumers' cognitive resources are limited. In this respect, information-economic approaches can play a major role in explaining and shaping consumer behavior in the context of dynamic pricing, but they are usually not sufficient, so that complementary alternatives to the so-called “information paradigm” are often required in consumer science

Behavioral Economics: Customer Confusion

The limitation of neoclassical and information-economic approaches in the consumer science context becomes particularly clear in a phenomenon for which the term “customer confusion” was established around twenty years ago.8 The empirical study by Dhar, which has now been cited almost a thousand times, showed that an increase in the Selection decisions perceived from the customer's point of view do not lead to a better clearing of the market, but on the contrary had a negative effect on it. While 40% of the customers stayed with an offer of six alternatives, of whom 30% then bought, with an offer of 24 alternatives 60% of the customers stayed, but of these only 3% bought. This almost complete market failure was justified by the fact that many test subjects were so overwhelmed or confused by the 24 alternative offers that they were no longer able to make a decision.

In the continuation of the scientific studies, it was then possible to differentiate between different types of customer confusion. In particular, “Choice Overload” and “Information Overload” should be mentioned here.9 Choice Overload occurs when the number of alternatives is unmanageable for the consumer. Information overload, on the other hand, occurs as soon as the alternative is characterized by too many features.

Since prices are also used as a source of information from the point of view of information economics, it is conceivable that the dynamic pricing approach could lead to a special form of customer confusion - more precisely a special form of information overload. This particular form could be referred to as “price confusion”. To investigate whether such a sub-form actually exists, an empirical study was carried out together with a large German food retailer.

For this purpose, consumers were confronted with various supplier-product-price combinations. The basic hypothesis was analogous to Dhar 10 that in certain situations too many alternatives could have a negative effect on buying behavior. Specifically, it was assumed that several different, i.e. inhomogeneous prices for identical products at different retailers lead to higher customer confusion than in the case in which the prices for these products are homogeneous at all retailers. To investigate this hypothesis, three categories of food products were initially chosen that consumers felt were difficult to assess. This was to avoid the test subjects being able to use internal reference prices when processing the price information. In addition, the information-economic importance of price information was emphasized. Price fluctuations of up to 30% in the respective stimuli were also used to vary the price homogeneity. In addition, the number of distributors offering was varied in order to achieve the greatest possible variance between the stimuli (see Figure 1).

illustration 1
Example of a price inhomogeneous stimulus

Source: own illustration.

The sample developed to test the aforementioned hypothesis comprised 236 respondents. The average age of the respondents was 36.05 years. To test the hypotheses, a regression analysis was calculated in which the stated customer confusion was the dependent variable and the number of dealers and the different price were the independent variables. The food categories served as control variables. As a result, the following effects could be determined:

In the case of inhomogeneous prices, consumers were more confused than in the case of homogeneous prices (scale: 1 = indicator for no confusion, 7 = indicator for high confusion):

M.homogeneous = 2.15 (SE = 0.075),

M.inhomogeneous = 2.64 (SE = 0.092) at p <0.001.

It could also be shown that with inhomogeneous prices and an increasing number of retailers, consumer confusion increased in line with expectations (p = 0.054).

Building on this, it can be assumed that the stronger the dynamic pricing approach takes place in everyday consumer life and accordingly leads to more different price information, the higher the likelihood of price confusion and the associated likelihood of situational or partial market failure. There is therefore a risk of “triple loss situations” in which the consumer does not get the desired goods, the provider does not realize the desired turnover and the state misses out on corresponding tax revenues.

Regulations in the event of market failure

In view of these results, the current demands of consumer policy to accompany the further development of dynamic pricing on a regulatory basis do not seem to be unjustified. At the same time, it should be noted that in markets where competition works, there is also the possibility of self-regulation.

So is there a regulatory need? The answer to this question is “Yes, but only if there are no alternative correction options, e.g. collective mechanisms that the customer can rely on and guarantee the functionality of the market.” In concrete terms, this means that the relevance of the individual price information for the purchase decision may be overestimated, because in fact consumers rarely know the exact price of a product even without dynamic pricing; but they buy it anyway.

As evidence for this thesis, which only seems strange at first glance, one can first of all take a look at the literature on the subject of “Price Knowledge” to see that most consumers have little idea about prices. This is shown by numerous empirical studies that have also been carried out in Germany, among others.11 From a naive regulatory perspective, this could suggest developing approaches, e.g. in the area of ​​consumer education, in order to reduce this incomplete price knowledge and thus avoid any disadvantages for consumers.

When is there (no) need for regulation?

Why is there probably no need for regulation at this point? Well, because consumers behave ecologically and rationally, even if they have no price knowledge.12 The term ecological rationality describes an adaptive form of rationality that leads to optimal results in the respective situation.

In order to clarify this fact in a specific case, one must first look at the economic logic of the food retail industry. Since the return on sales in the food retail sector is relatively low, with the corresponding fixed costs, a small loss of sales is often sufficient to compensate for the low profit margins. For some retailers, for example, it would be sufficient for 15% of customers to switch retailers due to a possible price increase for this retailer to realize losses. Building on this, it becomes clear that not all customers need to know all prices in order to be able to adequately sanction a dealer. It is sufficient if only a few customers know a few prices (you could call them “born market watchers”).

In the context of dynamic pricing, however, it should now be noted that this collective protective mechanism fails if the concept of dynamic pricing is accompanied by the concept of personalized prices (targeted pricing). Because the basic idea of ​​targeted pricing is precisely to individualize the respective prices. In this case, there could be a fundamental need for regulation if consumers are no longer able to compare the advantageousness of the respective prices either with prices that other customers have paid or with prices that have been paid in the past . In times when many companies are very interested in building long-term, profitable customer relationships with the help of the concept of "Customer Relationship Management", companies should also be interested in enabling their customers to make this comparison. Perhaps this is also the reason why such approaches are seldom observed in practice at the moment.13 If the companies do not have this interest, however, there would probably be a need for consumer policy to develop corresponding information-economic instruments and, if necessary, also to use them Help to flank behavior-based tools. In all other cases, however, one would initially assume that competition itself is regulating corresponding undesirable developments from the consumer's point of view.


In summary, from a consumer science perspective, it can be said that an isolated neoclassical analysis of dynamic or targeted pricing is not very fruitful and hardly up-to-date. If you add information and behavioral economic aspects to this, you come to the following theses:

  • The stronger the dynamic pricing approach takes place in everyday consumer life, the higher the probability of situational or partial market failure.
  • This probability can then be weakened if consumers have the opportunity to act ecologically and rationally.
  • If the concept of dynamic pricing is flanked by the approach of targeted pricing, this possibility does not apply - there is a risk of theoretically situational or partial market failure.
  • But: The more a provider relies on building long-term, profitable customer relationships and the stronger the competition, the less incentive they will have to use these two instruments.

It could therefore be that a functioning market cleans itself up or corrects itself at an early stage, so that there is no need for consumer policy regulation. At the same time, scientific studies should accompany further developments and contribute to empirically substantiating the theses mentioned and observing the corresponding market developments.

  • 1 Cf. P. Kenning, M. Pohst: Dynamic Pricing, in: WISU - Das Wirtschaftsstudium, H. 10/2016, pp. 1125-1129.
  • 2 Cf. i.a. P. Kenning, T. Eberhardt: Methods of Price Formation, in: J. Zentes et al. (Ed.): Handbook Handel, Wiesbaden 2013, pp. 609-629.
  • 3 Cf. C. Strünck et al .: Is the “responsible consumer” a myth? On the way to a realistic consumer policy, opinion of the Scientific Advisory Board on Consumer and Nutrition Policy at the Federal Ministry of Food, Agriculture and Consumers, December 2012.
  • 4 Cf. C. Strünck et al., Loc. Cit.
  • 5 Cf. G. A. Akerlof: The Market For "Lemons": Quality Uncertainty And The Market Mechanism, in: Quarterly Journal of Economics, 84th year (1970), no. 3, pp. 488-500.
  • 6 Cf. P. Kenning, I. Wobker: Affective and cognitive behavioral strategies for overcoming information asymmetries in the consumer goods trade - an empirical analysis with cartel law implications, in: Betriebswirtschaftliche Forschung und Praxis (BFuP), 64th year (2012), volume 6 , Pp. 626-642.
  • 7 See P. Kenning, L.Reisch: Alternatives to the information paradigm of consumer policy: A commentary introduction to a still dynamic consumer science field with consumer policy implications, in: Journal of Consumer Protection and Food Safety, 8th year (2013), no. 3, pp. 141-142.
  • 8 Cf. R. Dhar: Consumer Preference for a No-Choice Option, in: Journal of Consumer Research, 24th Jg. (1997), H. 2, pp. 215-231.
  • 9 Cf. R. Greifeneder, B. Scheibehenne, N. Kleber: Less may be more when choosing is difficult: Choice complexity and too much choice, in: Acta psychologica, 133rd Jg. (2010), H. 1, p. 45-50; as well as R. Grunder: Development factors of consumer confusion: empirical analysis of the influence of supply and person-specific factors on the perception of consumer confusion, Tönning 2006.
  • 10 R. Dhar, loc. Cit.
  • 11 Cf. H. Schneider, P. Kenning, V. Hartleb, T. Eberhardt: Consumers' implicit price knowledge - really better than their explicit price knowledge ?, in: Marketing ZFP - Journal of Research and Management, H. 4, 2009, p 219-233.
  • 12 Cf. D. G. Goldstein, G. Gigerenzer: Models of Ecological Rationality: The Recognition Heuristic, in: Psychological Review, 109th vol. (2002), no. 1, pp. 75-90.
  • 13 Cf. M. Schleusener, S. Hosell: Expertise on the subject of "Personalized price differentiation in online retail", SVRV - Expert Council for Consumer Issues, 2015.

The consumer policy perspective: current developments in online retail

Johannes Remmel

The subject of "Dynamic Picing - and the implications for consumer science and consumption policy" is of particular interest, because we know little about what will actually be in store for us in terms of corporate pricing in the coming years and what that means for consumers.

The Internet - more precisely, the commercial use of the Internet - has accompanied our society for around 25 years. At that time, no one had really been able to grasp the tremendous upheavals it would bring about for society, the economy and the citizens. The quote from Ron Sommer, the head of Deutsche Telekom at the time, is well known. In 1990 he said: "The Internet is a gimmick for computer freaks, we don't see a future in it". Many people thought that at the time - and were surprised.

Social change