@yaypie, what's the current breakdown on desktop vs. mobile traffic for Cake?
Our mobile vs. desktop split is actually really close to 50/50, with mobile typically just barely beating desktop.
We've taken a lot of care and spent a lot of time making sure that the Cake website works great on every device, but it hasn't been easy, and it imposes a lot of design constraints. In many ways I think those constraints result in better designs for everyone, but it can be a real challenge sometimes.
For us, I think a native app provides two primary benefits that our web app really can't compete with:
App Store distribution. There are a lot of people who simply don't want to use a web app, for whatever reason. Having a native app in the App Store allows those people to discover Cake, and having an app icon on their screen helps prevent them from forgetting about Cake.
Push notifications. Push notifications let users know when important things are happening and keep them coming back. As you said, @ChrisJenkins, browser push notifications are possible, but the problem is that the browser push notification user experience kinda sucks. The built-in authorization prompt to enable them is ugly and annoying, the notifications themselves are really ugly in some browsers, and iOS doesn't support website push notifications so they're not a good solution for iOS users.
There are other things that are possible in a native app and still can't really be matched by a web app, but they're mostly minor.
I think the best path for a service like Cake is to have both a great web app and great native mobile apps, and to not try to railroad people into using one if they're happy with the other.
Fascinating, Chris. This is kinda long but interesting:
One point he made is we've trained an entire generation to search for your service/app first in an app store. I'm always curious when people message me and say that they looked for Cake but couldn't find it, and I point out that it works just fine by going to cake.co. Some of them have said they didn't think of that, maybe because for services like Instagram you really have to have the app.
I get the feeling Apple has been trying to crack down on apps that aren't native enough, no?
His experience matches my own, and you see we come to similar conclusions about the state of apps on the web.
All of the problems mentioned which prevented him from just deploying a strict PWA?
The hardest store to get an app in, and the most expensive? Apple.
The userbase most likely to not think of using the browser? Apple. (And of course, given the state of Safari, no surprises there).
Speaking of the hypothetical Cake iOS app...it's now 100% less hypothetical!