>That puts the actual downloads for the month at 21.1 million, compared to 25.2 million for April 2016, back when the company was riding a user-growth hot strea

>That puts the actual downloads for the month at 21.1 million, compared to 25.2 million for April 2016, back when the company was riding a user-growth hot streak. Correct me if I'm wrong, but the title is a bit misleading - these numbers are year-over-year, comparing April 2016 to April 2017. So it's not a loss of $1M in one month. It's more that this month last year, they had 4.1M more installs. Judging by the graph, it looks like they only dipped ~1M installs between March and April 2017. That's not to say that this looks good, but it's not as disastrous as it seems.