Under-16 social media restrictions are removing millions of accounts from major platforms. Australia alone has deactivated, removed or restricted more than 4.7 million under-16 accounts, and Indonesia has reported 4.1 million TikTok and 600,000 YouTube removals. The risk for brands is not simply that social data changes. It is that their measurement depends too heavily on the slice that disappears.
Social media has long helped brands track conversations, identify needs and measure campaign impact.
But what happens when younger consumers begin disappearing from the platforms being measured?
Australia introduced under-16 social media restrictions in December 2025, while Brazil and Indonesia followed with age-assurance and access rules in March 2026.* The policies differ, but governments are clearly tightening how younger people access major digital platforms.
For brands, this is not only a policy story. It is a consumer insights story.
A digital landscape change is a data architecture test
Every brand seeking to understand younger consumers should ask:
What happens to our social media KPIs and brand signals when teens come off social?
For systems relying heavily on one platform, or social data alone, the exposure is direct. Fewer visible accounts can mean fewer posts, comments and behavioral signals.
A fall in activity may then look like declining brand relevance, weaker campaign performance or reduced category interest when, in reality, part of the audience has simply become less visible.
The issue is whether your measurement depends disproportionately on the part that moves or disappears.
Why i-Genie.ai’s digital signal data is diversified by design
At i-Genie.ai, we tested the potential impact using production data across Australia, Brazil and Indonesia. We found no structural impact on our social feed or scores.
That is because i-Genie.ai does not depend on one platform, or even one type of data, to understand consumers.
Our consumer insights AI platform draws from three complementary streams:
-
- Social conversation, capturing what consumers express across social channels, video sites, blogs, forums and communities.
-
- Consumer reviews, revealing product experiences, frustrations and purchasing language.
-
- Search behavior, showing what consumers want to know, solve, compare or buy.
The recent social media restrictions affect only part of the first stream. They do not remove search behavior or consumer reviews.
We sized that exposure rather than assuming it. The restricted platforms account for 84% of our Australian social feed, 96% in Brazil and 99% in Indonesia. But under-16s are a small share of users on those platforms: around 5% in Australia, 8 to 12% in Brazil and between 4% and 12% in Indonesia. Enforcement is also partial, with survey evidence from Australia indicating that over 60% of 12 to 15 year olds retain access, an effective enforcement rate of around 40%. Multiply those three factors together and the restrictions reach roughly 1.7% of our Australian social stream, 3 to 5% in Brazil and 1.5 to 5% in Indonesia.
Blogs, forums and news sit outside all three bans entirely. In Australian health categories, they account for 47% of our social feed.
Brand Pulse combines all three streams, search, social and reviews. Impact IQ uses search and social, with reviews available alongside a companion product. Trend Spotter draws from social, reviews and search, with search acting as the leading signal for emerging needs. CX Bench takes no social input at all, relying solely on reviews and eCommerce.
The scores are also structurally protected. Buzz is built on share of voice, so if teen posts fall at the same rate for every brand in a category, every brand’s share is unchanged and the math cancels. Only a brand whose teen appeal differs from its category average moves at all, by roughly 0.3 to 0.5 Buzz points spread across a 12-month window. The Brand Pulse Score takes no input from the volume of Buzz as mental availability, but instead the quality of the mental availability – so lost volume alone cannot move it. The realistic worst case is a few tenths of a point.
What our multi-source digital brand signal data shows
We compared smoothed social volume in Australia, Brazil and Indonesia with the United States as a no-ban reference market.
Across the relevant implementation dates, social volume held. Brazil increased after 17 March. Australia dipped before returning to its previous range. Indonesia softened modestly but remained within its recent band. The Brazilian increase reflects a Carnival-inflated February, and the same rise appears in the no-ban United States control.
None of the three markets showed the structural collapse we would expect if the restrictions had materially compromised our feed.
The bigger lessons for brands around multi-source digital signalling vs. social listening
Under-16 restrictions are only one example of how the digital landscape can change. Platforms alter APIs. Communities migrate. Algorithms change visibility. Regulations reshape access. Consumer behavior shifts. Search is migrating increasingly to LLMs, as should your data, which is why i-Genie.ai supports every major LLM as part of its search analysis stream.
Brand measurement systems built around single source or single channel data is vulnerable to every shift.
Reliable consumer insights require an architecture that can withstand disruption. When one stream becomes thinner, others must continue providing enough signal to understand what consumers need, how perceptions are changing and where the business should act.
So, what happens to your numbers when teens come off social?
For i-Genie.ai, the evidence is clear: our measurement remains resilient because it was diversified by design.
Discover how i-Genie.ai combines social, search and review data to deliver decision-grade consumer insights. Book a demo today.
* Reddit is only partially covered by the bans: it falls outside the scope in Indonesia, is part included in Brazil, and is currently challenging the Australian ban in court.
