Written Evidence Submitted by Dr Rebecca Mardon (Cardiff University), Professor Kate Daunt (Cardiff University) and Dr Hayley Cocker (Lancaster University)

Dear Digital, Culture, Media and Sport Committee,

This is a response to your request for evidence for the committee’s current inquiry into influencer culture. Drawing from the findings of our immersive, longitudinal, ethnographic study of leading UK-based social media influencers and their followers (specifically, our research investigated beauty and lifestyle influencers, with a primary focus on YouTube)[1], we present both published and unpublished findings in response to the questions posed by the inquiry. In summary, the discussions below suggest that:

• Consumers are generally aware of the types of arrangements that take place between influencers and brands, but feel that influencers do not always declare endorsements appropriately. Consumers report that some influencers fail to clearly disclose endorsements, whilst there is also significant variation in how influencers disclose endorsements, making it hard for consumers to discern whether content involves an endorsement and/or the nature of the endorsement.
• Many influencers delete and/or block critical comments from their social media platforms, and may block consumers who post these comments from accessing their future social media content. In doing so they prevent consumers from voicing their concerns surrounding endorsement disclosure.
• Some consumers actively report undisclosed or insufficiently disclosed endorsements to the Advertising Standards Authority (ASA) but feel that insufficient action is taken in response to their complaints, resulting in limited long-term impact on influencer disclosure.
• Current UK regulations surrounding disclosure do not sufficiently cater to all age groups, whilst national regulations are themselves problematic in the age of social media, where consumers regularly access the social media content of influencers from across the globe, each bound by distinct local regulations. We recommend that the committee consider these issues.

Q. How would you define ‘influencers’ and ‘influencer culture’? Is this a new phenomenon?

Influencer culture refers to a societal shift whereby consumers are increasingly seeking brand and product/service recommendations from their favourite ‘influencers’, rather than other information sources. Influencers are celebrities with large social media followings that they monetize by influencing their followers’ consumption behaviours. Common monetisation strategies include:

• paid for advertorial social media content produced in collaboration with brands that seeks to target the influencer’s followers with marketing communications messages.
• affiliate links that earn the influencer a commission when their followers purchase a product/service after clicking on a link posted to their social media profile.
• the sale of own-brand merchandise to their followers.

In addition to direct monetisation, influencers also frequently receive non-monetary benefits in the form of free or heavily discounted products and services from brands, which range from free makeup products to all-expenses-paid luxury holidays, which are given in order to encourage positive coverage of the brand in the influencer’s social media content. Whilst celebrities have long engaged in celebrity endorsements, appearing in brand advertisements in the mainstream media (e.g. TV advertisements) and on brands’ own media channels (e.g. the brand’s website), influencers have the capacity to promote products/services (both their own and those of other brands) on their own social media profile, directly targeting their own established, and often highly engaged, audience.

Whilst many influencers are traditional celebrities whose fame transcends and predates their social media presence (e.g. reality TV stars, sports stars), a sizeable segment are ‘social media influencers’ (SMIs), defined as “ordinary Internet users who accumulate a relatively large following on blogs and social media through the textual and visual narration of their personal lives and lifestyles”[2]. Unlike traditional celebrities, SMIs’ fame does not predate their social media presence, but rather stems from it. For instance, leading UK influencers Zoe Sugg (aka Zoella) and Sophie Hinchliffe (aka Mrs Hinch) are ordinary consumers who rose to fame on social media (YouTube and Instagram respectively). Unlike other influencer-types, such SMIs typically rise to fame within specific online ‘consumption communities’ or ‘tribes’ surrounding interests such as beauty, fashion or parenting, and are thus are members of a wider community[3],[4]. Despite originating as ordinary consumers, many SMIs’ online followings equal or exceed those of traditional celebrities (e.g., Zoella has 11 million YouTube subscribers, whilst Mrs Hinch has 4.1 million Instagram followers). Furthermore, research indicates that many consumers – younger generations in particular – are more influenced by SMIs than traditional celebrities, placing greater trust in their recommendations[5]. It is these SMIs, as opposed to influencers whose fame transcends social media, that are the focus of our research.

Whilst SMIs have existed since the advent of social media, the prevalence of influencers, the scale of their fame and influence, and their capacity to monetise their influence over their social media followers has increased significantly over the past decade. Indeed, early SMIs typically did not monetise their influence, or did so only via display advertising placed on their social media content – e.g. their blog or YouTube videos – that earned them a small income based on the number of impressions or clicks achieved. Consequently, the majority of early influencers saw social media content creation as a hobby and did not envisage that they would one day be able to earn significant amounts of money from their growing social media followings. However, as these content-creators’ audiences grew and marketers recognised their influence over their social media followers, SMIs began to be offered other monetisation opportunities such as paid-for brand advertorials. As their fame grew, the frequency of such advertisements grew, as did the fees influencers could charge advertisers, and many content creators quit their day jobs to become full-time influencers, typically managed by talent management agencies that guided their attempts to monetise their influence.

Q. Is it right that influencers are predominantly associated with advertising and consumerism, and if not, what other roles do influencers fulfil online?

Our research revealed that SMIs occupy multiple roles in relation to their viewers, and that the emergence of the influencer role can challenge the existing roles that led to the SMI’s initial success. We found that SMIs’ primary role is to serve as a ‘guru’ within a particular online consumption community (e.g. the YouTube beauty community), providing their viewers with informative, impartial, unbiased product recommendations as well as tutorials and advice on product use. Additionally, they occupy a secondary, complementary ‘friend’ role, with consumers developing parasocial relationships with influencers characterised by perceived intimacy. Thus, consumers were attracted to SMIs’ content because of these guru and friend roles, finding SMIs’ informative and relatable content appealing. We found that the emergence of the influencer role challenged these existing roles; SMIs’ collaborations with brands caused consumers to question their trustworthiness (challenging the ‘guru’ role) and their loyalty (challenging the ‘friend’ role). Thus, far from being the only role that SMIs occupy, the influencer role challenges their existing roles within the community, requiring careful management.

Many content creators reject the label ‘influencer’, preferring terms such as ‘content creator’ and arguing that ‘influencing’ is not their main role. Many SMIs attempt to balance the influencer role with their other communal roles (e.g., engaging in endorsement activities selectively and integrating endorsements into content in ways that do not negatively impact its entertainment or information value). Indeed, as we shall discuss below, consumers within the online consumption community may respond negatively when tension between these roles becomes apparent, and thus SMIs are often careful to moderate their own monetisation of their influence over their follower-base in order to avoid simultaneously eroding this influence by indicating a lack of trustworthiness or loyalty to their followers. Thus, SMIs’ relationship with their followers, and the way in which this constrains their own monetisation activities, sets them apart from other forms of influencer, whose origins do not require such careful management of communal roles.

Additionally, within our study of beauty and lifestyle influencers, we found that a range of important topics were addressed by many influencers, such as mental health, physical health, body image, sexuality, budgeting and investment, charitable causes and sustainable consumption. Thus, influencers’ capacity to influence is not restricted to increasing consumption; they also have the potential to influence consumers’ consumption behaviours and wider lives in more positive ways.

Q. How aware are users of the arrangements between influencers and advertisers? Should policymakers, tech companies, influencers and advertisers themselves do more to ensure these arrangements are transparent?

In a recent examination of consumer responses to endorsements by SMIs on YouTube[6], we found that consumers have high standards surrounding such endorsements and their disclosure. We observed clear expectations as to how SMIs should and should not implement celebrity endorsement. Where these expectations were not met, endorsements were perceived by consumers as transgressive, resulting in negative consequences for SMIs (e.g. consumers expressing negative sentiment towards the SMI, avoiding and even unsubscribing from their social media content) as well as for the endorsing brand. One such expectation identified in our study was that endorsements would be clearly disclosed to viewers. Consumers commented on YouTube videos that they felt contained undisclosed or insufficiently disclosed endorsements, clarifying their expectations surrounding disclosure and often justifying these expectations by making reference to the current UK regulations surrounding influencer endorsement disclosure. Forms of insufficient disclosure frequently noted by consumers included:

• ‘Forgetting’ to disclose an endorsement, adding appropriate disclosure to a social media post after its initial upload, often after the majority of consumers have viewed, and may have acted on, the content. (e.g., adding ‘AD’ to a YouTube video title hours, or even days, after its initial upload).
• ‘Hiding’ endorsement disclosure (e.g. writing ‘AD’ in white text on a light background, or in a very small font, in the YouTube video thumbnail, writing ‘AD’ at the end of a long video title, rather than at the beginning).
• Using incorrect and/or inconsistent terminology, causing confusion surrounding the nature of an endorsement (e.g., referring to products/services using terms such as ‘gifted’, ‘PR gift’, ‘PR stay’, or disclosing affiliate links using labels such as ‘affiliate’ or ‘aff’, rather than using the ‘AD’ terminology required by the ASA).

Furthermore, though this is not a current ASA requirement, consumers often expected influencers to verbally disclose an endorsement within video content, noting that they often multitask whilst watching YouTube videos in particular, and therefore may miss written disclosures.
 

We found that some influencers actively seek to educate their followers on the nature of their brand collaborations and/or the current guidelines surrounding disclosure (e.g. producing social media content explaining the ‘behind the scenes’ process of working with brands, or explaining the terminology they will be using to disclose endorsements). However, we also found evidence to suggest that some SMIs delete or block critical comments on their social media posts, preventing consumers from voicing their concerns and establishing a dialogue with the SMI surrounding endorsement disclosure. Many independent online forums have emerged as a result of this censorship, whereby community members discuss (amongst other things) instances of non-disclosure or insufficient disclosure by SMIs. Some consumers actively report these disclosure issues, and are often told by the ASA that they will investigate or speak to the SMI involved, however they are concerned that no punishment seems to occur, and that SMIs therefore seemed to repeatedly commit the same transgression. These consumers often conclude that reporting non-disclosure/insufficient disclosure to the ASA was not worthwhile.

These findings lead us to conclude that:

• Further enforcement of the current guidelines surrounding influencer endorsements is necessary in order to motivate influencers, their management teams, and the brands they endorse to comply with these guidelines. In particular, the ASA may benefit from imposing higher fines, and doing so more readily. Furthermore, if an influencer has been found to have broken the current rules surrounding endorsements, the ASA could also consider requiring them to upload an equivalent social media post notifying their audience of the ruling, thus 1) educating consumers and influencers alike on the current guidelines, 2) reassuring consumers that that their concerns are being acted on, thus motivating them to continue to report rule-breaking by influencers to the ASA, and 3) further discouraging the influencer from future rule-breaking.

• Further standardisation of disclosure mechanisms would enable consumers to more easily identify influencer endorsements and understand the nature of these endorsements. This may involve new guidelines requiring influencers to use a standardised terminology, and enforcing these guidelines when influencers use alternative, potentially confusing terminology. However, issues may still emerge surrounding how this terminology is incorporated (e.g. the clarity of the disclosure may vary due to font-size or placement within visual or video social media content, or where and how the disclosure is placed in written posts). Consequently, we perceive value in using standardised, platform-enforced disclosure mechanisms (akin to Instagram’s Paid Partnership tool) that clearly disclose an endorsement and enable users to distinguish between different types of endorsement. Indeed, given that some consumer confusion remains surrounding the terminology used to disclose influencer endorsements, this disclosure tool could enable users to click to learn more about the type of disclosure used (e.g. ‘what is an affiliate link?').

We’d like to raise two additional concerns surrounding endorsement disclosure by influencers, for consideration by the committee:

Influencer endorsement disclosure & children

Many children are avid consumers of social media content[7]. Indeed, many YouTube channels’ audiences are primarily children; the popular YouTube channel RyanToysReview, for instance, has amassed more than 29.5 million YouTube subscribers and its 9-year old star Ryan’s own-brand toy range is available from retailers worldwide. According to his mother, Ryan’s subscriber-base of child-viewers feel like they’re “on a play date with him and going on fun, pretend play adventures”[8]. However, how capable are these children of identifying endorsements by influencers like Ryan? Current guidance surrounding influencer endorsement disclosure by the ASA, CMA and CAP[9] does not consider whether these disclosure mechanisms are suitable for child viewers. Research shows that children have lower advertising literacy than adults;  they struggle to recognise adverts embedded in organic content[10], as is the case in influencer endorsements. Evidently children need more explicit disclosures than adults in order to successfully identify influencer endorsements, however current disclosure mechanisms are problematic.

First, our research indicates that the current variation in influencers’ disclosure mechanisms presents issues for adults, let alone children who are still learning to identify advertisements. As we shall discuss below, advertising regulations are country-specific and highly diverse and therefore consumers who view social media content from influencers based in other countries are exposed to a wider variety of disclosure mechanisms. Furthermore, local guidelines are often open to interpretation, with significant variations in how influencers’ endorsement disclosures are made. For example, videos featuring products provided to influencers free-of-charge by brands are variously labelled “ad”, “gifted”, “freebie”, or “PR sample”. The result is a complex and inconsistent system of endorsement disclosure that is difficult to fully comprehend. Since children learn to recognise adverts by learning methods of identification, exposure to highly varied forms of disclosure may make it difficult for them to understand what is and what is not advertising.

Secondly, the current UK guidance on endorsement disclosure by influencers refers to written disclosures. However, children access social media sites from an increasingly young age, often before they can read with confidence. For instance, according to research by Ofcom, in 2019, 51% of 3-4 year olds and 64% of 5-7 year olds watched YouTube videos. Further consideration therefore needs to be given to the age-appropriateness of disclosure mechanisms. We recommend making verbal disclosure mandatory for video content targeting children (indeed, given the findings reported above, verbal disclosure would also be beneficial for adult consumers).

International variations in influencer endorsement disclosure mechanisms

Despite the global audiences of many influencers, regulations surrounding endorsement disclosure vary significantly between countries. Influencers are typically required to abide by the regulations in their country of residence and/or the country from which they are uploading, and are not bound by the regulations in their viewers’ country of residence. Consequently, audience members are exposed to highly varied forms of endorsement disclosure, making it difficult for them to identify endorsements and understand the nature of these endorsements.

We perceive value in establishing consistent global disclosure regulations or ensuring a higher degree of consistency surrounding these regulations, given the international nature of most influencers’ audiences. We also see value in liaising with social media platforms to ensure consistent and easily identifiable disclosure mechanisms. For instance, the platform might introduce a labelling system to help users identify and distinguish different types of endorsements - akin to Instagram’s Paid Partnership tool, but covering different types of endorsements. Influencers could select from a set of pre-determined endorsement options (e.g. paid-for advertorial, PR sample, PR experience, affiliate link, own-brand merchandise) when uploading, resulting in an automated endorsement disclosure message. A platform-imposed approach to disclosure would ensure consistency between content creators, reducing the likelihood of consumer confusion. If it is not possible to achieve consistent regulations across geographical regions, this approach could also enable disclosure mechanisms to be tailored to the user, based on their selected country of residence, rather than that of the influencer, thus ensuring consistency for the consumer.

Given the confusion surrounding endorsement types and terminology, these endorsement disclosure messages could be clickable links, providing interested users with an explanation of the type of endorsement being disclosed. Here, platforms could not only contribute to the standardisation of influencer endorsement disclosure, but also provide access to further information that will enable consumers to educate themselves on influencer endorsements and the regulations surrounding endorsement disclosure in their country of residence.