Business Values Of PWA's

20 Jun15:55 – 16:30Stage: Main StageTalk

Checking session availability…

Hang tight while we load the latest updates.

In this session, Alan Semenov, Development Lead at Enonic, talks through the business value of PWA's and why companies should start adopting them.

Business Values Of PWA's

Alan Semenov at UXDX Community: Oslo. Video: https://youtu.be/-7r4NX-9gvw

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

What a progressive web app actually is

[00:00:00] Hello everyone, my name is Alan and I work for a company called Enonic. It's a Norwegian[?] company that works with CMS[?].

[00:00:22] Today we're going to talk about business values of progressive web apps, and here's the question. Who knows what progressive web apps are? One, two. Yes, great, you're my audience, you don't know anything about it.

[00:00:42] Okay, before we talk about what a progressive web app is, let's talk about what a web app is. According to Wikipedia, a web app is the computer program that runs in the client and that runs in the browser, which kind of implies that it has to have access to the internet, it has to be online and it has to run inside the browser.

[00:01:22] So when we start talking about progressive web apps, that's where the confusion starts. Because it's this word, "progressive", that causes a lot of confusion. People tend to get really confused. Is this just a sales speech? Is it just a cool word? What is that? Why should a web app be called progressive?

[00:01:58] But in fact, the progressiveness of a web app means that it doesn't need a browser to run. You can take the browser out of the equation. In fact, it doesn't even need the internet at all. It's a self-contained app, which can work offline, which will use up the entire space on your device, and which will still live even when you close the app. The process will still have access to all the resources of your device. So basically, for a regular user, it will look as if you would just be using a normal mobile app.

The state of the web then and now

[00:02:55] But why would a web app mimic a native mobile app? To answer this question, we have to look at the progress of the web over the last five, seven, ten years. Seven years ago, having wasted God knows how much money trying to build a web version of the Facebook app, Mr. Mark said the following: that betting on HTML5 was one of the biggest mistakes we've made. And then here we went on making the mobile versions of Facebook. And that was the year 2012. Not so long ago, but still. In terms of web technology, it's quite a long time.

[00:03:48] And we cannot blame him for saying that, because if we look at the state of the... Oh! Oh my God! Yes, I lost the projector, of course. Okay, but basically, we cannot blame him for the state of the web in 2010 to 2012, or even five years ago. It was quite miserable. No offline support, no access to device hardware. A website could only run inside the browser. A web app was not installable on the device. A web app was not designed to run inside a smaller device, an iPad or a cell phone. And there was no user engagement whatsoever, no interaction with the user. So yeah, that was quite a sad time.

[00:04:57] But if we look at the state of the web now, in 2019 or the last couple of years, things are looking much better. Web apps have offline support, so they can actually work offline, which sounds a bit weird. How can a web app work without access to the web? Web apps have access to most of the device's hardware: Bluetooth, camera, storage, almost everything. It depends on the device, obviously, on the platform. But still.

[00:05:43] They don't need a... Wow, okay. We're up. They don't need a browser, as I showed you. They can actually be installed on your device. You will actually have an icon on your home screen to launch the app. They are responsive, so they will fit the format they are running in. And user engagement via push notification. So basically, you will not notice the difference between a mobile and a web app. You can read more at this link and you will be shocked at how much your web browser or your web app is actually on Google Earth[?] nowadays.

Costs across the application lifecycle

[00:06:38] So if now, in 2019, web development has mostly caught up with mobile development, it means that you as a developer can build the same app both as a mobile app and a web app. And your user will not notice any difference, basically.

[00:07:06] But if you own a business, what are the costs? How much is it going to cost you as a business owner to build a mobile app or a web app? Should I build both? Should I maintain both? To answer this, we have to look at the application lifecycle and basically analyze the costs at every stage, for both technologies.

[00:07:35] Build. Building a mobile app means building at least two versions of it, one for iOS and one for Android. And these two platforms have totally different tech stacks, which means that you will need two development teams to build both apps, even though they will share the same code base. And mobile developers are expensive guys. So this will cost you. When it comes to building a web app, it's simple JavaScript and HTML and CSS. And any IT student can do that nowadays. So building a progressive web app is very cheap.

[00:08:33] Deployment is another sad story. If you build your two versions of the mobile app and you want to deploy them to the app store, this is a very long and cumbersome process. On the left is a diagram of submitting the app to the app store, to the Apple store, and on the right is Google Play. It will take you at least two days, in the best case, before your app will be in the store. So it's not easy. And of course, if you want to roll out an update, you have to go, not the entire way of course, but there's a separate process for that. So it will take time.

Discoverability, and the apps nobody downloads

[00:09:32] And then let's say that you succeeded. Your teams have built the apps, you deployed the app to the stores, and it's there. You will face another big issue, and that's getting your app discoverable, getting users to actually know that your app really exists.

[00:09:59] This is the indexual[?] statistics. And according to this, regular users don't download any apps in an average month. So if you try to remember the last time you downloaded an app, maybe it was a long time ago. And even if they did download your app, most likely they will open it once and just forget about it. This is the statistics showing how often people open the app after they have downloaded it. You see it goes all the way down. I'm not talking about Facebook, obviously. More than 75% of the apps that you downloaded will only be opened once.

[00:10:59] Here's even more mind-blowing statistics for you. That's probably the best one: no diagrams. There are almost 2.5 million apps in stores, but 60% of them have never been downloaded. So if you try to just get your head around the amount of work and money invested in this, you will be shocked. So the question we have to ask as developers, as business owners, is why spend money and resources on something that may not be used by anyone.

[00:11:47] So if you as a business owner want to get your app recognized, known, then you have to promote it, and it costs money. Again, remember we are talking about the mobile apps now, not the web apps. When it comes to web apps, they are easy to deploy and update, because when we build it, we simply deploy it to the server without having to go through any stages of approval or verification, nothing. And then when the app is on the server, it's very easy to share it. You just send the link to your app, because it's basically a website. So it will be indexed by the SEO, it will be indexed by the search. It's always just like a regular website. And you can send the link to any page of your app to your friends, basically. So it's very easy to get the app discovered.

Summing up the features and the business case

[00:13:04] That's my summing up. When it comes to the features, to the more technical aspects, I would say that progressive web apps have caught up generally with mobile apps. We talked about the offline support, we talked about the access to hardware. They are installable, they are responsive, they provide user engagement and they are safe. And last but not least, if you think that, okay, maybe it's safer to have my app in the app store, no, it's not really like that, because progressive web apps require HTTPS. So you should not care about the security.

[00:14:00] And when we talk about business aspects, we talked about that building a mobile app is expensive, while with a web app it's very simple. Maintenance, same thing. Deployment, quite hard and takes time for mobile apps. Promotion, again, costs money, and costs nothing for the web apps. Then installing the updates is also, for you as a user, going to take time to update. You have to go through the app store and log in and download the update and everything. While with a PWA you just go to the app and the app is on your device. No installations, nothing.

Success stories: Tinder, Pinterest, Alibaba

[00:14:54] So we talked about business aspects and it's nice to see tables. But you probably want to hear about some success stories of businesses that actually succeeded from using progressive web apps. Tinder cut[?] load times almost by three times. And the size of the PWA is much smaller than the mobile app.

[00:15:29] By the way, talking about the size. Do you know how big the Facebook mobile app is on your device, while we're on it? Any guesses?

[00:15:41] Audience: 200 megabytes.

[00:15:42] Alan: How big?

[00:15:43] Audience: 200 megabytes.

[00:15:44] Alan: More? More? No? It's 450 megabytes on your device, without cache, when you download it. And the mobile version is 4 megabytes.

[00:15:56] Pinterest rebuilt their entire mobile app as a progressive web app. And they improved basically everything: the revenue and the time that users spend on the site, on the website. It's a good example. Basically, increases in everything. It's faster, it's a better bounce rate, and there's an increase in how many pages a user views per session.

[00:16:47] And finally, to the e-commerce behemoth Alibaba and AliExpress. They almost entirely rely on progressive web apps, because they're in Asia with not so good internet. It's very important for them that users can use the app on low bandwidth. So as you can see, the statistics are just crazy. AliExpress improved the conversion rate by 100%. That's crazy. Just think about how much money this is.

[00:17:28] This is an empty slide for some reason. Okay, so if you want more statistics, if you want to surprise your boss, then go here and just show it to them. It's tons of cool stories about successful implementation of progressive web apps.

[00:17:58] And if you're still not convinced, here's last year's news for you. Facebook is testing a progressive web app for its mobile website. So the very same guy who just seven years ago said that our biggest mistake was to bet on HTML5 came back to it. And I think it's a very good proof that web development reached, maybe if not its peak[?], but at least it surpassed native development. Yeah, that's basically all I had. Any more questions?

Q&A

[00:18:54] Audience: Have you built a separate stack for the... I mean, and you have implemented...

[00:19:00] Alan: We never had a mobile app, so we've given up before we tried, because you have to think a bit ahead, right? And we didn't have any resources that were capable of building a mobile app basically. And if we wanted that, it would mean hiring at least two teams, and it's simply not worth it. It took us maybe a week to build our first one. It was very simple, and it just works.

[00:19:45] Audience: So what can apps do that PWAs can't[?]?

[00:19:55] Alan: They don't really like Apple at the moment, or the other way around. Apple didn't like progressive web apps up until last year, because they obviously saw this as some threat for the app store, which is actually a very good source of income for them. But since all other modern browsers added support for it, they didn't have any other choice than to jump on the train. But they were very late. So with every new release of iOS they add some features to it, but they're still kind of lagging behind. So if you have an iPhone and you install a PWA, you may notice that some things may not work. But if you're on Android, then you'll be fine.

[00:20:52] Audience: I have one question. What did you do in terms of driving adoption for a progressive web app? Because I think everybody just knows to go to the app store and download an app, but this is very new, progressive web apps. So was there any sort of messaging or marketing or something that you did to drive user adoption?

[00:21:08] Alan: You mean like me as a company, what would I do? First of all, it's the SEO[?], and it's the same as what you would do with websites basically. So this means any kinds of promotions on Facebook, on LinkedIn. You can basically just provide a link to the app or to some specific pages, something you would not be able to do with a mobile app, right? So you still have to promote, obviously, but it's much easier, because there's indexing. Mobile apps cannot be indexed. You have to point the user to the app, but with the web it's much easier.

[00:22:04] Audience: Thank you very much.

[00:22:04] Alan: Thank you.

Speaker

Alan Semenov

Alan Semenov

Development Lead

Enonic