While summer seasonality probably is a factor here, the larger issue is that the source of this claim (SimilarWeb) is not capturing users that switched from bro

While summer seasonality probably is a factor here, the larger issue is that the source of this claim (SimilarWeb) is not capturing users that switched from browser to the app when it was rolled out in mid-May. Some of my other sources suggest that traffic has been remained steady since mid-April.