End-user Experience as a Blind Spot in Brazil's Digital Markets Bill
Bill No. 4,675/2025 deliberately moves away from the consumer welfare standard as an analytical benchmark. As Victor Oliveira Fernandes, former CADE commissioner, argues, Art. 47-B introduces objectives that are sharply divergent from those of Law 12.529/2011.[1]
That shift is defensible on its own terms. But it leaves an unresolved question: the Bill names “promoting freedom of choice” as a goal without specifying how anyone would know whether that goal was achieved.
The European experience
Europe already ran this experiment and the results complicate the assumption that choice architecture translates automatically into user welfare.
Findings from the private sector suggest as much: a Nextrade Group survey of 5,000 consumers across 20 EU member states, commissioned by CCIA Europe, reports that most Europeans find their online experience worse than before the DMA, with perceived declines in map services (35%), search relevance (33%) and ad personalisation (39%). Two-thirds of frequent searchers say they need longer to find relevant content than before and six in ten report spending up to 50% more time on searches that used to be quicker – a pattern especially pronounced in Spain and Italy, where over 70% of respondents perceive searches as taking significantly longer, against an EU average of 62%.[2]
The same survey finds that 42% of Europeans who travel at least once a month say flight and hotel search results are less helpful than before, a direct consequence of rules that reordered how travel aggregators surface against gatekeepers' own listings. And a clear majority preferred what the DMA displaced: 59% would rather go straight to their apps than face a choice screen, and a similar share want to keep seeing rich, comparable shopping results rather than the fragmented alternative that followed.
That preference carries a price tag, respondents said they would pay, on average, between €113 and €191 to restore specific pre-DMA features, with willingness to pay present across all income brackets.
None of this means Europeans abandoned the targeted platforms: two-thirds still use at least three gatekeeper services daily. That combination, sustained use alongside a self-reported decline in quality, is what makes the European case instructive for Brazil: a remedy can be technically implemented and still leave users measurably worse off, without that friction ever surfacing in a compliance filing.
Why it matters for Brazil
The Aliel Machado substitutes require designated platforms to show neutral choice screens for third-party products and services (Art. 47-E, II, i) – the screen that asks users which browser or app they want, rather than defaulting them into the platform's own.
But this mechanism only works if people actually stop, read and choose differently. That is precisely what the European experience puts in doubt: users tend to treat the screen the way they treat cookie banners, clicking through to get back to what they were doing.
What the Bill has and what it lacks
Em seu favor, o PL nº 4675/2025 não é estático. As obrigações podem ser revistas quando os mercados mudarem significativamente (art. 87-B, §4º), os reguladores podem avaliar seu impacto e recomendar ajustes (arts. 47-F e 87-K, §2º, IV) e as empresas designadas devem apresentar relatórios de conformidade à nova Superintendência (art. 87-I).
To its credit, the Bill is not static. Obligations can be revisited when markets change significantly (Art. 87-B, §4º), cooperating regulators may assess their impact and recommend adjustments (Art. 47-F, 87-K, §2º, IV) and designated firms must file compliance reports with the new Superintendency (Art. 87-I).
What is missing is narrower and it becomes clear once you ask what those reviews would actually be looking at. Two distinct questions could be asked about a choice screen: (1) Was it built as ordered: neutral, visible, on first launch? (2) Did it change anything: did users switch or did they stay? The bill equips CADE to answer the first and leaves it without the means to answer the second.
The statutory tests – effectiveness, reasonableness, proportionality and competitive impact – are not defined in terms of end users. And the information reaching CADE arrives through compliance reporting, an instrument built to confirm that an obligation was carried out. That is a different question from whether it actually worked: a compliance report shows that the screen was built as ordered, not that anyone used it to switch or to stay.
If the stated goal is freedom of choice, the Bill needs a way to know whether choice actually happened, not just whether the screen appeared. Without that, Brazil risks repeating the European outcome: an obligation fulfilled to the letter and a user who never noticed the difference, or noticed and didn't like it.
References
[1] Fernandes, Victor Oliveira. Brazil's Calibrated Revolution in Digital Competition. ProMarket, November 12, 2025. Available at: https://www.promarket.org/2025/11/12/brazils-calibrated-revolution-in-digital-competition/.
[2] Nextrade Group. Impact of the Digital Markets Act (DMA) on Consumers across the European Union: Results from a Survey with 5,000 Consumers. September 2025. Available at: https://www.nextradegroupllc.com/impact-of-the-dma-on-eu-consumers.