Re: [w3ctag/design-reviews] Question: Review manifest-first Web Install API (Issue #1245)

dmurph left a comment (w3ctag/design-reviews#1245)

I worry a little bit about a mis-representation of stats that I have given. It looks like the standards position cited my click-through rates but not the generalized installed PWA usage rates. I don't see the citation of the 1.2% rate I gave, so IDK if the usage rates were there too, but I commented  at https://github.com/WebKit/standards-positions/issues/619#issuecomment-5625310796 more about the strong usage of installed PWAs we see. I'll quote it here:

> To clarify further, we definitely don't interpret low install CTR from browser UI as a lack of user interest in installed PWAs. Here are some approximate numbers from internal telemetry
> 
> Currently, **~7% of Chrome users across all platforms actively use installed PWAs** (measured on a 7-day aggregation window). Given public Chrome active user figures, this represents hundreds of millions of people.
> 
> On **desktop**, normalizing by user time shows:
> - **~5%** of global user time is spent with an installed PWA open in a standalone window (foreground or background). This is without normalizing for users that use PWAs at all.
> - **~2%** of global user time is spent with an installed PWA active in the foreground. Similar non-normalization for users with pwas installed.
>   - *(Note: This excludes background playback like music PWAs, which would push this higher).*
> 
> Both usage and engagement time continue to grow year-over-year.
> 
> A significant caveat: these metrics undercount enterprise usage, where telemetry is typically disabled. Enterprises are heavy adopters of PWAs, frequently force-installing internal and line-of-business apps, to the point where we built dedicated enterprise policies to auto-install, launch, and keep PWAs open on OS startup.
> 
> Our takeaway is that installed PWAs are deeply valuable to users, and the low install rate from passive browser UI demonstrates that browsers alone are poor mediators for discovery compared to developer-initiated, contextual flows ([see my earlier comment here](https://github.com/WebKit/standards-positions/issues/619#issuecomment-4983052114)).



-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/1245#issuecomment-5625482689
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/1245/5625482689@github.com>