European browser choice screens were associated with higher use of Firefox on mobile devices, according to a revised working paper examining a requirement under the EU Digital Markets Act.
Researchers estimate that Firefox usage was 91% higher on iOS and 13% higher on Android than it would have been without the mandate. Those are relative changes against an estimated counterfactual, not increases of 91 and 13 percentage points in market share.
The study, revised in November, examines a narrow but commercially consequential question: what happens when people are asked to choose a browser instead of relying on the option already set for them?
How the researchers made the comparison
The European Union’s Digital Markets Act places obligations on designated large digital platforms, including rules intended to give users more choice. A browser is the software used to visit websites, and a default is the application a device opens automatically for a task unless a user changes it.
Jesper Akesson, Kush Amlani, Raul Cepeda Suarez, Emily Chissell, Stefan Hunt, Michael Luca and Gemma Petrie used a statistical method called difference-in-differences to compare changes around the introduction of the choice-screen requirement. The method attempts to estimate what would have happened without the policy by using a comparison group.
In their National Bureau of Economic Research working paper, the authors say the larger estimated effect on iOS is consistent with the screens reaching existing users there, while on Android they mainly reached new devices.
A counterfactual cannot be observed directly. The estimates rely on the comparison and assumptions used by the researchers, and they measure Firefox rather than every competing browser. A working paper is preliminary research circulated for discussion.
Mozilla, the organization behind Firefox, supplied data for the study. Two authors are Mozilla employees, and Mozilla funded an earlier project involving some of the researchers. Those relationships are disclosed in the paper and are relevant context for assessing its findings.
Why default settings have commercial value
Many customers continue with the settings available when they first use a device. Changing software requires attention and a decision, even if alternatives are readily available.
For a browser provider, a default position can affect how often its product is opened. For businesses that depend on the web, the software people use can influence how sites are displayed and which privacy or technical features customers encounter.
A choice screen reduces one barrier by presenting alternatives at a moment when users are setting up or using a device. Its effect can still depend on the wording, order and timing of the options, as well as whether existing users see it.
The results do not establish that the same design would produce identical changes in another country or for a different software product. They also do not measure whether every user who switched preferred the new browser over time.
Relative growth is not market share
A relative increase measures the change compared with a starting or estimated reference level. For illustration, a rise from two users to four users is a 100% increase, but the change in share depends on the size of the whole population. The example is arithmetic, not a figure from the Firefox study.
That difference matters when a competitor begins with a small share. A large relative gain may coexist with a much smaller change in the percentage of all users choosing the product.
The researchers’ comparison also concerns usage, not a direct measure of advertising revenue, business profit or the total benefits of the Digital Markets Act. The design of default and choice interfaces remains one part of the wider debate over competition in digital markets.
In our earlier coverage of confirmation prompts, researchers found that an extra choice could change subscription decisions while adding checkout friction. That experiment studied a different kind of prompt and should not be treated as a test of the EU browser screens.
