Android, China, and the Wild Wild West

Last week, I talked about the importance for us industry observers, analysts, media, etc., to have a more informed discussion when it comes to Android. I think it is important when we analyze, from an industry and market viewpoint, that we do so with a holistic viewpoint.

My key point in last weeks column was to address the issue of Android platform forking. Android in its purist definition only refers to the AOSP or Android open source platform. Something anyone on the planet can take for their own and fork it, thus differentiating their Android platform and in many cases using the core Android source and making their own platform. Therefore, as it currently stands we have Google with a platform based on Android, we have Amazon with a platform based on Android and we have Barnes and Noble with a platform based on Android. Each of these platforms is their own unique ecosystem.

I make this point because when we say Android has X% market share we are talking about the total including all the forks. This is a key point, because when many make the claim that Android is winning the market share game, they often make the mistake of assuming that Android equals Google, therefore assuming that Google’s version of Android has the total Android market share. This is of course false, as Google’s version of Android, the one that benefits Google in a monetary or data gathering way (a.k.a a business model), has only a fraction of the overall Android market share numbers being referred to. Exactly how much we are not sure because even Google refers to Android falsely making it sound like the total installed base of Android devices on the market have some business benefit to Google and of course that is not true. My gut tells me that if Google did release the numbers of the global install base of Android devices tied to their services, thus qualifying as a Google Android device, the picture would not be as rosy as many make it out to be. No where is that more the case than in China.

The Wild Wild West

As I have been studying the Chinese Android market, the only way I can describe it is the wild wild West. Android is fragmented, un-unified, inconsistent, and otherwise fundamentally fractured in as many ways a platform can possibly be. In fact it is hard to even call Android a platform in China, and there is certainly no Android ecosystem there. There are dozens of app stores, tightly controlled ISP and heavily differentiated experiences and services bundled on the vast majority of Android devices, half a dozen different payment mechanisms, and a general lack of standardization.

The top app stores come from the likes of Tencent, 360, 91, UCWeb (which is a browser) app store and a number of other tier two heavily localized app stores. If I was an Android developer focused on China, I would have my work cut out for me making sure I was present in all the various app stores, or try to go direct to consumers (as many are trying to do), or working as close as possible with the ISP and carriers themselves. This model is somewhat feasible by the larger developers but very difficult for the upstarts and other smaller developers.

What is also very interesting about the Chinese market for Android devices is that the vast majority of the 38 million Android devices sold in China last quarter were extremely low-cost entry level devices. Now, in most cases, this is exactly the kind of scenario that Google would hope for. Google’s mobile business model depends on install base and the best way to do that is to have a plethora of cheap devices so hundreds of millions of people can jump on your platform and you can make some mobile search and ad revenue. The only problem is Google is not benefitting from Android’s success in China in even the slightest way.

The challenges of Google with China are well documented. Over the past few years Google has continually been closing offices in China and largely abandoning the region. Android has not helped relations or Google’s strategy–or lack of strategy–in that region and it doesn’t appear that it will anytime soon. The vast majority of Android devices sold in China have been stripped of all services tied to Google in any way. Here are some key points.

– Local browsers dominate the web browsing landscape
Google search engine market share is less than 5%
– 90% of new Android devices sold in China do not have the Google Play store on them.
– Many developers are choosing local in app advertising solutions over Google’s

China, and in particular the low-end Android segment, is one of the fastest growing segments in mobile. Every day China is accounting for more and more of the Android activations. Android in China has simply become such a customized and regionalized OS that I’d argue the point that Android in China should be considered its own fork. And due to the extremely fragmented and lack of standards around app distribution, I’m not that confident that Android has a sustainable position in the region outside that the devices are cheap. The vast majority of low-end Android consumers in that region are not investing into any specific ecosystem other than the likes of someone like Baidu, for example, which offers their services on a range of platforms, Apple’s included.

Other than Android devices being extremely low-cost, I’m not convinced, based on the data I have on the region, that Chinese consumers are loyal to the regional Android fork. A point, that offers more hope for standardized and unified platforms from competitors like Apple and Microsoft or even some platform not yet released.

The bottom line is, for now, Android is alive and well in China. It represents one of the fastest and the largest growth sectors for not just Android but the mobile market at large growing at about 300% year-over-year. Android is being taken by the natives and customized / implemented to benefit themselves and their heavily regional services. The vast majority of these devices have little to no benefit to Google. Android is doing well in China, Google is not. Something I find fascinating.

I paint this broad picture of Android in China for the hopes that we can have a more informed discussion when we discuss Android. Too many people associate Android’s holistic global success with Google and that is a disingenuous analysis. I’d love to be able to break out the individual Android fork market share, including the regional forks like China, India, and now Africa, but when the handset OEMs–and Google–are not sharing specifics. A situation I find entirely suspect. Although, the more I learn the truths about Android holistically across the forks and the regions, I am getting a sense of why the details are not being shared with us.

Published by

Ben Bajarin

Ben Bajarin is a Principal Analyst and the head of primary research at Creative Strategies, Inc - An industry analysis, market intelligence and research firm located in Silicon Valley. His primary focus is consumer technology and market trend research and he is responsible for studying over 30 countries. Full Bio

35 thoughts on “Android, China, and the Wild Wild West”

    1. I wish I could say. Those are both good questions and I just don’t know. I’d say it seems more like they just don’t care. Google isn’t showing any signs of attempting to be relevant in China, and I feel pretty certain that as long as the current governmental structure is in place, they will not be.

      I still think there is a lot of speculation as to why Google started Android. I have my own theories that I’ll flesh out here at some point in time but I’d say they are content to simply have Android balance out competition in the region even if they get no benefit from it.

      What matters to me when evaluating that region is the software by way of developers. But the current app store and software distribution issues with Android in China is chaos. So unfortunately this situation with Android in China is worse for developers rather than better.

  1. The real question is what is the actual market share of Genuine Android (i.e. Chrome + Google search + Google Play) as opposed to forked Android (Chinese, Amazon, B&N, etc)? Because, when comparing to iOS and WinMobile, only the former really counts.

    1. Exactly right, and it is something I am working on deciphering. Only it will have to be an educated guess using the data I can drum up. But I am going to try.

  2. Here is one way that Google benefits when Chinese manufacturers fork android with their own ecosystem: developers. Developers, developers, developers. These developers hone their Android skills, share them, and move from one marketplace to another. As these developers find they can take their skills to the Play store for Western cash, every kid in India and China’s going to be spending his day building apps.

      1. Most paid app will be hacked anyway. The most likely biz model is to bundle the app with hardware or free to download.

      2. They’re not coming out of China yet much, but they will. But these Android devices sold in China -are- running apps, and not just pirated apps from other app marketplaces. This is a pretty new thing over there, but it will grow fast – especially now that a fairly reasonable tablet can be had in Shenzhen for the equivalent of $40.

        As developer costs can be very low in India and China, they can afford to put more effort and polish into their apps. Overall they may lack experience, but quite a few are already very good and the rest will catch on fast. India has more experienced software developers, since we’ve been offshoring that work to there for many years. As India developer groups publish in their own right instead of jobbing it in, we should expect an explosion of apps.

        1. “we should expect an explosion of (Android) apps…” – symbolset

          You’re missing the point of the article. Many of the apps being made for the the various iterations of Android are not being sold in Google Play and are not compatible with many Android devices.

          1. People have been decrying this “fragmentation” problem since before Android was even publicly released on a phone. Here is an an example from the almost a year _before_ the first Android phone’s commercial release in October 2008: http://web.archive.org/web/20071125210416/http://rmh.blogs.com/weblog/2007/11/is-android-frag.html

            Always the prediction is dire. It does not seem to have had a negative impact on the platform – at all. To the contrary, it seems to have helped a great deal. Remember: at that time, _not one_ Android phone had ever sold commercially, and wouldn’t yet for a year. Now half a billion have, and 1.3 million more every day (maybe a Santa peak at 2 million this year?). They’re expected to pass over a billion sold by this time next year. http://www.zdnet.com/android-cumulative-shipments-to-hit-1-billion-mark-in-2013-7000004074/ That is a conservative estimate. And this insane growth is in a market where un-fragmented iPhones had already established a commanding share that has now been surpassed 5:1.

            Yet still we have these posts “the Fragmentation is killing Android!” Pshaw. Each of the various Android ecosystems are large enough to be self sufficient. Who needs a half-billion unit sales a year to maintain a market position in electronic devices? Nobody. Half a billion Android devices is on the order of $200-400 billion dollars in device product revenues. That’s quite a lot coming from nothing at all four years ago.

            “It’s not helping Google” you say. Google’s doing fine. Android has done for Google what Google bought Android to do: Google services are not locked out of mobile. That was the goal and we can consider it won. Verizon doesn’t Bing their Android phones any more. Turns out Andy Rubin is a bit of an overachiever. It doesn’t _hurt_ Google if every last person in China and India gets to use their Android OS for free. It’s not like Google didn’t pay for its development already, and aren’t making enough off of it to support a healthy profit as well as the aforementioned “not being submarined in mobile” benefit. So where is the harm? That poor people who are poor targets for advertising don’t see the advertising and drag down the price of the advertising? Why should Google care?

          2. I agree with you, but I think it is true only from a particular perspective. Fragmentation is a problem, but just for whom it is a problem is a more relevant discussion.

            I’ve always said Android fragmentation is NOT a problem for the customer. That is just the nature of cell phones and always has been. The customer is used to it. They find what they think they want, use it for a while, discover it isn’t doing what they want, and move on. as long as the fundamental reason for having a cell phone, smart or otherwise, is met—making calls and texting—everything else can be dealt with in other means until a better purchase can be made.

            Fragmentation doesn’t matter to the device maker or carrier because, as above, this is as it always has been. Besides, having the latest OS on the newest phone becomes a bullet point of the features. And it will last as long as cell phones have always lasted with customers, who are used to replacing phones that don’t add up but they are stuck with until they can’t deal with it anymore or the contract time runs out. then a new phone and the associated emotional rush that brings. Or not.

            I am also slowly discovering Google doesn’t care about fragmentation. It doesn’t matter to Google if any of the handsets ever get the latest OS, run the latest apps, ever get used for anything other than a phone. They don’t have to care. It is not their job. They never have to deal with the end user of the phone. That is the device makers’ job and sometimes the carrier’s. Oh, sure we hear Google give lip service to solving fragmentation, but nothing ever comes out of it.

            The only people that fragmentation matters to is 3rd party developers. They are the ones who have to figure out where the money is in the fragmented and segmented Android market and that is neigh on impossible. Eric Schmidt’s predictions not-withstanding, because of this issue Android will never become the predominant platform for developers because 75% market share for Android does not mean that 100% of that market is instantly addressable to them. If they are lucky, maybe 30%. There is no single Android ecosystem.

            I won’t even get into whether Android is of any interest for e-tailers. So far there is no reason for it to be.

            Now in terms of being relevant, should any of these groups _care_ about fragmentation? Sure. If Google wants Android to remain relevant, they need to start being concerned that only Samsung is making any money. Dell and one other company (I still can’t remember who, probably a waning cell phone maker anyway) is dumping Android because a free OS doesn’t help if it can’t move product.

            They need to be concerned that companies like Amazon, B&N, and most of the Chinese market is making Google’s participation in Android almost entirely irrelevant. Samsung is even contemplating a “closed” ecosystem (if news reports are to be believed) and I wouldn’t be surprised if they eventually made good on their tinkering with their own fork if not their own OS. That would be a devastating blow to Google/Android. At least Samsung sees the writing on the wall. If the other device makers learn fast enough, then this trend will continue. However, history shows most device makers don’t really learn fast enough if at all.

            But I could be wrong.
            Joe

          3. “Dell and one other company (I still can’t remember who, probably a waning cell phone maker anyway) is dumping Android because a free OS doesn’t help if it can’t move product. ”

            Dell can’t make an Android product that sells. That’s Dell’s problem, not anybody else’s. Just look at the Nexus 4, 7, 10 – and the companies behind them. That’s some rough company to compete with and some gorgeous products. Lots of people are making money on Android devices now – not just Samsung.

            I don’t know where you were headed with that “can’t move product” crack. You do know that Android devices outsold Windows devices this year, don’t you? By a good stretch. They’re moving product like crazy. They’re saving the tech economy.

            Here is another thing that Google gets with ubiquitous Android in China: competitor products can’t leverage China sales to attack them from that side.

          4. You said it yourself, Dell can’t make an Android product that sells. Using Android is providing no advantage.

            But your right. it isn’t enough to move product, one has to make a profit doing so. Which of those Nexus products are the ones selling at cost? Who is making a profit selling Android devices other than Samsung? At some point “going” Android is going to have to prove itself to the hardware makers or they can continue to lose money.

            Joe

          5. “Google services are not locked out of mobile. ”

            More to the point of the article, in China Google services are pretty well locked out.

            Joe

    1. “Google benefits when Chinese manufacturers fork android with their own ecosystem: developers…” – symbolset

      Your analysis is exactly backwards. The more fractured Android becomes in China and elsewhere, the HARDER it is to develop for the FEWER potential customer’s each separate Android market place can address.

      Where does a developer want to develop? For iOS, which has a single store with an addressable market of 500 million iPod Touches, iPhones and iPads? Or Android, where the addressable market is splintered into pieces? Don’t bother answering. Developers vote for iOS with every single App that they initiate at the Apple App Store instead of elsewhere.

      1. A lot of developers do choose iOS first. Some choose Android first. Most successful ones choose both iOS and Android, giving an addressable installed base of about one billion customers. Being exclusive to one or the other is of no benefit to the developer.

        You see Google Play is no further from a Chinese or Indian developer than their own local store is. They are both online. And it uses the same .apk files. So if their app is successful locally, why not put it in the Play store too? And Amazon’s market and whichever other ones as well. Compared to developing a popular app, submitting to the various Android markets is a trivial effort.

        1. “Google Play is no further from a Chinese or Indian developer than their own local store is.” – symbolset

          Actually, the exact opposite is true. Google Play is not available in many markets and particularly in most of China. Further, since many of Google’s software assets are stripped from various Android iterations, those phones don’t have access to Google Play no matter where they are located.

          1. You don’t need a device that has Google Play to submit apps to Google Play. You need a US credit card with $25 on it, a US remailing address, and a browser. Using a VPN to get your Internet address into the US address space is a likely additional cost. But these are all solved problems for these folks.

            You can even lease a set of Android devices to test on remotely.

        2. Just ask BBC how difficult Android is to develop for:

          http://www.bbc.co.uk/news/technology-20754182

          “It’s not just fragmentation of the operating system – it is the sheer variety of devices. Before Ice Cream Sandwich (an early variant of the Android operating system) most Android devices lacked the ability to play high quality video. If you used the same technology as we’ve always used for iPhone, you’d get stuttering or poor image quality. So we’re having to develop a variety of approaches for Android.”

          “the number one device contacting us is still the Samsung Galaxy S2, which can’t handle advanced video.”

          Joe

          1. I’m pretty sure the BBC iPlayer team is a Windows shop, and they don’t want to invest in transcoding their vast and growing library. Netflix has no trouble running good video on an SGS, SGS II, SGS III, or any other Android device I’ve tried it on. Anyway, these programmers aren’t going to be trying to do anything nearly that difficult.

      2. why you ask people don’t bother to answer? I thought moderators are the ones that are suppose to be arguing “sensibly”

        1. Sorry, it was intended as a rhetorical device. I was trying to say that the rhetorical question I asked had already been answered by the actions of developers.

  3. Refreshing to read an article created with critical thought processes and to read someone who actually questions why Google suppresses real figures. This projects a truer vision of OS market share. This also explains Android’s feeble web presence, despite Eric Schmidt’s claim of “winning”.

  4. Hi Ben, interesting article ! Android is so fragmented that I even usually download my own ROMS to change the stocked version. My question to you would be, how many are there actually “major android players” in China? Who are they? & how many are they compared to the iOS based on percentage of market share? This might give a clearer picture on how ANDROID has impacted iOS, but not benefiting google.

    Thanks in advance

    TW

    1. I’ve been looking to try and get more of a specific breakdown but the data is hard to come by and I am not sure how much of it actually exists. So many of the devices are just blank Android with no Google services and all regional services which is difficult to track. The major players are largely the handset guys like Samsung, Huawei, ZTE, Lenovo, etc. Market share of the hardware should be easier to acquire.

      What I am mostly interested in trying to generate is the percentage of overall Android install base that does count as genuine Google Android, but that is also difficult.

  5. As a consumer in China, I find Android very disapointing. If you buy an Android phone here you get none of the google apps….. and no access to them. Even the websites that allow downloads of .apk files are blocked. I could only get google maps by asking a friend in another country to download it, zip it and email it to me.

    But yet Note IIs, and Galaxy S3s are everywhere. Very expensive phones that are cut down to the functionality of Nokia Symbian phones of 6 years ago.

    Windows phones are similar. Very limited in the apps one can download.

    In my opinion, any smartphone sold in China should come with a warning on the box: “Warning: This is not a fully functioning smartphone”.

    1. Thanks for the comment. Your feelings are in line with a great deal of the feedback I got as well from talking with experts in the region. Hopefully this situation gets better but the question is whether or not Android gets better or other platforms that are more unified take the lead.

  6. Interesting perspective, but fundamentally flawed. The fact is that the “Google Android Platform” is far more prevalent as compared to the forked Android versions from Amazon/B&N. It really seems like “fragmentation” is fundamentally misunderstood – “Skinning” does not cause fragmentation, “Forking” & upgrade cycles do. Anyone who wishes to delve deep enough can easily find out that fragmentation is cyclical.

    You’re right in that Chinese Android devices don’t have access to the Google Play Store, but they’re essentially running the same Android OS (and not forks) – meaning developers have a distribution challenge facing them & not a development challenge. The problem in China is driven by Google’s problems with the Chinese government as opposed to a problem with the platform. This is obvious when you look at India – every single local manufacturer uses Android with Google services.

    At the end of the day, Android is a single platform and a dominant one at that. That’s not to say it poses a huge risk to Apple, they each have their core market segments that they serve effectively.

    1. I think you completely are missing where I am coming and what more critical analysis needs to be done to understand Android as well as its long term viability as a platform from a strategic standpoint and from a business standpoint.

      If you read the article prior, you would have noted that I am not referring to the skins. What I am pointing out is that there is a difference between Android at large and the Android that has meaning to Google from a business standpoint. The latter point is largely important if I am in investor looking at Google. I should not be tricked into thinking the entire install base of Android has business benefit to Google. So I am seeking to understand what percentage of devices that run Android are CTS compliant by a member of the OHA who has signed the AFA. That would constitute a Google partner. And it is not 100 percent of the Android install base. A number of my colleagues from IDC, and Gartner are currently working on a more educated Android break down.

      The big truth to this is that Android is not unified and that is my point and its also an important one from a strategy and economic standpoint. The bottom line is if I am a developer I can create an APK for a good chunk of those devices but I need to be smarter about how I distribute it. Because the bottom line is I can’t put it in the Play store and think I can take advantage of the total addressable Android market, because I can’t. From a economic, and platform standpoint this is key.

      That reality also points to a platforms vulnerabilities, to which in my reports to all the handset OEMs, I point out many vulnerabilities. Interestingly none of them disagree with me. It seems to be known that Android is a weak platform with lots of long term strategic holes by everyone making Android devices.

      A house divided against itself can not stand. That is what we are seeing, let alone that Android seems to be destined to the low-end by the masses, at least at the moment, which has huge economic implications for developers, advertisers, etc., because this demo is engaging very differently and is harder to make money off from a service standpoint.

      Overall this analysis is critical if we are doing true platform analysis and more importantly ecosystem analysis. We need to understand what platforms are sticky, create loyalty, and investments in by the end consumers. My point about China is that although, its Android there, there is very little ecosystem investment in Android due to their loyalty to horizontal services and their largely chaotic app store system. This in my mind creates an opportunity for competitors. So again, the Android picture is not as rosy from an ecosystem standpoint and ultimately we need to understand unified platforms verses others but also ecosystems. To my point that Android is actually multiple ecosystems using Android as a bios.

      I’ve read much of your analysis on Android and I see where you are coming from and don’t think we disagree on that point. I, however, am looking at Android very differently because these are the kinds of things the OEMs have me analyzing so they can make smarter decisions for their 5 year roadmap.

      1. I partially agree with your point. I think it is important to split Google-certified & non-certified versions of Android. As per Informa’s report that came out recently, about 33% of current Android sales are from China, of which about 41% are non-certified. Based on Gartner’s Android market share estimate of 72%, that comes out to about 10-12% non-certified vs. 60% certified. Of course, this is only one research firm and we need a lot more data to be certain. In tablets, Amazon has about 20% of Android tablet sales (B&N is in low single digits), but that seems to be dropping as broad based OEM growth takes over – This puts Amazon at 10% of the market vs. ~35-40% for other Android tablets. These figures are why I fundamentally disagree with the multiple platforms theory.

        Regarding the demographic angle, I don’t completely agree. For one, Apple and Samsung currently hold about 47% and 40% of the $400+ smartphone market. So the high-end smartphone market isn’t as one-sided as it is thought of (and behaves fundamentally differently in subsidized and unsubsidized markets or even those with lower subsidies, such as mainland Europe). So there are many factors that need to be taken into consideration before assuming a uniform low end demographic.

        And finally, there’s engagement. There’s no question that engagement is higher on iOS, and that is partially driven by average demographics. The issue is that most engagement statistics only take browser usage into consideration, which is skewed by the usability of the stock browser prior to Android 4.0. NetApplications data clearly shows sharply rising utilization on Android 4.0 devices.

        Of course, these are individual arguments. We essentially disagree on our core viewpoints, i.e. Android as a unified platform with smaller forked branches vs. Android as a BIOS. I guess we will only find out with time. In the meanwhile, it is always interesting to read multiple viewpoints, as it helps fine tune my research as well.

  7. Do you think Google is taking any measures to increase Google Play usage in China? Anything at all from increasing local user support or changing any of their engineering design? If not, why are they not making these efforts?

    1. It’s tough because the Chinese government is very anti-Google. So it seems that it will be very difficult to Google to truly make in-roads with their specific services into China, until they work things out with the government, which they seem to not be working hard at doing. Google essentially keeps closing offices in China, so rather than invest there, they seem to be retreating.

Leave a Reply

Your email address will not be published. Required fields are marked *