Dark Patterns 2026

Summary

The use of deceptive online design, so-called dark patterns, is a commonly recognised feature of online markets. Consumers experience them commonly on websites and apps and they can lead them to make decisions that are not in their best interests.

In 2021, Which? conducted consumer research to understand consumers’ experiences of dark patterns and we found that instances of harm were common. Five years on from that study, this report updates the evidence on how deceptive online design practices continue to impact consumers across the UK. Drawing on a nationally representative survey of 2,085 UK adults conducted in April 2026, the research examines awareness, exposure, behavioural impacts, and consumer attitudes towards dark patterns.

The findings reveal that potentially manipulative design remains widespread across everyday digital journeys, particularly within online shopping, social media, and subscription platforms. Key insights include:

• Widespread awareness and exposure: All surveyed consumers reported encountering at least one type of design feature that is a potential dark pattern. Two-thirds (63%) are aware of influence attempts, with scarcity claims (67%) and personalised prompts (66%) being the most frequently observed features.
• High awareness and confidence does not prevent detriment: Despite 79% of respondents feeling confident in navigating deceptive online design, 65% experienced negative outcomes or changes in behaviour, and 28% reported direct financial harm.
• Disproportionate impact on financially-strained consumers: Consumers who reported being under financial strain also reported higher levels of harm (60%) and severe outcomes (50%) compared to financially comfortable respondents (43% and 36%, respectively).
• Specific types of financial harm: Financial detriment resulted from continuing to pay for subscriptions longer than intended due to cancellation friction (20%), accidentally signing up to a paid subscription (12%), spending more than intended (12%), and accidental purchases (5%).
• Erosion of trust: Negative experiences have increasingly led to active disengagement, with 25% of consumers stopping use of specific platforms and 22% reporting a total loss of trust in the organisation.

Consumers strongly favour regulatory action and being able to have greater control over their online experiences, with 87% agreeing they should have the ability to turn off influence features and 82% supporting regulatory action to limit potential harm. These findings also highlight that consumer protection cannot rely solely on individual vigilance. More needs to be done to fully understand the consumer harm caused by deceptive online design, and how to tackle it.

1. Introduction

Websites and apps use a wide range of design features to shape how choices are presented to consumers. This ‘online choice architecture’ can be helpful, for example by making relevant information easier to find or simplifying a decision. However, it can also be used to steer, pressure or obstruct consumers in ways that make them more likely to take an action they did not intend to take, or which benefits the organisation at the consumer’s expense. These practices are commonly described as ‘dark patterns’ or deceptive design.

Dark patterns can take many forms. Some create urgency or social pressure, while others make certain options more prominent, set a particular choice as the default, make information harder to understand, or introduce friction when consumers try to act against the organisation’s interests. The effect of any individual feature will depend on how and where it is used: a countdown timer or personalised prompt, for example, is not necessarily harmful in itself. The concern is where design materially distorts or constrains consumers’ ability to make free and well-informed choices.

In 2021, Which? conducted consumer research to understand consumers’ experiences of dark patterns and we found that instances of harm were common. Since then, academic research has further documented the scale and systematic nature of dark patterns including the commercial incentives and third-party infrastructure that support their widespread use.  In the UK, the Competition and Markets Authority’s work on online choice architecture has set out how digital design can distort consumer decision-making and competition, particularly where multiple design features interact to create pressure or reduce meaningful choice. Meanwhile, government-commissioned research has begun to explore the behavioural impacts of specific practices such as defaults in online shopping, though this evidence is typically practice- or sector-specific. Research funded by the Which? Fund has found that some groups of vulnerable consumers may be particularly disadvantaged by dark patterns.

Despite these developments there has been little progress in the UK on how to better protect consumers from manipulative design features in online markets. The CMA is conducting more enforcement actions against businesses for practices that may be defined as online choice architecture, but these tend to be cases where the firm has engaged in misleading rather than manipulative behaviour and where the law is clear, for example instances of drip pricing. By contrast, the European Commission is expected to shortly bring forward a Digital Fairness Act, signalling a desire to take stronger action against manipulative digital design. 

To understand how consumer’s experiences of dark patterns in the UK have changed through the 2020s, this report updates our previous research to present broad evidence of the scale of consumers’ experiences across a range of dark patterns.

2. Methodology

This research updates and expands on Which?’s 2021 consumer research on dark patterns. It examines awareness and perceptions, exposure to practices, behavioural and financial outcomes, and subscription and cancellation experiences. The 2021 and 2026 studies are broadly comparable in purpose but do not use identical questionnaires or measures, so comparisons are indicative rather than strict like-for-like trends.

The research explores ten types of dark patterns.

Table 1: Types of dark patterns

Yonder, on behalf of Which?, surveyed a nationally representative UK sample of 2,085 adults between 2 and 7 April 2026. Unless otherwise stated, findings are weighted and the weighted base is 2,036 adults who use at least one of the digital services or activities covered by the survey. Some subscription questions were asked only of relevant subgroups and therefore have smaller bases.

The impact of dark patterns varies considerably across contexts and individuals, to assess the seriousness of reported outcomes, relevant responses to the question asking consumers what, if anything, had happened to them as a result of these design features were classified as low, medium or high severity (see Table 2). Respondents reporting more than one outcome were assigned the severity of the most serious outcome they reported. People who reported none of the coded outcomes were classified as experiencing no harm.

Table 2: Forms of harm by severity

To examine whether experiences of dark patterns differed according to consumers’ financial circumstances, respondents were grouped according to how easy or difficult they found it to keep up with household bills and essential spending. Those who said this was ‘very easy’ or ‘fairly easy’ were classified as financially comfortable; those who said it was ‘neither easy nor difficult’ formed a middle group; and those who said it was ‘fairly difficult’ or ‘very difficult’ were classified as financially strained. These groups were then compared to examine differences in the prevalence and severity of reported consumer harm.

3. Findings

3.1 Consumer experiences and attitudes to dark patterns

Consumer awareness of dark patterns is high. Nearly two thirds of consumers (63%) in our 2026 survey said they were aware that websites and apps sometimes use design features to influence behaviour while roughly the same proportion (62%) said they often notice attempts to influence their choices. This was consistent with the findings of the 2021 survey in which there was also a reported level of awareness of 63%.

When asked about specific types of potentially manipulative online choice architecture, all consumers reported seeing at least one feature. Seven of the ten features we asked about were encountered very or fairly often by a majority of UK adults, with scarcity claims and personalisation being the most common and being noticed by two-thirds of people, see Figure 1.

There has been some change since 2021 in the features seen most often by respondents. Activity messages, trick questions and scarcity claims were the three most commonly seen features in 2021 whereas our latest research shows a much broader degree of awareness.

Figure 1: Online choice architecture seen ‘very often’ or ‘fairly often’ by UK adults

Exposure was spread across different digital services. Online shopping websites were the clearest hotspot, with two thirds of consumers (66%) noticing potential dark patterns there. This was followed by social media platforms (40%), subscription services (26%), travel websites (21%) and ticket-booking websites (20%).

Figure 2: Websites or apps where dark patterns were noticed

Consumers generally reacted quite negatively to these features. Four in five (80%) expressed at least one negative sentiment. Annoyance was the most common response (59%), while 44% described the features as manipulative and 36% said they did not trust the information being shown.

These results suggest that deceptive design is not confined to particularly explicit or aggressive practices. Many commonly encountered features are embedded within ordinary digital journeys, including personalised prompts, scarcity messages, free trials and other forms of choice architecture.

3.2 The impact of dark patterns

Most respondents (79%) felt confident that they could use websites and apps without making choices they did not intend to make. At the same time, nearly two thirds (65%) reported experiencing at least one harm or negative outcome associated with dark patterns. The relationship is unlikely to be straightforward. More digitally active consumers may both notice these features more frequently and have more opportunities to encounter their effects. However, the findings show that high awareness and confidence can coexist with widespread reported harm. Consumer awareness and vigilance should therefore not be treated as the only safeguard against potentially harmful design.

Consumers reported a range of outcomes relating to their decisions and behaviour. Almost a third (28%) of respondents reported experiencing some form of financial harm because of a dark pattern. The most common financial harm was continuing to pay for a subscription for longer than intended.

Figure 3: Types of financial harm reported

People also reported having worse experiences, and sometimes spending longer on a task than they would like: 22% said they had difficulty cancelling, opting out, or changing a setting as a result of these design features, while 18% said they had found it difficult to choose what they wanted. Other ways in which choices were affected were that 10% said they made a decision more quickly than intended, and 8% said they chose an option they would not otherwise have chosen.

The impact of these dark patterns sometimes extended beyond the immediate decision and affected engagement and levels of trust. A quarter of UK adults, 25%, said they had stopped using a website or app temporarily or permanently, while 22% said they had lost trust in the organisation using them.

Consumer harm was also more common among people experiencing financial strain. The proportion reporting at least one mapped consumer harm rose from 43% among financially comfortable consumers to 60% among financially-strained consumers. The proportion of people experiencing some form of financial harm increased from 22% to 38%.

The difference was also apparent in the seriousness of the outcomes reported. High-severity harm rose from 36% among financially comfortable consumers to 50% among financially-strained consumers, while the proportion classified as experiencing no harm fell from 32% to 20% (see Figure 4).

These findings do not establish that financial strain itself causes greater harm from deceptive design. They do, however, show a clear association between financial pressure and both a greater prevalence and greater severity of reported harm. This is particularly important for financial harms: the same unwanted payment or unintended purchase may have substantially greater consequences for a consumer with less capacity to absorb it.

Figure 4: Harm severity split by financial circumstances

Subscription traps 

Subscriptions provide one of the clearest examples of a pathway from friction to financial harm.

We found 39% of UK adults had tried to cancel or change a subscription during the previous 12 months. Of these, 60% of those who had done so found cancellation easy, but 29% found it difficult. Common barriers included retention offers (39%), too many steps (32%) and having to confirm cancellation multiple times (30%).

Among those managing subscriptions monthly or who had tried to cancel or change one, around one in five said they continued paying longer than intended because cancellation was difficult.

Figure 5: Barriers experienced when attempting to cancel or change a subscription

The Digital Markets, Competition and Consumers Act 2024 establishes a new regime for subscription contracts, intended to give consumers greater control over recurring payments. Under new rules that will start to apply from January 2027, businesses will be required to provide clearer information before consumers sign up, issue reminders before free or discounted trials end or longer-term contracts renew, and make subscriptions straightforward to cancel. Consumers will also have a new 14-day cooling-off period after a free or discounted trial ends, or when a contract of 12 months or more renews.

3.3 2026 compared to 2021

At a broad level, the 2026 findings mirror the picture established in Which?’s 2021 research. Awareness remains high, exposure remains widespread and consumers continue to report negative experiences associated with deceptive design.

Levels of awareness of dark patterns appear unchanged between 2021 and 2026, with 63% of respondents in both surveys saying they were aware that websites and apps sometimes use design features to try to influence people into behaving in certain ways.

Similarly, in both surveys nearly two thirds (65%) reported experiencing at least one harm or negative outcome associated with dark patterns. Some narrower financial measures were also broadly stable: spending more than intended was reported by 13% in 2021 and 12% in 2026, while accidental purchases were reported by 6% and 5% respectively.

The largest differences between the surveys relate to trust and disengagement. Reduced or lost trust was reported by 15% in 2021 and 22% in 2026, while the proportion saying they had stopped using a website or app was 16% in 2021 and 25% in 2026.

Figure 6: Frequency of consumers' experiences of selected design

These comparisons should be treated cautiously because the surveys are not identical in design. The main contribution of the 2026 research is therefore a fuller picture of reported harm, particularly around friction, altered decisions, subscription cancellation and differences according to financial circumstances.

3.4 The interventions consumers want

Consumers expressed substantial support for measures that increase transparency and give them greater control over design features that might be manipulative.

Some 87% agreed that consumers should be able to turn these features off, while 84% agreed that organisations should be transparent about using them. More than eight in ten, 82%, agreed that regulators should act to limit potential harm, while 70% agreed that organisations should not use these features at all.

Figure 7: Consumer attitudes towards intervention

The findings suggest support for a range of responses rather than a single intervention, particularly measures that give consumers greater control, improve transparency and address features associated with consumer harm.

4. Conclusion

The results of our survey show that dark patterns remain a routine feature of online consumer journeys and can change consumers’ choices, sometimes leading to financial detriment when consumers make purchases they did not intend to. It is concerning that those in more strained financial circumstances appear to be more likely to suffer harm due to dark patterns.

It is notable that the most striking difference between our surveys in 2021 and 2026 is the increase in the number of people that have reacted to experiences of dark patterns by stopping using a website or by losing trust. It may be that people are becoming more intolerant of this kind of behaviour by businesses. However, it is also clear that consumers cannot be expected to protect themselves against these design features. They are deliberately created to influence choices to the benefit of businesses and we see very little change in their effectiveness with similar proportions of people spending more than they intended or making accidental purchases as a result of these features.

UK consumer protection law and regulation is not currently well designed to protect consumers from manipulative practices in online markets. However, the survey findings clearly show that people want change, with large majorities of respondents expressing a desire to be able to exercise more control over their transactional experience and for regulators to do more. 

At a European level, preparations for a forthcoming Digital Fairness Act signal a potential shift toward stronger, more preventative approaches to regulating manipulative digital design, including ex-ante obligations rather than reliance on enforcement alone.  However, policy debates in the UK are less advanced. More work needs to be done to understand how manipulative online choice architecture causes harm, how different design features interact to influence consumers’ choices, and how this varies in different market settings and for different products. This then needs to inform policy discussions about the best way to better protect consumers from manipulative dark patterns and whether it requires regulatory or legislative change.