• No results found

Ethical design: Persuasion, not deception

N/A
N/A
Protected

Academic year: 2022

Share "Ethical design: Persuasion, not deception"

Copied!
14
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

not deception

Received (in revised form): 7th May, 2021

Jeanette R. van der Lee

PhD student, Utrecht University and Netherlands Authority for Consumers and Markets, The Netherlands

Jeanette R. van der Lee is a PhD student at Utrecht University and the Netherlands Authority for Consumers and Markets. Her current research focuses on the psychology of supervision, bringing together social psychological research and regulatory practice. She has a research master’s degree in social and organisational psychology from Leiden University.

Universiteit Utrecht, Heidelberglaan 1, 3584 CS Utrecht, The Netherlands Tel: +31 611314334; E-mail: j.r.vanderlee@uu.nl

Dries Cuijpers

Senior Enforcement Official, Netherlands Authority for Consumers and Markets, The Netherlands

Dries Cuijpers is a senior project manager for the Netherlands Authority for Consumers and Markets.

A graduate in European and international law from Radboud University, he has more than 15 years’

experience as an enforcement officer, working for various regulators and in different sectors.

Autoriteit Consument & Markt, Postbus 16326, 2500 BH Den Haag, The Netherlands Tel: +31 6 31035871; E-mail: dries.cuijpers@acm.nl

Mareille de Bloois

Behavioural Scientist, Netherlands Authority for Consumers and Markets, The Netherlands

Mareille de Bloois is a project manager for the Netherlands Authority for Consumers and Markets

‘effective transparency’ project, which collaborates with businesses to increase the effectiveness of online disclosures. She studied social psychology at the University of Amsterdam and specialised in consumer behaviour. Her work is focused on integrating behavioural insights in ACM’s enforcement actions.

Autoriteit Consument & Markt, Postbus 16326, 2500 BH Den Haag, The Netherlands Tel: + 31 6 52527033; E-mail: mareille.de.bloois@acm.nl

Jessanne Mastop

Senior Behavioural Scientist, Netherlands Authority for Consumers and Markets, The Netherlands

Jessanne Mastop is a senior behavioural scientist and coordinator of the Behavioural Insights Team at the Netherlands Authority for Consumers and Markets, where her work focuses on online consumer behaviour and organisational decision-making and compliance behaviour. She has a PhD on the interpersonal effects of non-verbal behaviour from Leiden University.

Autoriteit Consument & Markt, Postbus 16326, 2500 BH Den Haag, The Netherlands Tel: + 31 6 52527082; E-mail: jessanne.mastop@acm.nl

Winnie van Heesch

Senior Enforcement Official, Netherlands Authority for Consumers and Markets, The Netherlands

Winnie van Heesch is a senior enforcement official and strategy adviser for consumer protection at the Netherlands Authority for Consumers and Markets. A graduate in political science from the University of Amsterdam, she has more than 15 years of experience in consumer protection and competition law.

Autoriteit Consument & Markt, Postbus 16326, 2500 BH Den Haag, The Netherlands

Tel: +31 6 31035811; E-mail: winnie.van.heesch@acm.nl

(2)

Elianne F. van Steenbergen

Professor, Utrecht University, The Netherlands

Elianne F. van Steenbergen is a professor, specialising in the psychology of supervision, at Utrecht University, where she is part of the Organizational Behavior Group research group. She is also a senior supervisory officer in the Behaviour & Culture expert team at the Dutch Authority for the Financial Markets.

Universiteit Utrecht, Heidelberglaan 1, 3584 CS Utrecht, The Netherlands Tel: +31 253 3082; E-mail: e.f.vansteenbergen@uu.nl

Abstract Companies are continuously developing and refining techniques to influence the online consumer. There is, however, a fine line between persuasion and deception.

This article discusses the boundaries of online persuasion, drawing on the guidelines developed by the Netherlands Authority for Consumers and Markets. It describes the pitfalls associated with knowledge disparity between sellers and consumers, and argues that information asymmetry, cognitive biases and personalised communications can all impair the ability of consumers to make an informed choice. Using real-life examples, the article demonstrates how the design of online sales environments can go wrong. The article goes on to describe the benefits of ethical design, and provides guidance on how to make ethical design choices and use consumer data to test whether the design of an online environment is unfairly influencing consumers. Using these guidelines, marketing professionals can make thoughtful choices in the design of their online environment that will benefit both their company and its customers.

KEYWORDS: cognitive biases, consumer protection, dark patterns, online marketing ethics, unfair commercial practices, fairness by design

INTRODUCTION

At what point does online marketing cross the line from persuasion to deception?

A recent legal complaint filed by the Norwegian Consumer Council against Amazon raises the question whether companies always draw the line in

accordance with the law — more specifically, consumer protection law. According to the complaint, Amazon makes it very easy to sign up to Amazon Prime, but imposes many hurdles when it comes to cancelling that subscription. 1

Amazon is not the only company to make contentious choices when designing its online shopping environment, potentially prioritising conversion over consumer protection. Amazon is one of the largest companies in the world, but any company with a digital presence, no matter their size,

faces ethical design questions. In today’s highly competitive digital marketplace, consumers can choose from almost infinite options. It therefore stands to reason that online sellers and online marketing companies will search for ways to attract consumers’ attention, retain them as customers, and get them to buy as much as possible. Assuming that most marketers have good intentions, where does it go wrong?

Part of the issue, particularly in the online world, is that companies have more knowledge and data than individual consumers do. This disparity is threefold:

companies have more information about the product than consumers do, especially in relation to innovative online services;

companies can use marketing techniques

that tie into consumers’ cognitive biases; and

companies deploy large amounts of data

(3)

about individual consumers and groups to design and continuously test those persuasive techniques. Today’s companies can leverage data and deploy persuasive techniques online at scale, while incurring relatively low costs.

This disparity in knowledge and data does not have to be a problem, but it can be misused to manipulate consumers into making choices they would otherwise not have made. As research has shown, 2,3 this can lead to economic harm for consumers;

declining consumer trust in online markets;

and the loss of people’s autonomy in decision making. Furthermore, consumers will often not be in a position to correct for these market failures. In order to prevent harmful outcomes, the business community must take responsibility and ensure it is compliant with consumer law.

To clarify the legal limits and thus facilitate the business community’s compliance with the law, the Netherlands Authority for Consumers and Markets (ACM) has published guidelines on the boundaries of online persuasion. 4 Since the publication of these guidelines, the ACM has intervened (eg with Booking and Expedia 5 ), to ensure consumer rights and interests in the digital sphere are upheld. Simultaneously, the guidelines also help safeguard a level playing field for businesses, in which they can compete fairly on price, quality and service.

So, what is allowed in persuading consumers and what is not? Sellers are allowed to persuade consumers through marketing, but they are not allowed to mislead consumers. Anyone selling products or services to consumers must abide by the consumer law of the country it sells in.

Misusing cognitive biases to steer consumers away from their best interests can easily constitute an illegal unfair commercial practice, as described in the European Union Unfair Commercial Practices Directive. 6

Besides setting out the legal context applicable to online marketing practices (in the European Union), the present article also provides online sellers with

practical advice on how to be compliant, for example, through the use of A/B testing to study whether their marketing practices fulfil the legal requirements of consumer protection law. To be clear, the article focuses on consumer law and the use and misuse of consumer data, rather than the collection of consumer data. Data collection falls under data protection law and is beyond the scope of this article.

This article will first consider what companies must do to be compliant with consumer law and what the benefits of ethical design are, both for companies and consumers. This will be followed by a discussion, including real-life examples, of three areas where sellers must be careful that persuasion does not turn into deception:

information asymmetry, design based on cognitive biases, and the use of consumer data. The article will end with advice for sellers on how to use consumer data for the ethical design of their online environment rather than for a solely conversion-driven design that runs the risk of violating the law.

This article aims to show that conversion and consumer protection can go hand in hand.

THE EUROPEAN LAW ON UNFAIR COMMERCIAL PRACTICES

What sellers are and are not allowed to do to convince consumers is legally established in the European consumer protection legislation. The Directive on Unfair Commercial Practices, Directive 2005/29/

EC, 7 is an important element of this body of law. This law focuses on the misuse of legally obtained data for targeting consumers and on how to use data to protect consumers.

It considers commercial practices as unfair, and therefore illegal, if they are ‘contrary to the requirements of professional diligence, misleading, or aggressive’. 8

According to the European Directive, a commercial practice is misleading if:

it contains false information and is

therefore untruthful or in any way,

(4)

including overall presentation, deceives or is likely to deceive the average consumer, even if the information is factually correct, in relation to one or more of the following elements, and in either case causes or is likely to cause him to take a transactional decision that he would not have taken otherwise … A commercial practice shall also be regarded as misleading if, in its factual context, taking account of all its features and circumstances, it causes or is likely to cause the average consumer to take a transactional decision that he would not have taken otherwise. 9

With respect to misleading omissions, the European Directive has the following to say:

A commercial practice shall be regarded as misleading if, in its factual context, taking account of all its features and circumstances and the limitations of the communication medium, it omits material information that the average consumer needs, according to the context, to take an informed transactional decision and thereby causes or is likely to cause the average consumer to take a transactional decision that he would not have taken otherwise … It shall also be regarded as a misleading omission when, taking account of the matters described in paragraph 1, a trader hides or provides in an unclear, unintelligible, ambiguous or untimely manner such material information as referred to in that paragraph or fails to identify the commercial intent of the commercial practice if not already apparent from the context, and where, in either case, this causes or is likely to cause the average consumer to take a transactional decision that he would not have taken otherwise. 10 Finally, the European Directive discusses aggressive commercial practices thus:

A commercial practice shall be regarded as aggressive if, in its factual context, taking account of all its features and circumstances, by harassment, coercion, including the use of physical force, or undue influence, it significantly impairs or is likely to significantly impair the average consumer's freedom of choice or conduct with regard to the product

and thereby causes him or is likely to cause him to take a transactional decision that he would not have taken otherwise. 11

There is a fixed list of commercial practices that is considered unfair per se, such as claiming a product is free when in reality costs are involved. For the full list, see Annex I of the European Directive. 12 However, the European Directive also contains open norms, which are broadly applicable and require interpretation for each marketing technique. For example, important information about a product should be presented clearly and in an easy-to-understand way. Of course, what is important and what is easy to understand depends on the context and the consumer.

To help with that interpretation, the ACM has developed guidelines on the protection of the online consumer, as described above.

Likewise, the focus of this article is on the interpretation of the open norms.

In short, again, sellers are allowed to persuade consumers, but they are not allowed to mislead them. To be clear, deception does not have to be deliberate as intent is not a relevant factor for a commercial practice to be in violation of the law. Marketers may think they are just using smart marketing techniques, yet these techniques may result in consumers being misled. For this reason, the ACM advises sellers to inform themselves about the actual effects of their marketing techniques. It is not sufficient to claim good intentions.

Regular marketing is used to inform potential consumers that a product exists and why they might want to buy it, for example, by providing information about the desirable qualities of the product, such as useful features, a low price or beautiful design.

In contrast, misleading online marketing

‘tricks’ consumers into buying, subscribing, etc through the design of the online environment.

ACM urges businesses to have a system

in place to test compliance and incorporate

checks and balances into their daily operations.

(5)

Compliance should not be an afterthought.

Moreover, compliance should not just be a task for the compliance officer. Marketers and others involved with selling the product to consumers also have a role to play in compliance. By being aware of the rules, establishing good checks and balances, and using data for consumer protection, marketers can persuade consumers to buy their goods, rather than mislead them into doing so.

Concluding this paragraph on the legal framework, it is important to stress that this article focuses on the application of consumer protection law. This does not preclude the application of other areas of law, such as data protection, to online marketing practices. Furthermore, national consumer legislation can differ and should be consulted before making changes in web design based on this article.

THE DANGERS OF UNETHICAL DESIGN VS THE BENEFITS OF ETHICAL DESIGN

Why should marketers be interested in ethical design? For a start, deliberately misleading consumers is unethical as well as illegal and harmful. Furthermore, misleading marketing practices can damage the market by hurting competition, insomuch as these practices can cause firms to compete on deceptive marketing claims rather than on quality and prices. 13 Additionally, being penalised by the authorities will have negative financial as well as reputational consequences.

But even if a business is not fined by the authorities, misleading customers is still likely to have negative effects on consumer trust and sales, especially in the long term. Research indicates that lower perceived ethicality of businesses is related to lower consumer trust, satisfaction and commitment. 14,15 Consumers may not realise they are being misled, perhaps not right away or perhaps not at all. However, as consumers become more experienced with online

shopping and associated persuasion, they are likely to become increasingly effective in recognising deception. Furthermore, as Paul Boag demonstrates, dark patterns may seem to work, but they will hurt businesses in the long term due to the substantial hidden costs relating to customer service, maintenance, support, return processing fees and social media backlash. 16

However, if the business behaves ethically, customers are more likely to feel satisfied and trust the company. They will then be more likely to purchase more, come back more frequently, and recommend the business to others. In short, ethical business is not only the right thing to do, but it is also a good business strategy. 17

INFORMATION ASYMMETRY,

COGNITIVE BIASES AND CONSUMER DATA

Three elements that play a role in the relationship between sellers and consumers are information asymmetry, cognitive biases, and consumer data (see Figure 1). In and of themselves, these elements are not problematic. There is the possibility of using them to the consumers’ benefit, but they can also be used to mislead consumers. The following sections define the line between persuasion and deception for each of the three elements.

Information asymmetry

When there is information asymmetry, the buyer and the seller do not have the same amount of information about a product.

Usually, the company selling the product has the most information. 18 The company knows how the product is made, how well it functions, how easy it is to repair, and other specifics about the product. The consumer may have some information about how well a product functions if there are enough valid reviews, but still has much less information than the company.

Especially with experience goods and

(6)

Figure 1: Elements influencing the relationship between sellers and consumers

credence goods, the consumer has certain information about the quality of the product only after it has been used — and perhaps not even then 19 (eg the validity of a sustainability claim).

According to economists, information asymmetry is not just bad for consumers, but for the entire market. If people cannot distinguish between good products and bad products, they will not be willing to pay above a certain price, even if the good products are worth more. Sellers with good products will leave the market, because they cannot get the price they should get for their products, leaving only the bad products on the market, with the entire market potentially collapsing as a result. 20 Information asymmetry exists between consumers in both online and offline companies. However, this disparity is enhanced in online markets. 21

For a well-functioning market it is therefore important that consumers are clearly presented with all relevant information about a product. According to European consumer law, consumers must be presented with the information they need to make an informed choice.

Cognitive biases and irrationality Simply having the information somewhere in the online environment is not enough.

The average consumer, like all humans, is subject to biases and irrationality that limit the extent to which they process information. Behavioural science has shown that humans are boundedly rational:

heuristics and cognitive biases influence decision-making as humans are unable to take all relevant information into account within an acceptable timeframe. 22

Using heuristics, or mental shortcuts, can be very useful, because it saves time and mental capacity, and can lead to good decisions. For example, when a consumer is searching for a new phone and there are many options to choose from, it is more efficient to choose a brand the consumer has owned previously or that a friend has recommended than to analyse all options to find the best possible choice. In this sense, it is actually rational not to analyse all possible information. 23

However, these mental shortcuts and

lack of information search also leave

people vulnerable to biases. Scientists

have uncovered many types of biases. For

illustration purposes, only a few will be

(7)

discussed here, but for an extensive discussion of human decision-making and biases, see for example ‘Thinking, Fast and Slow’ by Daniel Kahneman. 24 Table 1 provides a non- exhaustive list of biases used in e-commerce, illustrated with concrete examples of their abuse. Most biases can be misused in many ways and the examples in the table are for illustration purposes rather than all-encompassing.

When biases are misused to mislead consumers in the design of a website, this is known as a dark pattern (regardless of the intention behind the design). Mathur and his colleagues define dark patterns as

‘user interface design choices that benefit an online service by coercing, steering, or deceiving users into making decisions that,

if fully informed and capable of selecting alternatives, they might not make’. 25

Dark patterns are harmful because they have an impact on people’s welfare, regarding their finances, privacy, and the extra cognitive effort needed to make a choice or perform an action. They also have an impact on people’s autonomy, because it is more difficult for consumers to make free and informed decisions. Finally, they have a negative impact on collective welfare, because they hinder competition and price transparency, and erode trust in the market. 26

Dark patterns turn legitimate marketing strategies into deception. For example, it can be a legitimate marketing strategy to have a countdown timer showing customers how long they have until a deal expires. However,

Table 1: Common biases deployed in marketing with their definition and an example of abuse

Bias Definition Examples of bias misuse

Anchoring People use the first piece of infor- mation they receive as a baseline, even when that information is irrelevant to the decision

27

Advertising with a discount when the ‘original’ price was nev- er the real price. People use the original price as anchor and then the discount seems like a bargain. For example, ACM fined Seats and Sofas for deceptive pricing.

28

Bandwag- on effect

People want something more because others are buying it

29

False activity messages or fake reviews. For example, ACM ordered Bicep Papa to stop using fake likes and followers.

30

Default People stick with the default

option

31

The box for an extra product or service is already checked (eg cancellation insurance when buying plane tickets). For example, ACM fined World Ticket Center for, among other things, pre-checked boxes for travel insurance and cancella- tion insurance.

32

Framing The way in which the same infor- mation is presented can result in different choices, depending on how the information is presented

33

Framing subscription cessation as a poor decision. For example, the Norwegian Consumer Council accused Amazon of this type of framing (confirmshaming) to stop people from unsubscribing from Amazon Prime.

34

Salience People tend to focus on things that grab their attention

35

Making the option to unsubscribe small and inconspicuous, whereas the benefits of being subscribed take up most of the page. For example, the Norwegian Consumer Council accused Amazon of using this kind of visual interference to stop people unsubscribing from Amazon Prime.

36

Scarcity bias

People value a product that is scarce more than a product of which there is an abundance

37

False claims about, for example, the number of rooms avail- able on a hotel booking website. For example, after action by the ACM and other European consumer authorities, Booking.

com has committed to, among other things, removing false scarcity claims from its website.

38

Sunk cost fallacy

People continue to invest in some- thing they have already invested in, for example in time or money, even when it would be better to stop

39

Adding unavoidable service costs at the very end of the

buying process (drip pricing). For example, the Netherlands

Consumer Authority (one of ACM’s predecessors) fined Rya-

nair for, among other things, not including unavoidable costs

in its advertised prices.

40

(8)

it turns into a dark pattern and is considered misleading when the deal does not really end at that point, creating a false sense of product or service scarcity.

Dark patterns are often a combination of multiple characteristics and can exploit one or multiple biases. The characteristics of the dark pattern will be influenced by the specific context in which it is used.

Dark patterns consist of at least one of the following characteristics 41 :

• asymmetric: when certain options (ie the ones that the seller wants the consumer to choose) are more obvious than others, for example, making the opt-in button to the newsletter prominently visible and the opt-out button light grey;

• covert: when the customer does not realise their choice is being influenced, for example with a trick question;

• deceptive: when it makes customers believe something other than the truth, for example when it seems like an offer is only available for a limited time when the product is really always available for that price;

• hidden information: when it does not show the consumer relevant information, such as only showing extra costs that come with a product at the last minute;

• restrictive: when there is only a limited set of options available, for example customers only being allowed to browse a web shop if they make an account.

These five characteristics have been narrowed down to two types of dark

patterns: dark patterns based on modification to the decision space and dark patterns based on modification of the information flow. 42 In the former case, the website is designed in a way that influences consumer decisions. In the latter case, the information presented on the website influences consumer decisions.

For an extensive discussion of dark patterns, see the Federal Trade Commission’s

workshop, ‘Bringing Dark Patterns to Light’. 43

Example: Misleading design

Design can be misleading in multiple ways including, but not limited to, drawing attention to the choice the company wants customers to make, presenting the most expensive products first, and by default checking boxes that add extra costs. These design choices take advantage of the fact that consumers are more likely to make the choice that is easiest rather than look carefully through the whole website.

In 2015, the ACM fined World Ticket Center for unfair commercial practices. 44 In this case, the company was criticised for both misleading information (additional mandatory costs were mentioned later in the decision process) as well as misleading design (checkboxes for cancellation insurance and travel insurance were ticked by default).

Example: Information obfuscation An example of a dark pattern based on information is the use of scarcity cues. For example, when a hotel booking site adds the warning ‘Only one room left!’ when a hotel has only one room available for the period of interest to the consumer, then this constitutes a fair and useful warning. However, if there are plenty of rooms available, either via that booking site or elsewhere, then this is clearly misleading, as the customer does not have access to the same information that the hotel has, and has no way to verify whether the claim applies to that particular booking site or even if it is true at all.

In 2019, the UK Competition and Markets Authority took action against six hotel booking sites using various tactics to mislead consumers into booking. 45 Tactics used by the booking sites included some or all of the following:

• lack of clarity regarding the ranking of hotels —

hotels paid a commission to be higher up

(9)

in the site’s rankings, but this was not clear to consumers;

• advertising a lower cost than the final total cost — sites advertised prices that did not include mandatory fees, such as taxes;

• false discounts based on misleading price comparisons — sites presented the price of a basic room as ‘discounted’, when in fact it was simply cheaper than a more luxurious room;

• misleading claims about room availability

— sites claimed other customers were looking at the same hotel, implying that they were looking at the same dates as the consumer, when this was not necessarily the case.

Consumer data

In addition to understanding the biases to which consumers are subject, online sellers often also know a lot about their potential customers. Online sellers are able to collect a lot of data about their potential buyers, for example through customer accounts, and can buy data from third parties too. They can also use A/B testing to study which user interface design works best to increase conversion.

With tools such as these, companies can personalise online shopping environments to different types of customers and make adjustments based on what their data suggest will work best, in a way that is impossible in an offline store.

For example, a company may use targeted advertising towards a certain group that is more likely to be interested in a specific product, such as advertising a local restaurant to people who live in the area. A company can also use data about certain consumers’

preferences and previous buying behaviour to target them even more specifically; for example, advertising computer components to people who have recently bought a computer. These tactics are respectively known as segmentation and personalisation.

Segmentation is defined as the ‘process of identifying segments of the market and the

process of dividing a broad customer base into sub-groups of consumers consisting of existing and prospective customers’. 46 Personalisation is defined as ‘the ability of a company to recognise and treat its customers as individuals through personal messaging, targeted banner ads, special offers on bills, or other personal transactions’. 47

In short, segmentation is targeting specific groups within a population and personalisation is making something specific to a person based on the information a company has about that person.

Subcategories include targeted advertising and personalised pricing. This is allowed and can be beneficial to consumers, if done safely and transparently, because they get more relevant advertisements or get good suggestions for entertainment, such as movies they might like. 48 However, it becomes problematic when information about consumers is used to exploit their susceptibility or vulnerability towards certain practices to manipulate them into making choices that they would not have made otherwise; for example, when data are used to test which consumers are most susceptible to social proof and then presenting those customers with false information about what others think of the product.

It should also be clear to the consumer when targeted advertising or pricing is happening. For example, some consumers may be asked to pay a higher price for a product than others. This is allowed, if it is clear to the consumer that it is the case. If the consumer does not know that they are paying a different price to other people, this could be considered an unfair commercial practice.

Example: Consumer data transparency

The International Consumer Protection

and Enforcement Network is a network of

consumer protection authorities from over

65 countries. In 2018, it conducted a broad

investigation into terms and conditions for

(10)

digital products and services. This led to an open call to businesses to adopt best practices with regard to terms and conditions, but also led to further actions towards specific businesses. 49

One of those follow-up actions was a joint intervention in 2019 by 27 consumer authorities, led by those in the Netherlands, the UK and Norway. This endeavour sought to get Apple and Google to make provision in their respective app stores for developers to inform consumers about their apps’

privacy policies and how they handled customer data. This information is necessary for consumers to make an informed choice when downloading apps. 50

In 2020, Apple agreed to add a privacy policy segment to the product pages for apps and to include a summary of the most important points from the app’s privacy policy. App providers are now obliged to include such information on the download pages for their apps. 51,52 Google will follow in 2022. 53 This benefits companies with good privacy policies as it can give them a competitive advantage over companies with less fair privacy policies. It is also good for consumers, who are now able to make better informed decisions about which apps to download.

VULNERABLE CONSUMERS

The above elements do not exist in isolation only — they can also interact. For example, companies can misuse consumer data by studying which consumers are more susceptible to certain cognitive biases and then targeting those consumers with specific dark patterns.

When companies personalise their websites, advertising or pricing to specific groups, they should not take advantage of those groups’ vulnerabilities. Indeed, they have an obligation to ensure that their practices do not abuse those vulnerabilities.

According to ACM, companies that have more data about their consumers also have

a responsibility to use said data to protect those consumers. In what follows, the article explores how companies can best do this.

CONSUMER PROTECTION

The first step is to know what the law prescribes regarding online shopping. It is essential that online businesses familiarise themselves with the European Union’s consumer legislation and the legislation of the specific countries in which they operate.

But how can companies check whether their web design is ethical? The following advice draws on the ACM’s guidelines on the protection of the online consumer 54 to encourage companies and marketers to study the effects of their online environment on consumers.

Measuring for ethical design and information comprehension

The way an online environment is designed can influence consumer decisions. It is understandable that companies want to design their websites in a way that shows off their products, informs consumers about the desirable qualities, and shows consumers other products they might like too. When designing their website, companies should focus on presenting their products honestly rather than tricking consumers into buying them. When companies are designing their website, they should ask themselves whether the design and the information they present misleads consumers into taking decisions they otherwise would not.

Sometimes it is clear that a design choice will mislead a consumer, for example by lying about availability or making it much easier to sign up for a subscription than to cancel it. Sometimes, however, it is less clear.

To study whether this is the case, marketers

and others within a company can use the

data they are collecting anyway to study

whether consumers’ choices change when

they change the design. Many companies

already use such A/B testing to improve

(11)

conversion, but it can also be used for compliance.

When designing such a test it is important to think beforehand about what design change to test and what the goal of that design change is. For example, is the goal for consumers to be informed about the different options or for them to understand the cancellation policy? It is also important to think beforehand about how to measure whether that goal has been achieved. For example, it could be measuring whether consumers make different purchases, but it could also be a change in how long people stay on a certain page, whether they click a specific option, etc. Think through which behaviour says something about the question the study is trying to answer.

One way to test whether a design choice influences consumers’ decisions could be to establish a baseline. Design always influences people’s choices, so it is not possible to establish a truly neutral baseline.

However, it is possible to take the current situation as the baseline and test how design changes affect behaviour. For example, a company might be receiving many complaints about how hard it is to cancel a subscription. Making it unnecessarily difficult to cancel a subscription may be considered an unfair commercial practice, so the company needs to find a way to simplify the process.

First, the company would want to know how often consumers cancel with the current design. Once this is established, it would need to decide how to simplify the process. For example, if people have to confirm that they want to cancel three times, reduce the number of confirmations. Keep the rest of the design as it was to ensure changes in cancellation rate do not come from other design changes. Then, after the change is implemented, retest how many people are cancelling their subscription. Do more consumers cancel? If so, it is likely that the change in behaviour is caused by the change in the design.

Similarly, companies might want to test two different designs to study whether one design changes behaviour more than or differently from the other. Then, one of the two designs should be implemented and the outcome measured, after which the design change should be reversed and the second design change implemented and the results measured again. The company can then compare the results on the chosen outcome measure for the two design changes.

While doing such tests, however, be aware that consumer protection legislation always governs one’s marketing practices. At any time, any marketing practice, even a live version of an A/B test, must abide by the law, meaning that it cannot mislead consumers.

In addition to the question whether the online environment’s design is unfairly influencing consumers’ choices, businesses might also want to ask themselves whether consumers understand the information that is presented to them online. A business may not meet the legal requirements for providing consumers with essential information if the information is provided in ambiguous language or unclear or lengthy texts. Equally, disclosure requirements may not be met if the design of the information hides it from consumers through, for example, difficult-to-read fonts or colours, out-of-sight positioning of the information, or hiding the information in deep links.

It is a good idea to test whether consumers understand the important elements of the transaction, such as the return policy, the total costs, etc.

Example: A/B testing for consumer protection

ACM conducted a natural experiment with data from a platform that added a label that told consumers certain results were sponsored. 55 Suppliers could choose to pay a higher commission to the platform to end up higher on a ranking. The label was meant to inform consumers about this.

The platform had added the label on the

(12)

website, but not (yet) on its app. However, the ranking was the same on the website and the app. This meant it was possible to compare the effect of a design choice meant to improve transparency, namely the sponsored label. On the app there was no transparency and on the website there was more

transparency. The question is whether that transparency changed consumers’ behaviour.

In this natural experiment, ACM compared the purchases consumers had made from sponsored suppliers on the website (with the transparency label) versus the purchases made from sponsored suppliers on the app (without the transparency label). The study identified no statistically significant differences in purchasing behaviour between the two and thus that this design feature, which was meant to add transparency, did not change purchasing behaviour. While further research must be conducted to understand why consumers’

purchase behaviour did not change as a result of the information provided, it may be that consumers did not identify or understand the information. If this were indeed the case, it could indicate that the disclosure fails to meet the legal requirements.

DISCUSSION AND CONCLUSION This article has argued that companies and consumers benefit from ethical design. It will close with recommendations on how to enhance ethical design for companies, policy makers and researchers. Within any company, it is important that the company culture drives ethical design. For example, are targets and bonuses for employees set in such a way that they incentivise fair design?

Do employees feel they are in a position to point out unethical design choices without negative consequences? Are marketers and user experience designers being trained in the legal limitations of the commercial practices they develop and deploy?

Correspondingly, is the compliance officer or department involved in the early steps

of design? The business community at large can also strive for the adoption of an ISO standard on fair design and online influence.

More research on dark patterns is

recommended. Although there is a clear and general understanding that dark patterns and other unethical design choices are harmful to consumers, there is still little empirical research available on the harmfulness of specific practices. Further knowledge about specific harms will help policy makers and supervisory bodies to design better policies and interventions. Researchers, supervisors and companies can work together on such questions that are both theoretically interesting and practically relevant.

In conclusion, especially online, companies have a lot of power over consumers, which, consciously or unconsciously, they can wield to mislead consumers into making decisions they otherwise would not have made. This is not only illegal, but in the long run will also cost companies customers, money, and their reputation. Companies can avoid these negative outcomes by knowing the law and working towards fair outcomes for consumers, for example by using customer data for consumer protection rather than solely for conversion optimisation. Through thoughtful, ethical design, companies can create a fair online shopping environment that is good for them and their customers.

AUTHORS’ NOTE

Elianne van Steenbergen was involved in writing this article in her role as researcher at Utrecht University, not in her capacity as a supervisor at the Dutch Authority for the Financial Markets.

References

1. Forbrukerrådet (2021) ‘You can log out, but you can never leave: How Amazon manipulates consumers to keep them subscribed to Amazon Prime’, avail- able at: https://fil.forbrukerradet.no/wp-content/

uploads/2021/01/2021-01-14-you-can-log-out-

but-you-can-never-leave-final.pdf (accessed 6th July,

2021).

(13)

2. Day, G. and Stemler, A. (2020) ‘Are dark patterns anticompetitive?’, Alabama Law Review, Vol. 1, No. 1, pp. 1–45.

3. Mathur, A., Mayer, J. and Kshirsagar, M. (2021) ‘What makes a dark pattern … dark? Design attributes, nor- mative considerations, and measurement methods’, Arxiv, available at: https://arxiv.org/abs/2101.04843 (accessed 6th July, 2021).

4. Autoriteit Consument & Markt (2020) ‘ Guidelines on the protection of the online consumer:

Boundaries of online persuasion’, available at: https://

www.acm.nl/sites/default/files/documents/2020-02/

acm-guidelines-on-the-protection-of-the-online- consumer.pdf (accessed 6th July, 2021).

5. Autoriteit Consument & Markt (2020) ‘ Booking and Expedia inform consumers more clearly about their offers following action of European consumer authorities’, press release, 18th December, available at: https://www.acm.nl/en/

publications/booking-and-expedia-inform-con- sumers-more-clearly-about-their-offers-follow- ing-action-european-consumer-authorities (accessed 6th July, 2021).

6. European Commission (2005) ‘Directive 2005/29/

EC of 11 May 2005 on unfair business-to-consumer commercial practices in the internal market and amending Council Directive 84/450/EEC, Direc- tives 97/7/EC, 98/27/EC and 2002/65/EC of the European Parliament and of the Council and Regula- tion (EC) No 2006/2004 of the European Parliament and of the Council’, available at: https://eur-lex.

europa.eu/legal-content/EN/TXT/PDF/?uri=CEL- EX:32005L0029&from=EN (accessed 6th July, 2021).

7. Ibid.

8. Ibid.

9. Ibid.

10. Ibid.

11. Ibid.

12. Ibid.

13. Lindsey-Mullikin, J. and Petty, R.D. (2011) ‘Market- ing tactics discouraging price search: Deception and competition’, Journal of Business Research, Vol. 64, No. 1, pp. 67–73.

14. Leonidou, L.C., Kvasova, O., Leonidou, C.N. and Chari, S. (2013) ‘Business unethicality as an imped- iment to consumer trust: The moderating role of demographic and cultural characteristics’, Journal of Business Ethics, Vol. 112, No. 3, pp. 397–415.

15. Romβn, S. and Ruiz, S. (2005) ‘Relationship out- comes of perceived ethical sales behaviour: The cus- tomer’s perspective’, Journal of Business Research, Vol. 58, No. 4, pp. 439–445.

16. Boag, P. (2020) ‘Click! How to Encourage Clicks Without Shady Tricks’, Smashing Media, AG Freiburg.

17. McMurrian, R.C. and Matulich, E. (2016) ‘Building customer value and profitability with business ethics’, Journal of Business & Economics Research, Vol. 14, No. 3, pp. 83–90.

18. Mascarenhas, O.A., Kesavan, R. and Bernacchi, M. (2008) ‘Buyer–Seller Information Asymmetry:

Challenges to distributive and corrective justice’, Journal of Macromarketing, Vol. 28, No. 1, pp. 68–84.

19. Ibid.

20. Akerlof, G.A. (1970) ‘The market for “lemons”:

Quality uncertainty and the market mechanism’, Quarterly Journal of Economics, Vol. 84, No. 3, pp. 488–500.

21. Dimoka, A., Hong, Y. and Pavlou, P.A. (2012) ‘On product uncertainty in online markets: Theory and evidence’, Management Information Systems Quarterly, Vol. 36, No. 2, pp. 395–426.

22. Kahneman, D. (2011) ‘Thinking, Fast and Slow’, Farrar, Straus and Giroux, New York, NY.

23. Siciliani, P., Riefa, C. and Gamper, H. (2019) ‘Con- sumer Theories of Harm: An Economic Approach to Consumer Law Enforcement and Policy Making’, Hart Publishing, Oxford.

24. Kahneman, ref. 22 above.

25. Mathur, A., Acar, G., Friedman, M.J., Lucherini, E., Mayer, J., Chetty, M. and Narayanan, A. (2019)

‘Dark patterns at scale’, Proceedings of the ACM Human-Computer Interaction, Vol. 3, No. CSCW, pp. 1–32.

26. Ibid., ref. 3 above.

27. Tversky, A. and Kahneman, D. (1974) ‘Judgment un- der uncertainty: Heuristics and biases’, Science, Vol. 185, No. 4157, pp. 1124–1131.

28. Autoriteit Consument & Markt (2018) ‘ACM fines furniture retailer Seats and Sofas for deceptive pric- ing’, press release, 7th June, available at: https://www.

acm.nl/en/publications/acm-fines-furniture-retail- er-seats-and-sofas-deceptive-pricing (accessed 6th July, 2021).

29. Leibenstein, H. (1950) ‘Bandwagon, snob, and veblen effects in the theory of consumers’ demand’, Quarterly Journal of Economics, Vol. 64, No. 2, pp. 183–207.

30. Autoriteit Consument & Markt (2021) ‘ACM forces online store to stop using fake likes and fake follow- ers’, press release, 19th January, available at: https://

www.acm.nl/en/publications/acm-forces-online- store-stop-using-fake-likes-and-fake-followers (accessed 6th July, 2021).

31. Johnson, E.J., Hershey, J., Meszaros, J. and Kunreuther, H. (1993) ‘Framing, probability distortions, and insur- ance decisions’, Journal of Risk and Uncertainty, Vol. 7, No. 1, pp. 35–51.

32. Autoriteit Consument & Markt (2015) ‘ACM has fined World Ticket Center for displaying airfares incorrectly’, press release, 12th February, available at: https://www.acm.nl/en/publications/publica- tion/13848/ACM-has-fined-World-Ticket-Center- for-displaying-airfares-incorrectly (accessed 6th July, 2021).

33. Tversky, A. and Kahneman, D. (1981) ‘The framing of decisions and the psychology of choice’, Science, Vol. 211, No. 4481, pp. 453–458.

34. Ibid., ref. 1 above.

35. Kahneman, D., Slovic, P. and Tversky, A. (1982)

‘Judgment under Uncertainty: Heuristics and Biases’, Cambridge University Press, Cambridge.

36. Ibid., ref. 1 above.

(14)

37. Lynn, M. (1989) ‘Scarcity effects on desirability:

Mediated by assumed expensiveness?’, Journal of Economic Psychology, Vol. 10, No. 2, pp. 257–274.

38. Autoriteit Consument & Markt (2019) ‘Booking.

com commits to adjusting its website following action by European consumer authorities’, press re- lease, 20th December, available at: https://www.acm.

nl/en/publications/bookingcom-commits-adjusting- its-website-following-action-european- consumer- authorities (accessed 6th July, 2021).

39. Arkes, H.R. and Blumer, C. (1985) ‘The psychology of sunk cost’, Organizational Behavior and Human Decision Processes, Vol. 35, No. 1, pp. 124–140.

40. Autoriteit Consument & Markt (2013) ‘Netherlands consumer authority fines Ryanair’, press release, 20th March, available at: https://www.acm.nl/en/publi- cations/publication/11254/Netherlands- Consumer- Authority-fines-Ryanair (accessed 6th July, 2021).

41. Ibid., ref. 25 above.

42. Ibid., ref. 3 above.

43. US Federal Trade Commission (2021) ‘Dark Pat- terns Workshop, 29th April’, available at: https://

www.ftc.gov/news-events/audio-video/video/

dark-patterns-workshop (accessed 7th May, 2021).

44. Ibid., ref. 32 above.

45. Competition and Markets Authority (2019) ‘Hotel booking sites to make major changes after CMA probe’, press release, 6th February, available at:

https://www.gov.uk/government/news/hotel-book- ing-sites-to-make-major-changes-after-cma-probe (accessed 6th July, 2021).

46. Camilleri, M.A. (2018) ‘Market segmentation, target- ing and positioning. In: ‘Travel Marketing, Tourism Economics and the Airline Product: An Introduction to Theory and Practice’, Springer, Cham, pp. 69–84.

47. Imhoff, C., Loftis, L. and Geiger, J. (2001) ‘Building the Customer-Centric Enterprise, Data Warehousing Techniques for Supporting Customer Relationship Management’, John Wiley & Sons, New York, NY.

48. Ucuzoglu, J. and Hagel III, J. (2020) ‘Use your cus- tomer data to actually help your customers’, Harvard

Business Review, 24th April, available at: https://hbr.

org/2020/04/use-your-customer-data-to-actual- ly-help-your-customers (accessed 6th July, 2021).

49. International Consumer Protection and Enforcement Network (2018) ‘Joint open letter to businesses in the digital economy on the importance of standard terms and conditions for consumers’, 29th June, available at: https://icpen.org/news/902 (accessed 6th July, 2021).

50. Autoriteit Consument & Markt (2019) ‘ACM takes lead in international call on Apple and Google about app stores’, press release, 12th March, available at:

https://www.acm.nl/en/publications/acm-takes- lead-international-call-apple-and-google-about-app- stores (accessed 6th July, 2021).

51. Autoriteit Consument & Markt (2020) ‘Apple informs consumers about the data use of apps fol- lowing call from regulators worldwide’, press release, 8th December, available at: https://www.acm.nl/en/

publications/apple-informs-consumers-about-da- ta-use-apps-following-call-regulators-worldwide (accessed 6th July, 2021).

52. Competition and Markets Authority (2020) ‘CMA and global partners secure privacy changes to the App Store’, press release, 8th December, available at:

https://www.gov.uk/government/news/cma-and- global-partners-secure-privacy-changes-to-the-app- store (accessed 6th July, 2021).

53. Autoriteit Consument & Markt (2021) ‘Google to require providers to add information about data use to apps in its app store’, press release, 7th May, available at: https://www.acm.nl/en/publications/

google-require-providers-add-information-about- data-use-apps-its-app-store (accessed 6th July, 2021).

54. Ibid., ref. 4 above.

55. Autoriteit Consument & Markt (2021) ‘Sponsored ranking: An exploration of its effects on consumer welfare’, available at: https://www.acm.nl/sites/de- fault/files/documents/sponsored-ranking-study-acm.

pdf (accessed 6th July, 2021).

Referenties

GERELATEERDE DOCUMENTEN

preclude a plaintiif from establishmg that a reasonable the Northwestern Umversity School of Law durmg the ALI alternative design should have been adopted thdt would have sessions

Op deze noordgerichte helling van de zandrug tussen Brugge en Maldegem konden geen artefacten verzameld worden. Raakvlak adviseert voor deze locatie geen vervolgonderzoek, maar

Na de behandeling wordt u bewaakt op de uitslaapkamer en voordat u de ruimte mag verlaten wordt u eerst onderzocht zodat u in een optimale mogelijke conditie naar huis kunt...

[r]

Eight different persuasive pop-ups were developed varying on persuasion strategy (scarcity vs social proof), position (bottom left vs top left), exposure duration (4 seconds vs

To conclude on the first research question as to how relationships change between healthcare professionals, service users and significant others by introducing technology, on the

An increase in the world prices of exports and imports, due to the liberalisation of the food and agricultural commodity trade of the OECD countries, will likely benefit the

1) Whilst building and ship industries typically only produce unique products the airplane and car industry manufacture a great number of replicates. 2) All but the building and