Ask HN: Why are PWA user install stats so hard to find?

I work as a developer / consultant. Sometimes I show «install as desktop app» on Edge/chrome on windows and people are genuinely surprised, and don’t know about it. I’d bet the conscious usage is very low.

Chrome(OS) and others might use PWAs behind the scenes, so depending on how you count, maybe the numbers are much higher than I’d think?

Company I write code for has an in-house app that's used on tablets by employees on retail floors. It's a platform agnostic PWA. Getting the managers to actually create a home screen shortcut to use it full screen always requires someone to walk them through that. It's just surprising and not intuitive to most people that a page in a web browser can be made to act like a standalone app.

To be clear, it's not really critical. They can do their job perfectly well using the app in any browser. But having it full screen does improve the UX quite a bit.

I am also thinking that installing is exactly what people avoid when they prefer PWAs, so if installing is actually just creating a shortcut to home screen (and Windows equivalent), then maybe the verbiage should reflect that and we would see wider adoption.

On my RadioSide, when the user selects to add to home screen or install as an app, it makes an icon and that's that, from the user's point of view it looks like an app, works like an app, quacks like an app, without the risks and hassle.

Most users have no idea that it's possible thus presenting them with an "install this" they already get anxiety attacks of downloading and running stuff and taking risk, I am talking about average users.

As a (non-web) developer, I just don't trust PWAs: I assume they'll break their offline functionality in the most unfortunate moment while being not as tightly integrated into desktop as proper apps (even those using Electron). So I just keep using website as that has a more clear mental model.

Absolutely, I’ve yet to see something that works offline (I’m sure it’s technically feasible). But «so what» - few modern day apps do much use when offline.

In my whole life working in IT, I’ve seen two people use PWAs. And one was just trying it out to see if it was better than a website.

> Apple said PWAs had "very low user adoption"

Installing PWA isn't obvious for users on iOS Safari or even macOS Safari. It's very obscure

> A Chrome OS ... desktop PWA installs have grow by 270%

Google Chrome would "beg" you to install PWA if the web was developed in the correct way with correct manifest and ServiceWorkers.

It’s also subtly inhibited by Apple after installation. For example when I pull down on the Home Screen to get Siri Suggestions it will refuse to show me any PWAs even though they are the most used apps on my device.

> It's very obscure

and i might argue it's deliberately so!

Open Website. Click Share. Click Add to Home.

Doesn't seem that obscure and for me the issue is much more with websites not making it clear when a PWA version of the site is available.

You're not sharing it with anyone, so that does seem pretty obscure.

Share is the button to bring up the menu with myriad actions like bookmarking, adding to reading list, printing, running third party actions, etc. Has been for years. Obvious? Maybe not. Obscure? Also no. There are very few buttons you can click to begin with: it has to be either AA or Share.

Share section has always meant more than just share with other people.

It includes adding to favourites, publishing the page to other apps etc.

Took me months before finding "Find on Page". It's in the share menu, of course!

You can type what you need to search in the address bar, and then scroll down in the autocomplete to "On This Page".

It's been there forever, but still quite obscure

I would have never known the On This Page option existed if you didn't say so now.

I just found the find on page option in the share menu recently after a couple years of iPhone.

I don’t share with that at all, and I don’t share with my home screen either.

Why are Apple’s gestures and menus so undiscoverable?

I learned about this myself either by accident or from "X things you didn't know your iPhone could do" many years ago.

Otherwise it really is undiscoverable

Oh, and you omitted that you need to scroll down the share popup. A popup that primes you as scrollable sideways, don't forget.

When my family was discussing getting an older member a smartphone, I briefly thought about it and realized that the whole lore of what's srollable would be impossible to explain. If you don't want a feature used, put it "below the fold" in a popup.

It’s a pretty clunky call to action, though. The simplest thing for the user would be a button right in the message telling you a PWA is available. “Tap here to install” and it’s done.

Instead, you need to tell users “Hey, we have an app! No no stop, wait, don’t go looking in the App Store it’s not there. Instead you need to hit the share icon. Yeah it looks like a box with an arrow coming out of it, down the bottom of the screen. Yeah if it’s not showing you need to tap the gray bar at the bottom first. Then the share button. Then look through for something that says install to Home Screen. Yeah scroll down a bit and maybe scroll across until you see it. Yeah there you go.”

"Tap here to install" is how Chrome does it.

True. The context of this thread is discussing the process on iOS, and the hypothesis that it leads to lower adoption of PWAs on Apple's platforms. From the original comment:

> > Apple said PWAs had "very low user adoption"

> Installing PWA isn't obvious for users on iOS Safari or even macOS Safari. It's very obscure

No, that's not the sequence on ChromeOS. See this example

https://medium.com/@julianneagu/installing-pwas-on-chrome-ea...

Share doesn’t seem strange in that context?

Recent interesting thread about PWAs:

Ask HN: Who has had a successful PWA product?

https://news.ycombinator.com/item?id=40724774

Great thread, thanks!

My guess is that indeed the usage is low and people didn't bother to update stats or put more efforts into it. But the real reasons, in my opinion --

On the desktop, PWAs are simply less relevant. People are mostly online, and there is little difference between PWA and a "real" website, so people would just go to the website instead of "installing" it. Most "heavyweight" applications like VSCode, Photoshop and Premier Pro need to be run natively (although many of their features are available in the web version). Google Docs, probably, but otherwise there is very little incentive for using PWAs.

On mobile, many of the same points apply, although PWAs have more access to permissions and are often viable replacements to native apps (especially apps that are just websites anyway, like Uber Eats). I think mobile apps would be the ones driving the growth of PWAs. But of course... Apple. Apple doesn't want this ecosystem to grow to take revenue from the app store, and just made it difficult to develop and use PWAs. (There are legitimate concerns over how PWAs are used, but there are also solutions.) As a developer you don't want to spend time developing PWAs that only run on Android phones, so you might not bother at all.

I'd rather have tabs and forward back and history and bookmarks and extensions. Losing the browser & switching to an app seems like a colossal downgrade, like losing everything that's good. I do not understand PWAs.

> I think mobile apps would be the ones driving the growth of PWAs. But of course... Apple

Android is the dominant mobile OS in the world.

> As a developer you don't want to spend time developing PWAs that only run on Android phones, so you might not bother at all.

You will be developing two apps anyway, so...

The only ones I’ve ever used are gmail and google calendar on my work computer, and that’s because we don’t allow third party fat clients.

I’m not looking to use PWAs.

Because PWAs are mostly hype originally aggressively promoted by Google, but now even Google is tired of promoting them.

One of the few places where hype is still real is Hacker News, and even here people can't even come up with examples of good PWAs beyond, and I kid you not, Twitter.

Going from 1 install to 270 installs is a 270% increase, too. So I would take such stats with a grain of salt weighing about 270 tons. Google is dominating mobile market and Chrome is dominating the browsers market, and that's the best stats they can come up with?

> Going from 1 install to 270 installs is a 270% increase

That's a 26900% increase.

https://news.ycombinator.com/item?id=40960124

> Going from 1 install to 270 installs is a 270% increase

No, it's a 270x increase. In percent that's much higher.

Sorry, it was a brain fart on my part.

If we do the 270% increase from 1, the stats become funnier.

I personally very much enjoy using PWAs for things that are actually more like apps than websites. Things like Discord could benefit from a PWA version, but the ones I actually do use are for standard Fediverse pages, YouTube, Chrome Remote Desktop, Pokemon Showdown and a bunch more.

Sadly I do think that PWAs aren't very popular in the grand scheme of things since I think they often make more sense than a separate app that you download just for it to basically be a copycat of the website and to still not work offline.

I don't think anyone actually uses these in the real world.

My understanding is that a PWA is just a website with a few extra features. The ability to "install" it is one of those features, but it's still a PWA even if you don't install it and just use it as a normal website. So basically a website with a service worker that has some offline functionality is a PWA. Tons of people use those without even knowing it, and there's no good way to measure that.

I’m a little confused. Most people are talking about PWAs as though adding a website shortcut to your Home Screen is a PWA. When I search for what a PWA is on Google, it says they can be placed in the app stores. So are they in the app stores or do users have to do the shortcut thing?

Only Google and Microsoft app stores. Apple does not allow them.

There are some tools which allow you to package up a pwa into an app that is essentially a thin shell around a webview. Whether or not Apple accepts such an app into their App Store is another variable. For example:

https://docs.pwabuilder.com/#/builder/app-store

Guess the place to start is find out how many sites support PWA behaviour — could do this by finding all the sites in the HTTP Archive that have a manifest and scanning them

Then you’re going to beed to see what stats they’re willing to share

Quite a few sites seem to support PWA even on iOS — that’ how I use Instagram on my iPad

Because the amount of people that install PWAs is too minimal to measure meaningfully.

At my previous consultancy client, a B2B SaaS, we thoroughly implemented our products as PWAs...

A big, big part of what needed to be done was explaining what & why to users, we figured since it's B2B we could afford the extra end-user education.

In a B2C context though, yeah I think the adoption rate is just very low due to its obscurity, most people don't know it exists, and me, having developed multiple PWAs, don't even use a single one

What was the uptake like? Would be super handy to know how well it was tolerated or if there were any unexpected surprises.

I’m pretty much decided on Capacitor for my new B2B SaaS, but I honestly had forgotten about the possibility of PWA deployment.

Updated to say it seems people find anecdotes useful and interesting, my apologies for discouraging this.

Previously I wrote:

> Replies with personal anecdotes and opinions are not super helpful, sorry! I'm specifically looking for reliable empirical measurements and data.

K. HN is not your personal search engine or whatever. That’s not how social networks work. You aren’t just going to get crickets if nobody has what you’re after, which seems quite likely given you could count on one hand the number of organisations that could reasonably be in a position to have this data, and given that you would’ve searched before asking this.

It sounds like you’re trying to backsolve for a desirable answer. You can make quite educated / justifiable inferences, not much guesswork at all, to come to the conclusion that the numbers are for most intents and purposes immeasurably small.

Why would PWAs reasonably be used by anyone other than idealistic nerds? You’re kidding yourself.

But they are interesting, so I’m glad they’re posted.

Because they would easily grow bigger than the duopoly App stores

I'd hazard a guess that one key reason PWA apps are not more popular is the lack of an app develepment framework that would enable more sophisticated offline behavior.

I suspect the overlap between PWA users and those who value privacy and have the technical means by which to achieve it is high.

I searched for weather in my area in Google then added that page as a PWA in Android.

ok