10/08/2020

Online targeting gets a bad rap – but without it we would be lost online.  Automated systems now make decisions about a significant proportion of the information people see online. Yet it is also clear that online targeting can involve vexed questions about consumer data privacy, unfair marketing practices and the relentless targeting of vulnerable groups of consumer, if not outright fraud. 

As background reading for the Australian Financial Review’s conference on the ‘distributed consumer’, we will be looking at the gathering concern of regulator’s over ‘adtech’.  We will also be comparing the differences in approach between  some regulators, reflected in its Digital Platforms Inquiry and now ‘operationalised’ in court proceedings which the ACCC has brought against Google, Health Engine and others.

The first article focuses on the UK approach to adtech.  The next article will compare the UK and other overseas regulators approach with the ACCC’s approach.

The UK Government’s Centre for Data Ethics and Innovation released a report on online targeting in February 2020:

The CDEI is good background reading on how the adtech industry works, and it takes a balanced perspective on the ‘good’ and the ‘bad’ about ‘adtech’.

The good…

Targeted advertising has been with us for a long time. In the offline world, targeting was based on the likely readers of particular print media or viewers of particular TV shows, rather than personal information. 

The CDEI says online targeting is different from offline targeting in 5 ways:

  • Data: platforms collect an unprecedented breadth and depth of data about people and their online behaviours, and analyse it in increasingly sophisticated ways.
  • Accuracy and granularity: content can be targeted accurately to small groups and even individuals.
  • Iteration: online targeting systems learn from people’s behaviour to constantly increase their effectiveness in real time.
  • Ubiquity: content can be targeted at scale and at relatively low cost.
  • Limited transparency: the ability to accurately match people with content inevitably limits the broader scrutiny of that content  as fewer people see each item of content, and don’t know much about what other users are seeing.

The CDEI recognised the benefits of online targeting:

Online targeting helps people to navigate the vast amount of information available online by showing them content it has predicted they will engage with. It can also broaden people’s horizons through content “discovery”, by suggesting content that they wouldn’t have sought out but which might be beneficial to them. For example, a study has found that 79% of all consumers, and 90% of those under 30, agree that streaming services that use online targeting systems play a huge role in their discovery of new video content

Online targeting can also be a ‘power of good’ because it can help consumers make better informed choices.  For example the UK’s Department for Transport achieved an 11% increase in young men who thought it was unacceptable to let a friend drive after drinking, following an online public awareness campaign targeting them.  THE CDEI also gave examples of targeting to identify and get help to young people displaying social vulnerability in their social media feeds.

But above all, online targeting is part of many companies’ core business models and allowed them to innovative services to consumers that are now regarded as an essential part of our lives

In its polling and ‘dialogue’ groups, the CDEI found that consumers do not want to prevent online targeting, but want higher standards of accountability, transparency and meaningful control:

  • polling shows broad support for a range of uses of online targeting, with 54% of respondents finding the personalisation of online adverts acceptable and 68% finding it acceptable to recommend music in apps.
  • Consumers saw the main benefit of online targeting as helping them navigate content online.  They thought that it could broaden their horizons by showing them content they wouldn’t otherwise have sought out. They thought it enhanced their ability to find like-minded individuals. And they thought that it had economic benefits for people and businesses, by helping people find relevant products and giving them targeted offers or discounts.

While endorsing these perceptions about the value of online targeting, the CDEI was also concerned that there was a low level of understanding amongst consumers about how online targeting worked and how widely it was used in their feeds. The CDEI noted that ‘[t]he survey research shows only 7% of people believe that information about who people interact with online could be used in online targeting systems..

The primary concern of dialogue participants was about the potential for online targeting systems to undermine people’s autonomy and influence their behaviours or attitudes.   Participants said they wanted more control over how their data is used for online targeting.  Most found the controls in online platforms hard to locate and use and in any event, only 36% trusted that the platform would use their data consistently with their expressed wishes through these controls.

The bad…

The main concern with online targeting is that it amplify consumers’ impulsive responses – their “first order preferences”. Offline the process to ‘buy’ (e.g. completing the printed credit card application and mailing it in) can be a form of consumer protection – it is a more reflective process with time to change your mind. For example, experiments have shown that online targeting platforms inferring psychological traits from social media data (such as extraversion) can result in a 40% increased likelihood of consumers clicking on ads and a 50% increased likelihood of buying.  

Research following the publication of the CDEI’s report showed just how blunt and overwhelming these tools can be in the high-cost lending market

  • Over the past two years there have been 13 billion adverts or other digital material for “payday loans”.
  • Debt consolidation brokers use targeted keywords such as “gambling debt”, “wiping debt”, “all credit types welcome” to attract vulnerable consumers with poor credit scores to view and apply for high-cost loans.
  • One provider conducted 430 separate ad-tech campaigns on one day;
  • Online forms that misleadingly look like online enquiry are actually application processes;
  • The ‘small print’ terms and conditions are usually positioned below the onscreen ‘fold line’.
  • The ‘cowboys are being out-cowboyed’ – vulnerable groups are at risk of unknowingly using unauthorised cloned firms and believing fake social media accounts on Twitter or Reddit and falling victim to fraud when applying for credit.

CDEI’s recommended way forward…

The CDEI concluded that the UK’s current regulatory regime does not adequately address the potential harms of online targeting – a view that an increasing number of governments and regulators have reached about their own regulatory regimes.

But what is more interesting is how the CDEI frames the objectives in searching for new solutions.  Its recommendations “..have been designed to enable people, businesses and society to benefit from the use of online targeting, while mitigating the key risks posed by online targeting systems across the themes of autonomy and vulnerability, democracy and society, and discrimination.”  Consistent with its broad remit to promote AI in the UK, CDEI says that “[o]ur proposals aim to enable growth and innovation in the use of online targeting, maximising the benefits of data and AI for the UK society and economy.”

While effective enforcement powers form part of the CDEI’s recommendations, overall its approach is weighted towards upfront or ‘ex ante’ measures to improve accountability and transparency.  This reflects the broad policy settings for AI and technology regulation in the UK recommended by the  Furman Report:

“The approach should combine participation and consultation with the scope for regulatory enforcement … It should only intervene where doing so is effective and proportionate to achieve competitive aims. Where this is the case, the Panel wants to introduce a system where industry has greater clarity and confidence over what constitutes acceptable practice and the rules that apply.”

The key recommendations made by the CDEI were:

  • An Online Targeting Code of Practice to put some boundaries around the targeting techniques. The code should be systemic, focused on the processes organisations use to target content. Platforms would need to record the purpose and desired outcomes of the online targeting systems, how the online targeting systems work, what data they use, and any key design features, risk management approaches, user controls and performance data such as the proportion of views of different items of content that were driven by the online targeting systems;
  • Empowering the regulator to appoint independent third-party auditors to provide an ongoing monitor of compliance;
  • Requiring online advertising platforms to maintain advertising archives for a specified period of time detailing how they target consumers, especially advertising related to age-restricted products, credit and housing as they have wider societal harm implications. There is existing precedent for such an approach - Facebook unveiled an “ad library” in 2018 in response to criticisms over its lack of transparency in using online targeting in political campaigns;
  • Encouraging market solutions, such as the emergence of data intermediaries – data representatives – to promote user empowerment. These third-party intermediaries would advocate for the interests of individual users, create standardised user controls and could be set up as trusts with fiduciary responsibilities to their members’ interests. Query however whether such an approach would work effectively in high cost lending markets where vulnerable groups do not necessarily recognise they are at risk of online targeting.

In a follow up article, we will compare the UK approach to the ACCC’s approach to adtech.

""