Why are Android devices missing on the web?

[minimo3]

According to IDC (and all the other analysts) Android enjoys a command lead in terms of worldwide smartphone market share. IDC for example pegs worldwide smartphone market share as follows: Android at 75% and iOS at 17%. Source: IDC http://www.idc.com/getdoc.jsp?containerId=prUS24108913.

Similarly in tablets, Android is supposed to have a 56.5% marketshare vs iOS at just 39.6%. Source: IDC courtesy of Business Insider http://www.businessinsider.com/android-ahead-of-ios-tablet-market-share-2013-5

Logically then Android should have significantly more users on the internet right? So how does this explain the latest Net Applications browser market share numbers:

Source: Net Applications

iOS still dominates with a 60% usage share that has stayed relatively constant. I can think of several reasons, in order of liklihood:

1. Analysts' guesses of smartphone/tablet marketshares are way off (entirely possible since no one except Apple reports unit sales/shipments). Analysts rely on calling up stores, randomly surveying a few thousand participants, or checking large distributors
2. The majority of Android device buyers in developing countries buy it as an upgrade to a feature phone and continue to use it as primarily a phone, text messaging device, without using the advanced features like apps/web surfing etc (I've seen many examples of this first hand as I travel)
3. Android users don't like to browse the web on their smartphones/tablets because of the poor experience (preferring instead to use their laptops or desktops). Its possible, but I don't think its very likely, especially since many users in developing countries may not possess laptops or desktops
4. Android users root their device and modify their user-agent string to impersonate a different browser. Yes, some geeks do that, but its definitely not widespread among the masses

If I were to guess, I'd think its a combination of 1 and 2.

 

[T5BRICK]

The numbers on your linked article are only from 1st quarter 2013. Remember that there are 6 years worth of iOS devices floating around out there. It took a while for Android to gain traction.

 

[paulsalter]

Not sure how other devices report data, my android devices don't report me as using a mobile browsers (various websites that say what browser I am using)

mine is shown as Chrome 27, the same as if I check Chrome on my desktop, which means it will not be included in these figures

 

[minimo3]

The numbers on your linked article are only from 1st quarter 2013. Remember that there are 6 years worth of iOS devices floating around out there. It took a while for Android to gain traction.

Fair enough, here are the cumulative numbers then:

In Sept 2012, Apple reported 400m total iOS device activations since the day iPhone 1 was launched in 2007 (source: http://venturebeat.com/2012/09/12/apple-84-m-ipads/), the next day Google reported 500m total Android device activations (Source: Andy Rubin https://twitter.com/arubin/status/245663570812100608).

In terms of run rate:

Android is supposed to hit a cumulative 1 billion devices by end of 2013, whereas iOS is forecast to hit 1 billion in 2015.

So total Android devices should have far exceeded iOS devices by now

 

[The-Real-Deal82]

Could it be perhaps that Android supply many low end smartphones and a certain percentage of owners are not using them for web based activities? I have older family members who own phones like the Samsung Ace and low end Sony devices and they don't even realise you can go on the internet with them lol. The figures look good for Android in terms of live devices out there, but I think my original point holds some weight as many budget phones are sold on the platform too.

 

[Jessica Lares]

The internet experience on an Android device isn't as good as on iOS. Mostly because the browsers have their own gestures built in, so it makes using the mobile version of sites really messy.

One example would be Google News. It has a pretty nice swipe interface. Well, you CAN'T use it in Dolphin because you swipe to the left, and you get your bookmarks, and if you swipe to the right, you get extensions.

Safari is simple, and because it is, it's nicer to use.

 

[minimo3]

Not sure how other devices report data, my android devices don't report me as using a mobile browsers (various websites that say what browser I am using)

mine is shown as Chrome 27, the same as if I check Chrome on my desktop, which means it will not be included in these figures

Yes, thanks for reminding me, I forget to mention that. I've checked the user agent string on Chrome and it seems like Chrome on smartphones report as "Chrome/27.x.x.x.x Mobile" whereas Chrome on tablets do not include the "Mobile" suffix. I noticed that on the Nexus 7 tablet. So, the chart may be displaying only Chrome usage coming from Android smartphones and not tablets. Still that doesn't completely account for the huge disparity in usage statistics.

 

[paulsalter]

If I check on my S2 using the stock browser, I am reported as running Safari

 

[Jessica Lares]

If I check on my S2 using the stock browser, I am reported as running Safari

Oh the irony!

 

[sentinelsx]

The internet experience on an Android device isn't as good as on iOS. Mostly because the browsers have their own gestures built in, so it makes using the mobile version of sites really messy.

One example would be Google News. It has a pretty nice swipe interface. Well, you CAN'T use it in Dolphin because you swipe to the left, and you get your bookmarks, and if you swipe to the right, you get extensions.

Safari is simple, and because it is, it's nicer to use.

One browser is hardly enough "fact" to stamp the browsing experience as subpar on android with so many choices around.


It is really simple. Compare the number of high end android phones versus low end ones. Now consider how bad of an experience the low end devices bogged down with skins alone give. Compare it to the high end experience each iPhone gives.

I am amazed that iOS is only at 60%, i expected 75% at least. Given the number of iOS devices still selling a lot (especially ipads), and given how different OEMs lock down the experience on android tablets (amazon wants you to run amazon store stuff, not the browser all the time for example), and the fact majority of consumers buy cheap $100-$200 outright android devices in many parts of the world (or as $0 promotions from carriers) on voice only plans, i am not surprised.

 

[skratch77]

A lot of androids get detected as iOS devices because they are using WebKit browser engine that apple uses.

 

[Jessica Lares]

One browser is hardly enough "fact" to stamp the browsing experience as subpar on android with so many choices around.

Of course. I was just using Dolphin as an example. The new Go Launcher browser suffers from the same issue. You should be using pinch-to-zoom to zoom in and out of the webpage, not to go to another tab, which is what I'm getting at. When it comes to third-party launchers, too many are trying to be the most fanciest.

 

[dejo]

A lot of androids get detected as iOS devices because they are using WebKit browser engine that apple uses.

Do you have a source to back up this claim?

 

[paulsalter]

Not sure I fully understand the chart on the first post

Why is Android listed as a browser (which Android browser is this meaning)?
Why is Apple /iOS not listed (obviously Safari is iOS only though)

The other browsers could be used on either

So Android could be the Opera & Chrome & Android figure

Similar iOS could be a combination of them

This is about Browser usage, not OS usage on the web

 

[Abazigal]

To be honest, I don't really do much web-browsing on my iphone either. Most of my internet experience is in apps like flipboard, tapatalk and misc apps that mimic the browser experience for certain websites that I frequent.

Do these count as time spent on mobile safari?

 

[Greg.]

Probably due to the iPad's dominance, since tablets are the main users of internet browsing. I do use the browser on my Nexus 4, but use Facebook, Twitter, Reddit Sync, Feedly etc. more, which I guess doesn't come into the statistics?

Also didn't/doesn't the iPad lack a YouTube app, so people generally use YouTube via Safari instead? If I look at my latest data usage cycle on my Nexus 7, the YouTube app was the biggest user of data.

Just thinking out loud, because the statistics seem kind of odd. I realise that there are a lot of Androids used as dumb phones, since they're cheap, so that's partially an explanation.

 

[strausd]

I don't have a full reason why this is the case. But for my usage, no browser on my Nexus 4 has been as good as Safari on my older 4S. User experience goes a long way here.

 

[adder7712]

There's a Safari string in the user agent of the default Android browser. Perhaps browsers of many Android devices are identified as Safari?

Plus on Firefox, the browser reports it's Firefox on Ubuntu!

 

[cynics]

I've read that some Android browsers report safari if the user has it set to desktop mode.

I used a desktop user agent when using my Android tablet. Mobile versions suck on tablets IMO and a lot of website won't give you the option to switch to desktop. Rooting is not required to change this setting btw.

Is there a website that I can visit that can tell me what my device is being reported as?

 

[Abazigal]

Probably due to the iPad's dominance, since tablets are the main users of internet browsing. I do use the browser on my Nexus 4, but use Facebook, Twitter, Reddit Sync, Feedly etc. more, which I guess doesn't come into the statistics?

Also didn't/doesn't the iPad lack a YouTube app, so people generally use YouTube via Safari instead? If I look at my latest data usage cycle on my Nexus 7, the YouTube app was the biggest user of data.

Just thinking out loud, because the statistics seem kind of odd. I realise that there are a lot of Androids used as dumb phones, since they're cheap, so that's partially an explanation.

YouTube (and similar apps like jasmine) exists for the ipad (though it was removed in ios6, and there are probably ios devices running around without them.

 

[paulbennett95]

I know a ton of people (mostly older relatives) who have Android phones, who essentially use them as touch-screen feature phones, ignoring all web browsing and apps. When I had my first smartphone last summer (an Android one) I never used the web browser because it was terrible. Now that I have an iPhone, I use it all the time (I have an android tablet too, and use that for web browsing at home [Firefox ])

EDIT: It seems Apple licenses out WebKit or something of that sort, and many browsers use WebKit, so maybe that's another reason?

 

[hallux]

Not sure I fully understand the chart on the first post

Why is Android listed as a browser (which Android browser is this meaning)?
Why is Apple /iOS not listed (obviously Safari is iOS only though)

The other browsers could be used on either

So Android could be the Opera & Chrome & Android figure

Similar iOS could be a combination of them

This is about Browser usage, not OS usage on the web

100% agree. This looks like someone making the numbers look the way they want them to look. You can't say "Safari had x% of users on the web" and then turn around and call out an entire OS and say those were the ONLY users. On a related note, there IS a Chrome for iOS, so lumping Chrome in with only Android would be inaccurate to a point. The best solution would be to pull the OS out of the browser string, though that may not be accurate either (as evidenced by Firefox reporting Ubuntu).

 

[Liquorpuki]

Could it be perhaps that Android supply many low end smartphones and a certain percentage of owners are not using them for web based activities? I have older family members who own phones like the Samsung Ace and low end Sony devices and they don't even realise you can go on the internet with them lol. The figures look good for Android in terms of live devices out there, but I think my original point holds some weight as many budget phones are sold on the platform too.

This is the reason

Low end Android smartphones have become the new feature phones

This is especially the case in developing countries where a big chunk of Android market share are whitebox Android handsets or phones sold without a dataplan

 

[minimo3]

100% agree. This looks like someone making the numbers look the way they want them to look. You can't say "Safari had x% of users on the web" and then turn around and call out an entire OS and say those were the ONLY users. On a related note, there IS a Chrome for iOS, so lumping Chrome in with only Android would be inaccurate to a point. The best solution would be to pull the OS out of the browser string, though that may not be accurate either (as evidenced by Firefox reporting Ubuntu).

In the chart, when NetApplications lists 'Android' they are referring to the stock Android browser (aka the one that ships with Android). When NetApplications refers to 'Chrome' they are referring to the Chrome browser running on Android. When NetApplications lists 'Safari' they are referring to any browser running on an iPhone or iPad including Chrome for iOS. This is because all browsers running on iOS are required per the developer guidelines to use the embedded Safari control.

Note: Safari, stock Android browser, and Chrome are all based on WebKit (which Apple developed out of KHTML/Konqueror) so all of them will list 'WebKit' in the user agent string. Many Android devices will also list 'Safari' in the user agent string. However it is fairly easy to discern which browser one is coming from by looking at the platform information, usually listed in parantheses near the beginning of the useragent string. For example:

User Agent String:
Mozilla 5.0 (Linux; U; Android x.x.x.x)...... WebKit..... Safari.... ----> Android stock browser - shown in Chart as 'Android'

Mozilla 5.0 (Linux; U; Android x.x.x.x) ..... WebKit.... Chrome x.x.x ----> Android tablet running Chrome - show in Chart as 'Chrome'

Mozilla 5.0 (Linux; U; Android x.x.x.x) ..... WebKit.... Chrome x.x.x Mobile ----> Android smartphone running Chrome - show in Chart as 'Chrome'

Mozilla 5.0 (iPhone; U; iOS x.x.x.).... WebKit.... Mobile Safari ----> iPhone running Safari or any 3rd party browser - show in Chart as 'Safari'

Mozilla 5.0 (iPad; U; iOS x.x.x.).... WebKit.... Mobile Safari ----> iPad running Safari or any 3rd party browser - show in Chart as 'Safari'

 

[paulsalter]

IWhen NetApplications lists 'Safari' they are referring to any browser running on an iPhone or iPad including Chrome for iOS.

So the figures for Safari are being overinflated?

Not sure how popular Chrome is on iOS, but in a report comparing popularity of browsers, If an iOS user is using Chrome, it should be reported as Chrome and not Safari