Thinking About Web Browsers and Stickiness (Premium)

After an interesting experiment this weekend, I might be returning to Google Chrome. But whatever happens, this episode was a good example of why stickiness matters so much to platform makers. And what can happen when your products aren’t sticky enough.

So, what does it mean to be sticky?

Stickiness is an often hard-to-characterize aspect of a product or service that makes it difficult to switch to the competition. Building in stickiness is core to any platform. And all platform makers do this to some degree.

Stickiness takes many forms, of course. In a perfect world, you will love some product—say, a Sonos speaker–so much that you want to buy more of them. And in doing so, you are becoming a “Sonos household,” at great expense, and are now less likely to switch to some other speakers. In part because Sonos is great. And in part, because doing so would be prohibitively expensive. Sonos is sticky.

Platform makers also game the system to make their products and services stickier. In the Microsoft world, for example, product bundling was about extending the stickiness of popular Microsoft products—Word, Windows, and so on—to other products and services. It’s why the Microsoft Office suite exists, because users would otherwise be slow to embrace certain lesser-known Office offerings. And it’s why Microsoft bundled a web browser, a media player, and a communications tool in Windows. It wasn’t enough for Microsoft to just show up in these new markets. It extended the stickiness of popular products to new ones.

We see this in the smartphone world, too, of course. Google requires hardware makers to include key Google apps—and, as important, to make them the defaults—if they want the “full” Android experience with the Google Play Store (which they do). And it’s why Apple doesn’t allow users to configure non-Apple default apps in iOS: It lost the stickiness battle in the PC wars, and it doesn’t want to make that mistake again.

On the flipside, I’ve argued that digital personal assistants are not all that sticky. That they should all work nearly identically and with whatever smart devices and services that you own. The stickiness there is in those other things, not in the assistants. And if were to switch from, say, Amazon Alexa to Google Assistant (or vice versa), I bet the transition would be easy and fairly seamless.

But what about web browsers?

When web browsers first appeared in the 1990s, they weren’t particularly sticky in the sense that the web wasn’t yet mature or full-featured. But sensing a possible extinction event for Windows, Microsoft moved quickly to bundle its own web browser, Internet Explorer, in Windows. (Apple later did the same on the Mac, and even Linux distributions will come with some default web browser.)

Microsoft’s strategy worked, at least at first. And while it later ran afoul of antitrust regulators around the world, the damage was down: Netscape was defeated, and web browsing became a feature of Windows and other desktop systems.

Things changed, obviously. And without getting too bogged down in the history, let’s just say that a combination of things had to happen for web browsers to become sticky again. First, Microsoft had to basically abandon Internet Explorer, giving rise to more functionally useful third-party web browsers like Mozilla Firefox and, later, Google Chrome and Apple Safari. And second, it had to do this just as the web was maturing into a major platform in its own right, into the thing that Netscape saw as the future.

When it comes to web browsers, the biggest failure for Microsoft is that the terribleness of Internet Explorer literally helped its customers overcome the natural stickiness of using what was bundled in Windows and installing, using, and then preferring third-party web browsers instead. Think about that. You as a power user may have no compunction at all about testing different products regularly. And you know that I preach the importance of this kind of continual evaluation, of embracing change. But normal people? They tend to just use the thing that comes with the thing. It’s a fact that Microsoft relied on for decades.

So consider the incredible fact that the number one desktop web browser, Google Chrome, is not preinstalled on about 97 percent of all desktop PCs (which includes Macs). And that Google Chrome is nonetheless responsible for about two-thirds of all desktop-based web browsing. In at number two, Mozilla Firefox is not preinstalled on any desktop PC or Mac, and yet it still has over 13 percent usage share, ahead of bundled browsers like IE, Edge, and Safari (on Mac).

Chrome’s success on the desktop is thus all the more impressive, and it says a lot about the changing habits of PC users. As Terry Myerson noted during an explanation for Windows 10 S, Microsoft knows from its telemetry data that Windows 10 users spend over half their time in the system using a web browser. The problem for Microsoft, of course, is that that web browser is usually Chrome or Firefox and not one of its web browsers.

Looked at from a different angle, our usage changes over the past 20 years show that Windows has become ancillary to the web for many, an interesting reversal. And that the web, and not Windows, is stickier these days. That is, many of could utilize the same web apps and services on a Mac, a Chromebook, or even in Linux, and be perfectly happy.

The web is sticky because it’s vital. But again. What about web browsers?

I’ve argued recently that web browsers aren’t all that sticky. Yes, there is some effort to switching browsers, but most modern web browsers can import almost anything from other browsers, which makes the process of switching easier. I find it relatively painless to move between Chrome, Firefox, Edge, and even Safari (on the Mac) from a data portability standpoint. The trick is finding the browser you prefer. (And adopting a mobile strategy that makes sense, too: If you use a certain web browser on your PC, using the same browser on mobile will offer the most seamless experience.)

As you may know, I recently switched to Mozilla Firefox. And I really like this web browser: The look and feel is very Edge-like, which I prefer, as is the excellent text rendering. It performs as good as Chrome, and it uses fewer resources. The new Firefox has a lot going for it.

But like Brad, who also switched to Firefox but ended up switching back to Chrome, I have certainly run into various issues. For example, I very much prefer using Chrome’s desktop shortcuts for web apps in Windows, and I rely on these apps every day. And Grammarly, which I also rely on, isn’t working correctly in the new Firefox.

So, I’ve been compromising. I continued using the Chrome-based web apps—Google Inbox, Google Calendar, and Twitter Lite, primarily. And I created a WordPress web app shortcut, too, so that I can use Grammarly when I post articles (in Chrome now). What this means is that I’ve been using both Firefox and Chrome in Windows. Not just Firefox.

And then it occurred to me. If I am OK using Firefox like this, would Microsoft Edge work in a similar fashion? After all, some of the things I like about Firefox is that it looks like Edge, and renders text like Edge. With Microsoft recently reporting battery life results that show Edge outpacing Chrome and, by a wider margin, Firefox, maybe it would make sense to look at Edge again.

So I did.

To test Edge correctly, to use it like I was using Firefox, I made Edge my default browser on both my primary desktop and on the Surface Book 2. I wiped out all the bookmarks in Edge and started over from scratch by importing my Firefox bookmarks as well as my Chrome-based bookmarks and other data, and then configuring things in Edge as I preferred, with the right extensions, Google Search, and so on. And then I tried to use Edge. Again.

Sorry, Microsoft fans, but it took only a few minutes of use on both PCs to remind myself why I’m not using this web browser. The performance is pokey compared to that of Chrome or Firefox. And there are tons of goofy little things you can’t do, like configure the home page or new tab page the way I’d prefer. (I use an extension called Momentum in Chrome and Firefox). The developer tools aren’t as good as those in Chrome either (and that was true of Firefox, too, actually). Edge just isn’t as good as Firefox (or Chrome). It’s not even close.

But this made me think.

I recently wrote that Edge’s battery life advantages over other browsers have not helped it gain usage share. And that this may be an interesting issue for Windows 10 on Snapdragon: After all, the only meaningful advantage of an ARM-based Always Connected PC over an Intel version is that the former will get better battery life.

But PC battery life is generally fine for most people. Plugging in a PC is not onerous, and we do this with all our devices, especially phones. In my recent Surface Book 2 review, I noted that that high-end PC SB2 delivered over 15 hours of battery life. But 10ish hours is normal these days for PCs, overall.

And if the choice is between 20 hours of battery life with slow performance (Snapdragon) and 10-13 hours of battery life with great performance (Intel), most will choose the latter, I bet. That will be an issue for Qualcomm to address, of course. But it may also be an issue for Firefox.

Looking back at the Edge battery life results, I see that Edge narrowly beats out Chrome, with 16 hours of life vs. 13:30. But Edge is otherwise pretty terrible compared to Chrome. (Edge more dramatically beat Firefox’s battery life; in that case, it was under 10 hours.) So when you compare these three browsers holistically, it is Chrome, not Edge—and not Firefox, actually–that is the best browser overall. Obviously.

And now I’m considering simply using Chrome. It will be one less thing to install and manage, both on my PCs and my mobile devices. I will lose some things I really do like, the look and feel, the text rendering, which I love. But the convenience—the stickiness—of Chrome is real. And it’s real for the right reasons: It’s just better. One chooses Chrome because it’s better; one doesn’t just use Chrome because it’s bundled with the OS. (Though this probably happens on Android, I guess.)

So we’ll see how it goes. I’m going to experiment this week going back and forth between Chrome and Firefox, on both Windows and mobile. But I think I’m going to find that the matrix of compromises that either choice triggers will favor Chrome. Not because Firefox isn’t great. But because Chrome just delivers so wonderfully where it ultimately matters most.