AMP Index update

The following graphs show the most visible global phone brands and the approximate percent of value they captured since 2007.

Screen Shot 2013-08-21 at 8-21-6.17.19 PM

The graphs each show trailing four quarter average of shares of units shipped, revenues, operating profits and smartphones shipped. I also averaged the four shares into a single share number called the AMP index (Asymco Mobile Performance).

Note that this data ignores those “other” vendors whose financials are not available.

I placed all AMP indexes on the same graph and looked for some patterns. The result is shown below.

Screen Shot 2013-08-21 at 8-21-6.58.01 PM

It’s not much to go on, but there seems to a pattern where each vendor reaches a peak share, though that peak varies greatly in value. If the pattern holds, we could perhaps anticipate a peak for Samsung in one year or so.

I’m searching for some hypothesis of what is happening and what is causing this pattern, if it is a pattern. The instability of the market is certainly a pattern. (Many markets would show horizontal lines for many years rather than what we see here.)

But the constant turnover and the periodicity of success and failure is a characteristic of either a market where there is a hidden hand (operators? regulation? etc.) or one where disruptive change is extremely consistent and rapid.

  • Anon

    Why is Samsung included on the 1st set of graphs but not the 2nd (AMP)?

    • Shaun Wilson

      It is there, it’s the yellow line…

    • Shaun Wilson

      It’s the yellow line…

  • obarthelemy

    I don’t think operators nor regulation have much of an impact. As far as I know, regulation is both light and fair; it probably barely impacts total sales, let alone the repartition of sales between vendors. Operators are retailers, and as much as OEMs would like to think retailers are “partners”, retailers are wh*res selling whatever the customer wants.

    I think the patterns we’re seeing combine 3 main elements:
    1- ecosystems war: Being outside of the iOS+Android duopoly is clearly not a winning proposition.
    2- pricing transition: Smartphones used to be fairly price-insensitive, because they targeted affluent markets and were extensively subsidized. Both aren’t so true nowadays. Also, the market is maturing and 2nd and 3rd tier hardware that used to be unusable is now OK to good, increasing pricing pressure on top brands.
    3- features uncertainty: it seems obvious that OEMs still don’t know which features sell, and which don’t. Huge phones have been mocked since day 1, but were a surprise hit. Screen resolution has reached ridiculous heights yet still seems to be a strong selling point, whereas sound quality and power is very limiting, yet still overlooked. Ditto for camera, SD slot, casing materials, battery size and “removeability”, FM… The fact that everyone is getting components from the same handful of suppliers doesn’t help breakthrough features, either.

    On top of that, branding is in flux. All major players bar Apple seems to want to duke it out on the features and feeling front, there’s no hipsters vs hood vs bourgeois vs rich vs family segments like there often are in clothing/cars/food…

    • davel

      The fact that Apple has no formal relationship with China Mobile has nothing to do with the volume of phones they can sell..

      So yes, operators have no impact.

      • Tatil_S


  • Pedro Andrade

    Horace, would it be possible to extend this study backwards? I think it would be extremely useful to your analysis. As you often say, what happened in 2007 with the iPhone launch was a dramatic blossoming of “jobs to be done” by a mobile phone and with it a shift of the relevant skills required to provide them. Many of the titans of the past succumbed in this new environment, which is exactly the time you show. But how did things look before? Was it always like this, or have there been periods of AMP stability in the past?

    Cheers and keep up the excellent work.

    • It would be a great idea to extent back. It’s a lot of work but if it can be organized as a crowd-sourced project then it would help me a lot.

      • Pedro Andrade

        I would be glad to help. Maybe we should do it for yearly numbers instead of quarterly, at least at first. It also appears important to clearly set out the methods to reach the current data, so we can extend backwards following the same criteria. How would you organize the “crowd”, Google Docs?

      • tmay

        This site is very interesting to myself as somewhat more adept at manufacturing, but looking at the charts, would it be possible to integrate profits of all the players to predict the rate of change for the market and the mature market threshold?

  • johninseoul

    The AMP Index chart is fascinating – was there a post on how it is calculated?

  • handleym

    An obvious suggestion as to the pattern is that companies (or more precisely the top management at companies) become fixated on their way of doing things, and stay with it even as the world changes — skating to where the puck is today.

    So Nokia kids itself that dumb phones and Symbian will remain profit centers even after the iPhone, Blackberry that “professionals” will still demand a keyboard, MS that people desperately want to replicate the Windows desktop on their phones, and so on (this latter NEVER hit any sort of peak, which should have told MS something even before Win8).

    One could certainly make up a non-ludicrous story that Apple is prey to this same hubris. I personally don’t think so; I think they’re enough aware of the next big thing (wearable computing devices, and the personal cloud/network that ties them together) that they’ll transition smoothly. They’ll probably lag a year or so behind the leaders (that’s pretty much always their pattern) and, again following pattern, the reason will be that they want to do the job right, while the internet ship of fools will insist that it’s because they have no vision, didn’t see the future coming, and that it would be all be different if only Jobs were still alive.

    Samsung is the wildcard in all this. Judging from their past behavior, they will embrace wearable computing with the enthusiasm of a crack addict visiting Columbia, and the skill of a dyslexic new-born monkey. Expect an endless variety of poorly thought out gimmicky gadgets, none of which work well together, all of which lose any sort of vendor support three months after they’re released.
    Will the world go for this and allow them to actually make money? That’s the part I don’t know. My guess is that the developed world has enough experience with computers (and the hassles of incompatibility vs the payoffs of smooth interop) that this will have little traction; but it may be able to separate a number of Indians. Chinese, Peruvians, etc from their cash before they learn the downsides personally.

    • obarthelemy

      Speaking of monkeys, all reviewers give top marks to several Samsung devices, so maybe your vision of Samsung is very off the mark ? Also, the developed world also has experience with single-vendor, locked-in, non-interoperable proprietary ecosystems, so you’re probably off there, too.

      • Dan

        What if Samsung’s watch or other wearable computing devices only work with Samsung phones, or are even more limited by requiring S-Apps (not open to competing Android apps)?

        The current rumors are that the Samsung watches require Samsung phones.

        Wouldn’t this be single vendor lock-in?

        Samsung is becoming more like Apple in some (not all) regards.

        Lets be honest:

        Google wants Android to be open and dominate market share so they can sell more and better targeted ads.

        Samsung wants to sell more devices, phones, watches, HDTVs etc…

        Samsung can sell more devices by being “half-open” than “all-open”.

        If Samsung owned Android, it wouldn’t be open at all, it would be exclusive to Samsung.

      • obarthelemy

        Indeed, if Samsung watches only work with Samsung phones, that would be lock-in.
        As to “half-open” selling more than “all-open”, If the watch ends up being a phone accessory, I’m not sure how limiting it to Samsung phones increases sales ?
        As for a “Samdroid” being *obviously* proprietary, Tizen isn’t proprietary. Bada was though.

      • davel

        I would think they would go this way. Currently Samsung is Android. So they can do this.

        At what point does Google stop paying any attention to Android and just focus on Chrome?

      • peter

        Regarding the monkey thing, the difference is that Apple uses some form of ‘inner compass’ in designing products, whereas Samsung iterates its way to better products using external sales (and what seems to work for others) as its compass. Different approaches.
        It does mean that Samsung 1.0 products may seem like amateur attempts, whereas in reality they are more a form of public beta testing (although the paying customer may not know that, but such are the risks an early adopter takes).

      • obarthelemy

        That’s mainly a variant of FUD:
        1- Samsung v1.0 products are mostly excellent (Galaxy S, Galaxy Note, Galaxy Tab 7, Galaxy Tab 10).
        2- There’s attempts at trying to build the myth that because Apple release few products, they’re better from the get-go. that flies in the face of Antennagate, native apps, …

        The way I see it, Apple decide what is good for us, Samsung lets us choose (and Android even more so). Trying to rationalize “our products are more perfecter because there’s no choice” doesn’t chime with reality.

      • peter

        By choosing one Samsung phone over another, you are providing a valuable service to them as they will iterate the more popular models/features. For example, if customers ‘vote’ in favor of 5″ models then they will try a 6″ model. If that does not sell then they will iterate in a different direction. To me (and presumably Apple) a 6″ phone looks silly, but who knows maybe there is a usage model that I do not understand and it will sell like crazy.

        I think that an iterative approach to hardware can work quite well, provided that you can do it fast and abandon your failures quickly. Where I think the approach is more troublesome is in software (trying different OSs, skinning and trying different build in applications). There is a real cost from the resulting complexity, stranded past customers who cannot upgrade might not be so happy and it is more difficult to spot the winning features in a software stack.

      • obarthelemy

        Again, choice:
        1- If OS updates are a major need, get a Nexus or a Google Edition phone (HTC One and GS4 currently, so Samsung is in there).
        2- For non-nexus/GEd phones, expect 0 or 1 major updates.

        Note that OS updates are a bit overblown
        – I still have to run into an app that won’t run on Android 2.3+ (also, Google supply lazy devs with libraries to backport the latest API levels to older ones)
        – Google have been quite good at taking features out of the core OS and moving them to easily-updated apps or services.
        – 3rd-party tools (launchers, notifications, lockscreens, keyboards…) can be easily downloaded from the PlayStore to tweak and upgrade a lot of stuff.

        We’re comparing Apples and oranges here (pun intended): iOS locks down many things, so OS updates are vital to keep the phone vaguely up to date. Android on the contrary is modular: core apps update w/o OS updates, users can change most anything via regular apps from the PlayStore.

        The only caveats are OEM-supplied software (as if anyone were really using those, I can’t think of anything that hasn’t a better PlayStore equivalent), and hardware-specific apps (pen-enabled apps are few besides Samsung’s,). The biggy is Samsung’s certified security infrastructure KNOX (supported only on the GS4, and not a mainstream product).

        The latest major OS-dependent features were performance tweaks: project butter in 4.1 (8/12), and TRIM support last month (4.3). There seems to be stuff brewing on the security front.

      • perfection

        Must be sad for all the Android developers if Google is adding zero OS features and APIs with their OS updates, while they watch hundreds of new and useful APIs added to every new iOS release. Though, perhaps, from your view, the Android 2.3 API was perfect.

      • Kenton Douglas

        Very well said. When a lot of people talk about Android ‘fragmentation’ it just underscores the fact that they have very little understanding of the architecture.

      • Walt French

        @obarthelemy:disqus wrote, “Note that OS updates are a bit overblown- I still have to run into an app that won’t run on Android 2.3+”

        Looks like Gingerbread (2.3) was released late in 2010, so nearly 3 years ago now. As you note, there’ve been quite a few performance enhancements and features since then, but developers would have to do quite a bit of work (not sure why you call libraries to back-port more recent features for “lazy” developers).

        This is almost no concern for Google, who want to ensure lowest-common-denominator access is at least good, nor is it hard for developers unless their functionality really WANTS some more recent feature that’s maybe a bit tricky to back-port.

        What it does is to shorten the support/upgrade cycle for users. In a fast-changing industry, that’s not all bad, but it *IS* a bit of planned obsolescence, one of the less attractive features of Americans’ way of doing business—not something to celebrate.

      • obarthelemy

        I think the problem is way overblown, and going to solve itself anyway:
        1- as I said, ever fewer stuff is linked to the OS (for Android anyway), so the penalty for running an older version is ever lower.
        2- the pace of updates is slackening: we’ve had 4 major Android versions in 2009, 2 in 2010, 2 in 2011, 1 in 2012, certainly 1 in 2013….
        2b- I’d argue that the *importance* of updates is also lowering, though I’d love to be proved wrong.
        3- The hardware is still evolving rapidly. If you’re making do with 3 year old hardware, a 3 year old OS is not your most pressing concern, especially when it still runs, though less smoothly, everything. Your hardware is your most pressing concern. OTOH, my brother in law is still happy with my old HD2… (How on earth could HTC get it so wrong… they’d have been fine just iterating the HD2 design like Sam is doing with the Ses… FM radio notwithstanding…)

        Side note: OS updates have historically been free so people want their free lunch, but most would be hard-pressed to articulate *why* they want it (what jobs to be done they need it for). I’ve had gizmos update on me (usually once per gizmo: Xoom, Note, Note 10.1, plus a Nook Color I play with CyanogenMod on). None of those updates ever enabled any new use, and I mostly didn’t notice any changes at all.

        I understand the iOS situation is *very*different, with keyboard, notifications, icons, base OS apps,… baked into the OS. Again, not so in Android. Even the latest gimmick (remote wipe and geotracking, I’ve been using an App for that and more for years) is just an update to some daemon, and running on all versions starting with 2.3.

        And finally, Apple and Android count differently: Apple release a new OS version “for all devices”, except not really: they cut features on older hardware. Google *add* some features to older OS versions. Less flashy, but very nice too.

        Edit in conclusion, I think the Android version thing is mainly an Apple PR gimmick. Not important IRL, and not clear cut at all.

      • Kizedek

        “I think the Android version thing is mainly an Apple PR gimmick. Not important IRL, and not clear cut at all.”

        That’s right, it’s not clear cut at all, as you tell it. You are doing a lot of obfuscation.

        What is clear is that 3 yr-old hardware doesn’t have the capability of new hardware. Technologies improve beyond mere raw power: chips improve, RAM, power management, GPUs, etc. …at least on the Apple side.

        Now, whatever you want to call it — PR, FUD, different strategy, whatever — it is clear that 3 yr-old hardware should differ from new hardware in some fundamental ways. I don’t know why you are playing obtuse and pretending to be the only person on the planet who doesn’t see this.

        Moreover, what is clear is that the majority of Android devices are, and continue to ship with, an older version of Android.

        Finally, what is clear is that Google has to approach the “issues”, whatever these may be (and you obviously think they are minimal to non-existent), in a different way. It is putting some features into apps vs. the core OS, apparently. Why? *Because* the lack of control over OEMs and carriers has led to a situation where hardly any user can count on getting any update, at least not easily and quickly. Because Android OS Fragmentation is a fact.

        Therefore, some “features”, whatever they may be, and where ever they may be, be they in the core OS or siphoned off into apps (who really cares?), are more usable, or only usable, on newer hardware.

        Certainly, some new features should be expected that cannot “run” well, or at all, on 3 yr-old hardware. Apparently, it is a terrible thing that Apple develops new features like this, and makes it easy for all iOS devices of 3 yr-old devices and newer to enjoy at least SOME of the new features and improvements (of which there are many, for all devices, no matter how old).

        Now, iOS developers can take advantage of new features, improvements and APIs, AND they can make their apps “backward compatible”, ie. “compatible with” or “downloadable and installable” for older devices or versions of iOS. They just have to check the boxes. However, as noted, certain features may or may not be available to certain devices. That’s understood (again, by everyone except you, apparently).

        However, IF a developer’s app is based all around new features and APIs as its USP, then they will likely raise the installation requirements, and users of older devices or versions of iOS may not be able to download it at all (the developer will purposely not make it “backward compatible”).

        Surely, the situation is very similar on Android. Your obfuscation has failed to convince us that it is any different — certainly not better in any way: an Android app developer, including Google as it siphons off certain core OS features, has to make the app “backward compatible” for older hardware or versions of Android going back 3 or so years to, say, 2.3. In fact, he has much more of an obligation to do so than an iOS developer who can count on the fact that 90%+ of iOS users are running the latest version of iOS.

        Now, clearly, such an app (one written quite recently for use on 3yr-old hardware and/or versions of Android) is not going to run “just as well” on every device and every version of Android; nor will every new feature be “just as usable” (if available at) to every Android device running every version of Android back to 2.3?

        At least one would hope not. If this is just a made-up issue with no practical relevance whatsoever, if it is just a “PR gimmick” or FUD, then, as Perfection commented, it shows there has been no real improvement in the Android OS or APIs for 3 years, other than to add cosmetic improvements and things that simply require only raw power to run more smoothly on later hardware and/or versions of the OS. Take your pick.

      • davel

        Galaxy S was not a 1.0 product. Look at the docs in the Apple vs Samsung case that netted $1B for copying. Their current phone lineup is a direct result of Samsung deconstructing Apple phones ( down to placement of icons ) before they blew up. If Samsung comes out with wearable products as is rumored I will be very interested if they learned anything or go back to their usual ways and just spam every possible combination of features. The phablets seem to make money for them, but how many iterations did they spew out before they found one that the consumer liked? You can make the case this is innovation, but there is no thought to this or at least considered thought.

      • obarthelemy

        The famous $1B is fairly random, several of the involved patents have been invalidated since… we’ll talk about this when the process has run its course, maybe ? Also, locality seems to play a large role in US court decision, reverse outcomes have happened in other countries.

        Phablets were successful from the first one (“Galaxy Note”). As were tablets.

        No clue about watches, I’m very doubtful about the whole market but I’m getting old and blasé. I’ll see what the nephews and nieces are buying ;-p WIth so many smart watches already out, and fo so long, anything that comes out now is bound to mainly a re-hash of what’s already out there. I’d love to be wrong though.

      • attention

        “Also, locality seems to play a large role in US court decision, reverse outcomes have happened in other countries.”

        Common misconception, but you should be paying more attention. e.g.

  • Sacto_Joe

    I would only argue with the last few line segments for Apple. One can see in the data points very clearly the seasonality of the iPhone sales over the last two years, that is, ever since Apple shifted the iPhone release date three months closer to the Winter quarter. lt is obvious that this pattern is going to repeat, although it is not know whether or not growth will continue. Thus, the last line segment should be running roughly parallel to the x axis, albeit perhaps with a dashed line to indicate a projection.

  • Chaka10

    The components of the AMP index are all percentage-share based (profits, unit sales, revenues…). Given its still strong, some say “obscene”, GMs and brand attraction, Apple could reduce its iPhone prices significantly, literally overnight (and many argue it should, as soon as 9/10). Its percentage shares and AMP Index would shoot up and those of Samsung would dive down, giving a very different picture, for Apple and for Samsung. IOW, Apple has control over the trajectory of its AMP Index, in a way perhaps different from NOK, HTC, RIMM …. Apple obviously knows that, and has the luxury of making the decision whether to go for share or gross dollar profits. I understand that the foregoing may be viewed as providing a strong, obvious conclusion Apple should quickly go for more share even if it means less profit. I remain unconvinced of that — my main issue remains one of market definition, which is a key underlying factor for the share numbers. That however, is the key strategic management decision for Apple.

    • obarthelemy

      Other OEMs, especially Samsung, are in the exact same position for their high-end products: they get wonderful margins and have a lot of room to lower prices.
      All of them prefer to leave the high end with high margins, and try and work out a midrange proposition. For obvious reasons.

      • Chaka10

        “Other OEMs, especially Samsung, are in the exact same position for their high-end products”

        Well, perhaps we’ll see Apple put that to the test….

      • obarthelemy

        Actually, they’ve already put the test on Apple :-p

      • Chaka10

        I really, really didn’t want to get in the same tired debate with you, but you throw the same old global market share chart up …. No, OEMs are not “in the exact same positions for their high-end products”.

      • Math

        Samsung (and Apple) “sell” $600 phones for $200, because they make the rest from the network (though Apple may well get slightly more). An extra $200 dollar discount for people taking two contracts is meaningless.

        Find me a store selling 2 S4’s for the price of one out of contract and we’ll make a killing on eBay.

  • Walt French

    Horace wrote, “I’m searching for some hypothesis of what is happening and what is causing this pattern, if it is a pattern.”

    Methinks that you’re reading a bit too much into the noise in the case of Apple, and too much into brand generally. Brand matters a LOT, but that strategy & position matter more.

    In creating the touchscreen smartphone market, Apple was a clear winner from its strategic leadership. It took a while for the market to coalesce around apps, and for Apple to ramp up its production, so even a badly disrupted firm like RIM kept gaining until 2010.

    Apple thought it had a 5 year window before competitors had touchscreen smartphones of comparable quality, but was obviously blindsided by Android, and especially the company most adept in pivoting from its feature-phone past, Samsung. The marriage of a modern OS plus Samsung’s global smartphone presence and the supporting technologies obviously was going to succeed.

    Why didn’t LG and HTC succeed in the same way as Samsung? I presume some combo of bad luck and not having either the retail presence of Sammy nor the financial wherewithal to jump in with both feet.

    And you must understand Nokia better than almost anybody, but your chart makes it look like it was fiddling around with high-end phones but simply wasn’t quick enough, while Samsung came in and ate its low-end business.

    I think the Apple story will be much clearer once we see the results of the 5C — in perhaps a year or two. That device should be Apple’s “VolksPhone,” a device that competes well for middle-class Europeans or wealthier people in the developing nations.

    I’m hoping that Cook takes Jobs’s 1997 advice to Apple: take everything of value you can from the leading product and repurpose it into the next great thing. Their crown jewels today, of course, are not the Mac OS, but the iPhone, iOS and iTunes; with their design language in the mix.

    Apple has already repurposed iOS into tablets and the AppleTV, and seems to have a vision for a wearable next device / phone peripheral. One already has, and two that could well be, disruptors. Myself, I’m hoping that Apple is willing to risk cannibalizing its high-end line (while continuing to improve it), by repurposing their resources into a mass-market, hundreds-of-millions-per-year device.

    This has implications for Apple’s future course. While I expect the new Mac Pro to be a very limited-sale device, I expect it to inspire a new TV and/or game platform at the high end. (Why else would they be advertising such a niche device at theaters?) In phones, I expect innovations to be of the sort that pay out in small amounts over hundreds of millions of devices; amortizing an on-phone Siri, or buildout of Apple.Com across the globe’s networks for local presence, rather than simply more DPI or GHz and fancy sensors on the flagship phone; the flagship will be the test bed, as it has been with the Siri beta.

    • peter

      Looking at Horace’s charts, all of the companies presented are quite capable builders of hardware (and increasing complexity of hardware is not all that disruptive anyway).

      I think where the various companies got tripped up were in (1) operating systems (which are difficult to develop and turn into a platform – Nokia), (2) insufficient economies of scale (which limits buying power, design budgets and margins – LG, Sony, HTC and Motorola) and (3) addiction to high margins (the inability to sell to lower margin customers – RIM).

      The risk for Apple is in (3), if the iPhone 5C is priced sufficiently low then this risk might not materialize for them. The risk for Samsung is in (1), I’m not sure what they can do to encourage Google’s continued commitment to Android.

      • Kenton Douglas

        ” I’m not sure what they can do to encourage Google’s continued commitment to Android”. I don’t see their commitment altering to Andriod as a platform. Their focus always was (and will continue to be) web services – nothing has changed.

      • peter

        Google wants to maximize internet traffic to their services and generated advertising revenue, they need access to the iOS market but do not need to compete with Apple. Samsung, however, does compete with Apple and needs Android to stay competitive with iOS 7/8/… in the phone and tablet space.

        A cynical Google might be able to maximize profit by letting Android trail iOS in terms of features because the required processing power will arrive 2 years later in low-end devices (which are the bulk of the Android market). In addition, not being cutting edge and copying-what-works is a far simpler strategy that can still prevent outright dominance by Apple in the mobile space.

        For Samsung being dependent on Google’s continued efforts and good will is a risk.

      • Kenton Douglas

        Yes, access to iOS is very useful, but it’s not a priority ahead of Android. Also, they’re going to have to compete with Apple in advertising (think iAd, iTunes Radio, which is advertising backed).
        Android is already feature competitive with iOS, and ahead in certain areas such as scalability/flexibility due to it’s modular (Linux-based) design. You need to identify these “features” where Android trails iOS. Being cutting edge is what Google are all about recently. I don’t think they look to Apple for any market guidance. Whether the initiatives they’re pursuing (Project Glass, Driverless Cars, Project Loon, etc) are ultimately successful is a different question.

        If anything Samsung needs to worry because a ‘cynical’ Google will lock down the services and development environment on Android (already on the way via Google Play Services) and push their own devices (Nexus, Motorola, Google Glass) more.

        I don’t think they’re going to let their investment in Android (which supports around 1B users already) flounder because the perception is they’re going to make more money from prioritising a third-party OS to which they have no guaranteed access.

      • peter

        What I’m saying is that there is not much benefit for Google in being ahead of of iOS because it takes forever for the latest Android version to show up in the installed base. The vanity projects you describe are good marketing, but not much of a business. The real challenge for Google is to find enough ‘white space’ on mobile devices that can be sold as ‘advertising space’ (judging by their mobile ad rates, this is not something they have yet cracked).

        The cynical Google you describe must be a constant worry over at Samsung; from their point of view being locked out by Google is as bad as being locked out by Apple. I would expect Samsung to try to get more control over their own destiny by trying other operating systems (e.g. Tizen), perhaps forking Android the way Amazon did, or entering into a deal directly with Google (who might need a bit of co-funding for the ongoing development of Android) — anything really to make sure they are masters of their own destiny.

        From what I can see the investment in Motorola ($12B, maybe $6B after cash and asset sales) is currently floundering. Who knows what else might flounder as Google tries to find out ways to monetize mobile devices.

      • Kenton Douglas

        OK, I see where you’re coming from. Although I think they’re trying to change the delivery of services (and core apps) on Android by separating those from the core OS – which is why the latest update (v4.3) was actually small (feature-wise), since the core features and apps are now updated independently. There’s also a possiblity they’ll also slow the release cadence to maybe one version update a year.
        So the base updates by carriers/OEMs will – I think – become less time critical.

        In terms of revenue, I think they’ll have to increasingly look beyond advertising, since that market will become more commoditised like any other. It’s their biggest challenge to date. I read the nett purchase on Motorola is nearer $2B ($1.5-1.7B) once you factor in the the tax assets/credits acquired (from a loss making company), i.e. about what they paid for YouTube. Ultimately though, like all the other major players in Tech, they’ll probably end up with a reasonably diverse revenue mix. Primarliy ads yes, but some income from hardware, some from Play Store services and apps, some from the Cloud Platform (IaaS/PaaS), some from Apps for Business, etc, etc. The key will be the mix. I think moderate success is certainly possible from the ‘vanity projects’ also:

        For Samsung the problem isn’t the underlying OS or even the apps (which can be readily ported). It’s the web services and the global cloud infrastructure needed to support them. That takes time and expertise to get right. I suspect their best option (Plan B) would be to fork Android and get mapping, search, etc from Microsoft. As mentioned porting the apps would be trivial. Tizen doesn’t make any sense to me if you can start from (Open Source) Android as a base – and have access to an established development community. So overall, I don’t think Samsung are in a bad situation. However, in the very long term web services will be the key in my opinion. Therefore, they might as well start that long journey while they have a Plan A and B available. They have a developer conference next month. I think we’ll see a move to emphasise a services layer that binds Samsung devices (phones, smart TVs, etc) on top of Android (or any given OS). Should be interesting for a sign of their chosen direction.

      • peter

        I checked Google’s 2012 10K. Right now about 85% of their revenue is advertising, most of the rest is Motorola. Google’s interests as an advertising business are not necessarily aligned with the needs of Samsung as a hardware business, while as a mobile phone vendor Google is an outright competitor of Samsung.

        It will be interesting to see what will happen next, your comments give some food for thought. To add another example, if Apple were to introduce a TV product, would Google quickly produce the type of Android extensions (or Apps) that Samsung would want to use in their TV sets?

      • Kenton Douglas

        Yes, there should be some fun times ahead with the realignment of some existing relationships 🙂

      • obarthelemy

        Low- and mid- range devices are the bulk of the market in general, not just of the Android market: Nokia are having good success with their low-end line; both under WP and Symbian.

        Android covers the gamut from very low end ($75) to higher end than the iPhone. That’s mainly an OEM/hardware issue: Andorid, like most other OSes (windows, linux…) is fairly hardware agnostic.

    • Chaka10

      @Walt French. Off topic, but I wanted to thank you for your commentary a couple of weeks ago on the purported statistical “proof”” on the Tech-Thoughts blog piece, “Proof Of The IPhone’s Dependence On Carrier Subsidies: A Market Share Analysis”.

      As you pointed out, the author’s attempt to calculate the “purchasing power of consumers in a given country” in ref to GDP is highly problematic. If I may paraphrase your point, the calculation is not done on a per capita basis and in any case doesn’t reflect wealth/income distribution demographics. I would add, if I may, that author’s “PP Index” (dividing nominal GDP by PPP-adjusted GDP) simply yields the PPP adjustment factor, which is an adjustment for purchasing power differences of CURRENCIES, NOT CONSUMERS, in different countries.

      [@Obarthalemy’s positng of a market share chart from that blog reminded me to post this comment, and I impose on Horace’s blog to do so because I don’t have a Google account.]

  • mieswall

    Weight: Reissuing this metric seems to validate the unweighted nature of it. It makes me wonder if really a share in global phones, for example, is of any importance, much less of the SAME importance as a share in smartphones (same as a share in global vehicles, including trucks, really don’t mind to BMW, but instead in high performance cars).

    Also a discussion would be useful about the equal weight in profits vs revenues, after all, the purpose is to be sustainable. And finally the never ending comparison between market share (number of phones) and profits.

    Taking in account this belief of properly “weighted” averages (the weighting factors would be a nice discussion for reach this finely tuned “sound pressure machine- spl”), in my view the situation of Apple is brighter of what AMP shows now.

    Amazing posts this week, Horace. Always a pleasure to read them.

  • sbono13

    Horace, is it possible you have revenue share (green line) and smartphone share (yellow line) mislabeled in the top figure? I’ve been trying to figure out how you calculated a 30% smartphone unit share in the most recent quarter, which would imply total smartphone shipments of 104m, but of course, even weirder is the notion that Apple would have higher unit share than revenue share, given what we know to be the highest ASP in the industry (excluding Vertu, maybe).