I’m rarely found praising the Advertising Standards Authority (ASA), but praise where it is due, their Monitoring Report on Online HFSS Ads (1), is a good first crack at deploying automated avatars to monitor online advertising – it also blows to smithereens the obvious nonsense from the Advertising Association (AA) that kids barely hear a whisper of advertising for products in high fat, sugar or salt (HFSS) when online (2).
Before we look at how the ASA just sunk the AA battleship let’s nerd out on a little methodology.
In simple terms they took a fresh computer with no usage history, made it go to web pages popular with kids, so it built a history profile similar to a kid and then using an automated process it went to popular websites and You Tube pages repeatedly over a short period time to see what adverts were on show. They did the same for different profiles such as teens and younger kids.
The obvious hole is that it excludes logged in social media environments such a Facebook, which we know both kids and junk food advertisers love (3). This was a smart move, sending a bot into a logged in environment can be technically tricky and the Facebook notion that it does not “officially” have any under 13’s makes the whole thing a tangled web which they are wise to avoid on this first attempt. They acknowledge this and plan to address it in future iterations – so full marks to them. For everyone else let’s remember we are only seeing half the story and it’s already a total horror.
The other concern on their method is the absence of Google. The data Google gleans mostly from search powers the placement of advertising on You Tube and over 2 million other sites - it is the most influential targeting system online. To effectively train their avatars I feel they should go to Google and search for topical keywords for the profile demographic – such the trending kids movies, toy and hobbies. This is important as brands often target kids based on kid-like interests expressed as search terms - this is particularly true with kids entertainment branded cereals, yogurts, happy meals etc.
The ASA monitored nearly 100,000 adverts, and the results are juicier than a strawberry Starburst, and just as likely to rot your teeth.
They found nearly 1,000 breaches of the current regulations in only 2 weeks
38 brands were found to be likely breaking the rules including many of the UK's largest advertisers (McDonalds, Asda, Kellogg’s, KFC)
They found that 33% of the websites and 95% of the You Tube channels which they identified as clearly aimed at kids, were running HFSS ads
3.5% of the ads were for food and drink
69.9% of those were for HFSS products
Firstly, just to be 100% clear, the ASA go out of their way to point out that these avatars are far more intense users of the web than real kids and so the data should not be used as a measure of individual exposure. They’re right, they tell us nothing about exposure per child. But as a well-structured study of a large sample of ads it gives excellent insight into the nature of the market, and that’s going to trouble the Advertising Association…
It sinks AA’s claim that we have “strictest and most effective rules in the world to restrict the exposure of children to HFSS advertising” (1) err, no we don’t it is utter chaos and widespread abuse with no consequences for repeat offenders.
It torpedoes their suggestions that only 23.9% (1) of the ad seen by kids online are HFSS –yeah yeah, nice try, but actually the data reveals it to be 69.9%
Then we come to their battleship, the Advertising Association’s flagship claim that kids only see one half of one second on HFSS advertising online per day. This claim is based on a similar method to Kantar’s analysis commissioned by the Government’s 9pm watershed Impact Assessment. It estimated the market for advertising food and drink online to be worth £89.7m per year (4).
We’ve already analysis in the report we authored with Dr Mimi Tatlow-Golden for the Obesity Health Alliance (5). Any reasonable analysis of multiple credible data sources show that this calculation significantly understates the size of the market, by at least 8 times, and the actual overall exposure by at least 16 times. This is confirmed by this new ASA data, their sample of nearly 100,000 ads show that 3.5% of online advertising is for food & drink and with the total online ad market market worth £13.44bn (6) that makes the market for advertising food and drink online worth £470m, five time more than the Kantar estimate - and that doesn't even include social channels, influencers and advergaming.
I agree with the Advertising Association on one thing, that the solutions to the crisis of childhood obesity need to be “evidence-based”. We’ll here it is, good solid evidence - we have clear wide-spread use of effectively unregulated junk food ads for deserts, takeaways, crisps, sugary cereals and confectionery polluting our children’s screen time – the current regulations do not work – it’s a mess – families have had enough (7).
We need a 9pm watershed – we need it now - it is that simple.