Consumer data, targeted advertising and user choice

Most consumers know that the online platforms they use collect data about them to some extent. Some people welcome the personalisation of advertising as they are more likely to receive adverts that are relevant for them, and so they are comfortable with what they receive in return for giving up their data. However, many other people are uncomfortable about the amount of personal data that is collected and want to have greater control over its collection and use.

This report explains why a pro-competitive intervention to give consumers greater control over their data at Google and Facebook is necessary and presents the findings of our research estimating the value of the reduction in harm that would result from giving consumers more control.

Introduction

The business models of some of the world’s biggest companies are based upon digital advertising. In the UK, Google had more than a 90% share of the market for search advertising in 2019, while Facebook had more than half of the market for display advertising [1]. The ability of these firms to collect masses of data about individuals allows them to target or personalise the adverts shown to a consumer, and so charge higher prices to businesses for that advertising space. 

Most consumers know that the online platforms they use collect data about them to some extent [2].  Some welcome the personalisation of advertising as they are more likely to receive adverts that are relevant for them, and so they are comfortable with what they receive in return for giving up their data. However, many people are uncomfortable about the amount of personal data that is collected about them and want to have greater control over its collection and use.

The UK government is currently consulting on the powers to be given to the Digital Markets Unit(DMU), which will promote competition in digital markets. The proposals include giving the DMU the ability to make interventions that would increase consumers’ control over their personal data. Such interventions would be pro-competitive, since the collection of large amounts of consumer data is a source of competitive advantage for the largest online platforms, but they would also reduce harm for consumers who would value more control. This report explains why apro-competitive intervention to give consumers greater control over their data at Google and Facebook is necessary and presents the findings of our research estimating the value of the reduction in harm that would result from giving consumers more control.

The harm caused by excessive data collection

Which? research has consistently found that consumers feel disempowered in their dealings with digital firms because there is a lack of transparency about what data is collected and how. Consumers feel particularly strongly about the collection and use for targeting of third-party(or ‘off-site’) data, which is collected about a user from or through another organisation. We found that a third of users (34%) were unaware that Facebook tracks user activity on other websites and apps, and eight in ten (79%) were unaware that Facebook matches profiles to customer lists uploaded by companies [3]. When informed about it, many users were shocked about the extent of the data collection [4].

‘Oh my God... it’s basically everyone... everything’ 

- Male, aged 37 years

The consequence of the lack of transparency is that people do not feel they have given their informed consent for their data to be collected and used to target advertising at them. Only 6% of users felt that they had given informed consent for Facebook to track them on other websites/apps, and only 5% felt they had given it for their profile to be matched to customer lists [5]. This frequently gives rise to feelings of frustration and annoyance [6].

 ‘I don’t like it, and it’s not with my knowing consent. I might have clicked a box in the terms of agreement thing [for organisations] somewhere, we have to do that to do anything online, but it doesn’t mean I really consent.’

- Female, aged 63 years

As a result, many people feel that these methods of data collection for targeting are not acceptable. We found that half of Facebook users (53%) thought it was not acceptable for their profile to be matched to a customer list in any circumstances, and 58% thought that all third-party website and app tracking was unacceptable [7].

This undermines the legitimacy of targeted advertising. We have repeatedly found that as people become more knowledgeable about the way in which data is collected and used to personalise the adverts that they are shown, then they become less comfortable with it. In one survey, 52% of Facebook users initially told us that they were happy to see targeted advertisements. After they were informed about data collection methods and had engaged with their settings to see the list of third-party organisations from which Facebook collected data about them, this fell to just 37% [8].

The consequence is that many people are reluctant users of services that they feel they cannot do without, having little choice but to accept the status quo as ‘the price you pay [9]'. It has led to an unhealthy situation with low levels of consumer trust in some of the largest tech firms, but with consumers unable to choose different services. Just 12% of UK consumers in August 2021 said they trust social networks, whilst 55% do not trust them, giving a net value of -43. This is considerably lower than other sectors, such as online marketplaces (+16) and broadband/home phone services (+22) [10]. In a recent survey that asked consumers about a wide range of brands, we found Facebook was the only brand for which more of its own users distrust it to treat them fairly than trust it to do so [11].

The limits to competition

One way in which these problems would be resolved is if consumers could switch to other platforms that would give them more control over their data, either by providing a service without taking so much personal data or by giving greater compensation to users for their data. Such companies do exist. For example, the search engine DuckDuckGo does not use personalised advertising, while the start-up browser extension Gener8 [12] allows users either to limit how much they are tracked online, or to share their data and be rewarded for it as advertising revenue from the targeted adverts the consumer sees is passed back to them.

However, while these companies offer alternatives for consumers who want greater control over their personal data, the nature of digital markets means that they can only exert limited pressure on Facebook and Google. Strong network effects and large economies of scale and scope give the largest platforms enduring market power. This is further enforced by the amount of consumer data collected by these platforms as it gives a competitive advantage when improving the quality of their own service and generates greater revenue through, for example, ever more sophisticated targeting of advertising [13]. As these markets currently function, it is impossible to envisage how competitors with different, less privacy-intrusive business models could grow sufficiently to resolve the harm currently being experienced by consumers in markets that rely on digital advertising for revenue.

The largest platforms have taken some steps to give their users greater control over how their data is used, but it is clear that these have not been sufficient for their users to be comfortable with how their personal data is collected and used to target advertising. This variously reflects that the level of control is insufficient, that users find it hard to execute and that default settings are likely to lead to greater collection of data than many consumers want. Facebook allows users to control some elements of data collection, for example, the Off-Facebook Activity tool allows users to monitor and control online tracking. However, our research found that only 11% of Facebook users were aware of the ‘off-Facebook activity’ setting, and users do not have the power to turn off personalised advertising [14]. Google offers users more control as they can choose not to receive personalised adverts, but concerns have been raised relating to the complexity faced by consumers wanting to exercise this choice [15]. Our research found that only 17% of Google users had changed their settings, and some of these clearly demonstrated they did not understand the choice they had made [16].

The failure of competition to provide consumers with adequate means to control how much data they share means that other measures are needed to ensure that all consumers are given an easily executable, informed choice about how much of their personal data is collected and how it is used.

Making informed choice easier

The Competition and Markets Authority (CMA) recommended interventions that would make it easier for consumers to exercise control over their personal data. This included a Choice Requirement Remedy (CRR) that would require platforms to give consumers the choice not to share their data for personalised advertising. The right to choose not to receive targeted adverts would mean that a platform should have an obligation to provide its core service without collecting any data beyond that which is necessary for the performance of the contract with the consumer. This pro-competitive intervention, to be overseen by the DMU, would only apply to the largest digital platforms assessed for whom the acquisition of customer data is a source of competitive advantage. The platforms must have been assessed as having substantial, entrenched market power in online advertising that provides it with a strategic position, and the expectation is that this would apply to Google and Facebook.

In addition to the potential competition benefits that such a demand-side remedy could bring, it would also address the harm felt by consumers when they feel disempowered in their dealings with these platforms and when they feel too much of their personal data is being taken for what they receive in return. To investigate how much benefit this would bring to consumers, Which?, alongside Accent and PJM Economics, conducted research to estimate the value that users of Google and Facebook would place on the ability to have greater control over their personal data that is used for targeted advertising [17].

In such cases in which there is no market value then so-called ‘stated preference’ studies allow avalue of a policy or action to be estimated. These studies are an established alternative method to value non-market goods [18]. They are used widely in many areas of public policy such as environmental, health, transport, and safety policy, and they can involve asking people either their willingness-to-pay or willingness-to-accept for a hypothetical option. 

In our study, we used both willingness-to-pay and willingness-to-accept methods and asked people to make hypothetical choices between receiving targeted adverts and sharing their data or receiving generic adverts. In the willingness-to-pay scenario, they were asked how much they would pay in the form of a small monthly fee to be able to only receive generic adverts, while in the willingness-to-accept scenario, they were asked how much they would need to receive in monthly rewards to accept being served targeted adverts. Willingness-to-accept methodologies are less commonly used in valuation studies since people tend to be less sensitive to gains in money, like rewards, than to losses, like a subscription fee, and this tends to lead to higher estimated values. However, in this instance, the willingness-to-accept rewards is a more plausible scenario than people paying a subscription fee to use Google or Facebook, and it is the scenario described by the CMA when it recommended the introduction of the CRR. Given this, we include both methods, but use the willingness-to-pay, fee scenario as the preferred, more conservative approach.

The research results support many of our findings from other work. More than 90% of Google and Facebook users know that the platforms collect data about them and that the advertising they see is targeted on the basis of this. Users varied in how useful they found the targeted adverts, but 50%of Facebook users and 45% of Google users said they were useful very or fairly often. Despite this, levels of comfort with sharing personal data with Google and Facebook for the purpose of targeted advertising remained fairly low. On a 1 to 10 scale in which 1 was not at all comfortable and 10 was completely comfortable, about two-thirds of people were more uncomfortable than comfortable (ie chose 5 or less), and about a fifth said they were not at all comfortable (chose 1).

The consequence of this is that many users would prefer not to receive targeted adverts. In the absence of any financial incentive, only 27% of Google and Facebook users said they would prefer to receive targeted adverts [19]. When a financial incentive is introduced to encourage users to choose targeted adverts, then the proportion of people who do so increases markedly. Of those whose choice involved a fee to receive generic adverts only, the proportion choosing targeted adverts increased to 81%. For those offered a reward to receive the targeted adverts, it increased to 55%. People were, unsurprisingly, more likely to choose targeted adverts when the financial incentive was larger. Overall, these findings demonstrate the diversity in preferences across platform users. Some hold clear preferences either to always receive targeted adverts or to never share their personal data. However, there are many who are currently uncomfortable with the current situation, but who would be accepting of data collection for targeted advertising if a financial incentive made this a more equitable trade-off.

In terms of how large this financial incentive would need to be, Google and Facebook users said they would be prepared to pay a fee of £1.09 per month on average to have the control offered by the CRR. Aggregating this estimate for all UK users of Google and Facebook for a full year gives a total annual estimate of the value of the CRR of £1.14 billion [20].

The rewards scenario, in which people were given the chance to receive financial compensation for accepting targeted adverts, gives even higher estimates of the value of the CRR. On average, the estimated monthly rewards for a person to choose to receive targeted adverts was £4.03. Aggregated across the UK this gives an annual total estimated value of the CRR of £4.21 billion. 

We believe these are conservative estimates of the value that Google and Facebook users would place on the CRR because:

• The study assumes that users who have already reviewed their data settings would receive no additional benefit from the introduction of the CRR. 35% of Facebook users told us they had reviewed their data preference settings, and 31% of Google users had done this. For all of these users, we assume their individual value of the CRR will be zero. This is despite the CRR providing a more transparent, more accessible, and wider choice between data-sharing options.
• Users aged under 18 were also given a zero valuation, but they are expected to value having greater control. In fact, the analysis indicates that younger users place a higher value on the CRR than users aged over 65, so users’ valuation of the CRR may increase in the future.
• The survey sample was obtained from an online panel who, on account of having signed up to reveal information about themselves for small amounts of money, may be more likely than the general population of users to be willing to share their data.

Conclusions and recommendations

The current levels of control that individuals have over the collection and use of their personal data by the largest online platforms has led to widespread discontent among consumers. Many are uncomfortable with their data being collected and used for targeted advertising, and they find the methods of data collection unacceptable. Consumers want to be able to make an easily executable, informed choice about how much of their personal data is collected and how it is used. 

Consumer control would be enhanced by the implementation of a choice requirement remedy for the largest platforms that serve targeted advertising. Our research has found that the potential benefit of this remedy is substantial. Depending on the methodology used, the average estimated value to consumers from being given more control is between £1.14 billion and £4.21 billion per annum. This is additional to the competition benefits that would arise from reducing this source of market power.

It is imperative that the government follows through with its intention to legislate for a new pro-competition regime for digital markets, and the Digital Markets Unit is empowered to introduce remedies that promote competition by giving consumers greater control over their personal data.

Footnotes

[1] Competition and Markets Authority (2020) Final report of the online platforms and digital advertising market study  ↑

[2] In a survey of Facebook users, 92% said they knew that data about what they do on the platform (known as first-party or ‘on-site’ data) was collected by the company (Which?, 2021, Are you still following me?)  ↑

[3] Which? (2021), Are you still following me?  ↑

[4] Which? (2020), Are you following me?  ↑

[5] Which? (2021), Are you still following me?  ↑

[6] Which? (2020), Are you following me?  ↑

[7] Which? (2021), Are you still following me?  ↑

[8] Which? (2021), Are you still following me?  ↑

[9] Which? (2020), Are you following me?  ↑

[10] Which?’s regular tracker survey of consumer sentiment, finances, trust and wellbeing. Fieldwork is carried out by Yonder on behalf of Which?. The sample is circa 2,000 UK consumers per wave, weighted to be nationally representative  ↑

[11] Which? Consumer Insight Tracker May 2021, Online Poll weighted to be nationally representative, carried out by Yonder with 2,091 respondents. Question text; How much do you trust or not trust the following companies to treat you fairly?  ↑

[12] Gener8  ↑

[13] Digital Competition Expert Panel, 2019. Unlocking digital competition, Report of the Digital Competition Expert Panel  ↑

[14] Which? (2021), Are you still following me?  ↑

[15] Competition and Markets Authority (2020) Final report of the online platforms and digital advertising market study  ↑

[16] Which? (2021) The value of the Choice Requirement Remedy. 42% of Google users who had changed their data preference settings incorrectly believed it would result in them seeing fewer adverts  ↑

[17] Which? (2021) The value of the Choice Requirement Remedy  ↑

[18] HM Treasury (2020) The Green Book  ↑

[19] This was 26% of Google users and 27% of Facebook users, with the difference not being statistically significant between the two platforms. Here, and throughout, the proportions are those for the respondents after they had been shown information about how Google and Facebook collect personal data from them. These informed choices form the basis of our analysis, but even prior to being told about data collection methods, 57% of Google and Facebook users had a preference to receive generic adverts only  ↑

[20] We estimate the total number of UK users aged over 18 to be 41.4 million for Facebook and 45.7 million for Google. This is based on the CMA’s estimate of Google, and Facebook’s UK reach and ONS population estimates  ↑

About

Which? is the UK’s consumer champion, here to make life simpler, fairer and safer for everyone. Our research gets to the heart of consumer issues, our advice is impartial, and our rigorous product tests lead to expert recommendations. We’re the independent consumer voice that works with politicians and lawmakers, investigates, holds businesses to account and makes change happen. As an organisation we’re not for profit and all for making consumers more powerful.