What's the issue?

The ASA places high expectations on marketers of age-restricted products to comply with targeting restrictions designed to limit children’s exposure to such ads.

The CAP Code is already clear on targeting types of media. Targeting ads for age-restricted products to a child profile on a mixed-age site is not necessarily a breach of the CAP Code. Age-restricted ads for alcohol, gambling etc. can be served in media where adults comprise 75% or more of the overall audience.

Conversely, the Code bans ads for HFSS products in children’s media (where 25% or more of the audience are children). Likewise, gambling and alcohol ads can't appear in media which appeals particularly to under-18s (where 25% or more of the audience are aged 17 or younger).

The ASA acknowledges that it has less experience in considering dynamically served (ie addressable) age-restricted ads in mixed-age online media when the audience is >75% adult. It recognises CAP’s guidance on Age-restricted Ads Online that marketers of dynamically served age-restricted ads should:

• use media- and audience-targeting tools to target their ads away from groups aged under 16/18 (depending on the nature of the product advertised)
• exercise special care where no data is available regarding the age of a user.

What's the development?

The ASA carried out a proactive monitoring sweep using six avatars to assess how dynamic ads for alcohol, gambling and HFSS products were served to adult and child audiences on websites and YouTube channels attracting a mixed-age audience but predominantly composed of adults. The avatars represented online profiles of: a 6-7 year old, an 8-12 year old, a 13-16 year old, an adult, an adult and <12 year old shared profile, and a neutral profile with no browsing data to indicate the user's age.

The monitoring focused on:

• mixed-age online media - consisting of non-logged in websites and YouTube channels, with adults comprising 75-90% of the audience but therefore still with 10-25% children in the audience (ie spaces which attract both child and adult audiences)
• dynamically served online display ads for alcohol, gambling and HFSS products.

The ASA identified 248 Gambling ad impressions and 182 HFSS ad impressions served to the six avatars, with no alcohol ads.

The ASA has, following the sweep, published a report calling on advertisers to make better use of audience and targeting tools to help minimise children’s exposure to dynamically served age-restricted ads in mixed-age sites (those being ads for HFSS products, gambling and alcohol). It also calls on third parties in the online ad sector to ensure the data and modelling on which these tools rely are as accurate and effective as they can be to support advertisers in minimising children’s exposure to age-restricted ads online.

The ASA recognises that its request goes beyond the approach taken under the current CAP Code but says "it is a legitimate regulatory objective to seek to minimise children’s exposure to age-restricted ads generally and therefore wants [sic] to see advertisers of these products use available tools to more effectively target their ads away from children, even where the vast majority of an audience is over 18."  It suggests marketers of age-restricted products and media distributors could do more to protect children from age-restricted ads online.

What does this mean for you?

For HFSS and gambling advertisers, ad tech providers and media owners, there seems to be a veiled threat here: if advertisers and ad distributors don't do more to use tools, data and modelling to minimise children's exposure to dynamically served age-restricted ads, then more prescriptive measures are likely to be introduced.

The report notes that, even in the current framework, the ASA can take enforcement action "where it feels advertisers and the third parties they engage have not done enough to direct dynamically-served age-restricted ads away from child audiences in online media where 75% or more of the audience is adult." Nevertheless, the ASA will ask CAP to consider whether a more prescriptive approach is merited for regulating dynamically-served age-restricted ads.

ASA rulings in this area already demonstrate a very protective approach to preventing children seeing inappropriate ads in children's media. Even if advertisers use the targeting tools made available on sites such as YouTube, when inappropriate ads appear in children's media, the advertisers will still be responsible. This is demonstrated by rulings against Warner Bros and Spotify which confirm:

• it is the advertiser who is responsible for the targeting and that responsibility can't be delegated to YouTube etc.
• the ASA will take a relatively broad approach to the media which they consider to be "likely to appeal to children" and they'll focus on the media rather than the advertiser's attempts to target.

The Warner Bros ruling is interesting because, despite their use of 18-44 and 'no kids' parameters, the ad still appeared on a lullaby playlist and WB were still found responsible for the inappropriate targeting. That reinforces that it's not the steps taken to target the ad that matter but where the ad appears: "although Warner Bros had taken steps to target the ad responsibly, because it appeared around content that was likely to be heard by young children, who were likely to be distressed by it, the ad had been inappropriately targeted".

The ASA's analysis of the data from the sweep suggested that, while other factors could play a part in the way ads were targeted, that data and modelling underlying audience targeting tools made available by YouTube and other third parties, were not as effective as they could be in restricting access to ads by the relevant groups.

Advertisers need to bear in mind the ASA's scrutiny and focus on this area when appointing their media agencies, and when working with them to ensure media buying is appropriately targeted, using a combination of all the data, blocklists and selection tools which are available.