Categories

The AMP Index for Q3

The Asymco Mobile Performance (AMP) index is an unweighted average of:

  1. Share of all handset units sold (global)
  2. Share of smartphones
  3. Share of value (revenues)
  4. Share of profits

For major phone vendors. The raw data for each share is shown in the following charts (note change of vertical scale: each gridline represents 10%).

Note that the vendors are arranged in a particular way: top row are entrants, bottom row are incumbents with late incumbents from Korea arranged in the middle.

Taking the average of each of the lines above, the unweighted share index (Units, Smartphones, Value and Profit) is shown below:

And here it is in sparklines:

Like many indices, this is a generalization that measures a section of reality. The value is in its use as a benchmark and trend analysis.

  • http://www.informationworkshop.org Mark Hernandez

    I’m glad to see you’re still maintaining your “composite market share index.” I’ve left tons of comments on blogs and news sites when they cite only one market share out of context and I’m starting to finally see other people do the same, and I’m also noticing that blogs and news writers are picking up on it too.

    Our world is increasingly complex, and oversimplification is the problem we must constantly fight, as it’s the enemy of clarity. At best we’re always in search of the optimal balance between being concise and being informative. I’m always delighted to read your one-liners that both encapsulate so much, and intrigue us to learn more.

    In a way even the AMP index is an oversimplification, but part of its payload is the implicit understanding that it represents and array of market shares.

    Thanks Horace for your tireless efforts in upgrading everyone’s understanding of the marketplace, and allowing us all to come along for the ride as you continually find new ways to sort it all out.

    • http://twitter.com/bennomatic bennomatic

      A nice change to AMP might be a dynamic feature allowing the viewer to provide weight to each of the share metrics. My personal weight might be:

      Share of all handsets: 40%
      Share of smartphones: 10%
      Share of revenues: 10%
      Share of profits: 40%

      That might not make a significant difference in the general direction of each company’s curve, but it certainly could affect the amplitude.

      Hm. Maybe I’ll play with that this weekend and post my version of the AMP charts.

  • http://joeharris76.blogspot.com joeharris76

    Horace,

    I’d like to see the small multiples graphs each drawn to the same scale. You could then make the Apple and Nokia squares double height to illustrate the amazing growth and decline of each.

    Joe

  • davel

    Do you draw any conclusions from Samsung drawing even with Apple on these charts?

    • http://twitter.com/bennomatic bennomatic

      I think the answer to this question may be related to the answer to mine, below. I would wager that these numbers are units shipped into the channel rather than sold to consumers. For Apple, shipped = sold. For Samsung, they magically shipped a huge volume of units last quarter, but they’ve provided no data on sell-through. Because these are “share” indicators, any increase in share that Samsung got from those metrics will of course decrease the shares of everyone else in the pool.

      Of course, Apple also indicated some slow-down in iPhone 4 sales which they had blamed on the rumor mill. It’ll be interesting to see how these numbers change this quarter. With Apple selling every 4s they can make, it’ll be up to Samsung to actually sell the units they shipped last quarter or the channel won’t have room for new units.

    • http://www.asymco.com Horace Dediu

      Samsung had a great quarter and Apple was transitioning to a new model. We have to wait to see the results for the next quarter to see if this is sustained.

    • Anonymous

      Apple seem to have a problem finding their second wind. A single device is only going to go so far, while Samsung does address all market segments (keyboard, very small, huge screen, low-end and very low-end…).
      Plus I’d argue Samsung is grabbing a clear advantage in screen technology, with AMOLED pulling up to IPS in pixel density, while offering much better contrast. There’s no other hardware disruption going on (bouncy vs shattery is not really a disruption, people seem to not care, surprisingly, and anyway it would count against AAPL too), so Apple is left with having to use software to make up for the hardware issues. It’s a harder sell: Siri for example only works well in English, and in the US for the search engine. The US *is* a huge market, but the WOW ! effect of AMOLED works everywhere.

      • Anonymous

        Apple now has the top 3 selling phone handsets in America. they are slowly but steadily pushing down their cheapest phone into the lower priced/prepay market segments.

        We will have to wait until after the next quarter results to be sure, but I’m quite confident in predicting that apple this quarter will be the leader in production capacity of smartphones thanks to 5 years of hard work. Production capacity for 35 million+ smartphones a quarter is not something a rival company can just buy, it’s something that has to be built over a long period – and of course you have to have a matching steady growth in demand to match it.

      • http://twitter.com/bennomatic bennomatic

        I agree that there may be a middle path between one mobile phone offering and 150, but I think the market segmentation which it appears that you are suggesting is an advantage for Samsung definitely has its downsides. I think it was a really good tactic when phones were just phones, and the only differentiators were size and style, given reasonable performance, battery life and durability. But scores of offerings mean lower margins for the vendor, and confusion for the consumer.

        My bet is that, as soon as smart phones hit some magic number (maybe 51%, maybe 60%) of the phone market, we’ll see Samsung’s offerings drop to just a few in order to achieve profits more like Apple’s.

        With regards to display technology, I agree that great displays do make a difference. However, I’d suggest that once you get past a certain point, there are diminishing returns. The iPhone 4/4s displays are quite beautiful, and while there may be some WOW factor in some competing displays, we may be talking about the Pepsi Challenge here. Just because something is brighter in a certain kind of light or has more saturated colors does not mean that it’s better, and in the long run, the display is just another component, as long as it’s good enough. This year’s WOW is next year’s HO-HUM.

        With regards to Siri not working in other languages, is that a fact? I know I’ve seen issues where people with heavy non-English accents can’t talk to Siri in English, but I had thought that at least a handful of other major languages were indeed supported.

        That being said, Siri has already gone through rapid improvements. The fact that much of the processing is cloud-based is a big deal, because the improvements can take place without any updates to the phone, and it seems like Siri can do new things and give better responses each and every day. I’d assume that language support will follow a similar trajectory, but of course, we’ll have to wait and see.

      • GeorgeS

        “Siri for example only works well in English, and in the US for the search engine.”

        You do understand what “beta” means, don’t you? Doing Siri in other languages, especially non-Indo-European languages*, will require a lot of effort, not just in programming, but basic research.

        *There’s nothing particularly “easier” with Indo-European languages, but their basic grammars are similar and quite different from languages like Chinese, Japanese, Arabic, etc.

      • davel

        Regardless of the labeling, after a short period of time it should work. Beta or not. Else why tout it and release at all?

        Apple has a reputation to protect. They will be given the benefit of the doubt for a time, but they need to work out the issues.

        They also need to make a Mandarin version. That will drive sales in China through the roof.

      • Anonymous

        “There’s no other hardware disruption going on (bouncy vs shattery is not really a disruption, people seem to not care, surprisingly, and anyway it would count against AAPL too), so Apple is left with having to use software to make up for the hardware issues. ”

        So your argument is, essentially, that people buy phones for the HARDWARE, not the software? Good luck with that philosophy.
        Perhaps you’d like to use it to explain why Apple is struggling so much in the tablet market with what, only 90% or so market share?

      • Anonymous

        Easy: up to now, competitors have been selling essentially identical or inferior hardware, at the same price. And, that’s 70% share these days, not 90. Watch what happens in the next couple of years, as Apple’s stranglehold on component supplies lessens, and low-price competitors can emerge. Hint: look at the smartphone market.

        But actually, my argument is that software stuff (Facetime, Siri) is a harder sell than hardware. If Facetime, which was supposed to be the next big thing, has had any impact, I’m unaware of it. Everybody is still on Skype (software = network effects); Siri only works in a handful of languages, and in even fewer countries, so customers in 5 countries are all excited about it, the rest of the world is “meh”. Better screens, IO, batteries,… work everywhere. Basically, 20 software packages (which all platforms offer by now) account for 90% of users; and killer apps (MS Office, Angry Birds…) are very rare. FaceTime isn’t one of those; Siri might become one, if it ever gets universal, works well, and has no competition. right now it isn’t universal, doesn’t work that well (my friends usually have to repeat twice), and.. it already has competition for 3/4ths of the use cases.

        Also, people may start to value standard IO more: the latest Samsung Galaxy Note requires a $2 cable to connect to a TV or a HD (different cables though ^^), can get data to/from any USB or DLNA device without need for any extraneous software on either side, and a standard USB-microUSB cable… How much would that cost for an iPhone ? How much extra stuff would you need to install ? How many times would you be caught out ? OTOH, I still don’t understand how thinness matters in a tablet, which is essentially sedentary, especially a 10″ one with a reflective, low-contrast screen. Such marketing/PR flukes are bound to be corrected.

      • davel

        The tablet market share includes B&N as well as Amazon. The others are of no consequence for now. Also it appears some vendors are already folding.

        Siri will grow and will be a huge deal if they put it in the Chinese market.

        As for your cable to the tv for a mobile device would a wireless solution not be better for this niche? I do not know what the capacity constraints are for wireless, but that would be an ideal physical layer in my eyes.

      • Anonymous

        Wireless is not only possible but built in, my phone could do dlna server and client out of the box, for free. Depending on wireless landscape, or at work when visiting clients… Cables are always handy though, you never know what features,security, old stiff you’ll come across.

      • unhinged

        From what I’ve heard, AMOLED screens don’t have a long working life, with degradation after 18-24 months. If this is true, the benefit of such a screen is not as high as you claim. First-time buyers of the device are sure to find it irrelevant, but the resale value of the device will be nowhere near that of an iPhone.

      • Anonymous

        I’m sure Apple forums are full of rumors about how AMOLED sucks. Never saw any of those anywhere else though, so I’m assuming it’s bunk.
        You are right in that iPhones offer very good resale value. I’m not sure how TCO for the typical couple of years works out though, with the much higher purchase price, higher software costs, higher peripheral costs, and fragility of the thing. I’d be interested in hard figures !

      • davel

        We will find out in a year or two.

        I had read where the issue with AMOLED was volume. The suppliers could not ship in the volume Apple needed.

        This may change, although I read where there is another LCD screen with different materials that Apple may use.

      • Anonymous

        ” I’m not sure how TCO for the typical couple of years works out though, with the much higher purchase price, higher software costs, higher peripheral costs, and fragility of the thing. I’d be interested in hard figures !”

        The hard figures are, what?

        – The leading edge iPhone at any point in time costs about the same as the leading edge Android phone. I’ve no idea what you mean by “higher purchase price”

        – The only peripherals bought by the vast majority of phone users are BT headsets (which cost the same for everyone) and cases (which again cost as much or as little as you wish on both platforms)

        – I’ve no idea what expensive apps you imagine come under “higher software costs”. Are you upset at the cost of, I don’t know, iMovie at $5, because there are just so many alternatives on Android that do a better job and cost less?

        The only item on your list that makes a lick of sense is the fragility of iPhone which (for 4 and 4S is larger) than for most phones, but not to an extent that materially affects the TCO.

      • Anonymous

        - Price: Top end Android is a lot less expensive than iPhone: just got a 48GB+swappable cards Galaxy Note for 550euros, 64GB iPhone is 850;
        – many people i know do buy extra sync/charge cables, sd cards
        – and spend easily 20-50 bucks on software that’s free on android (mine came with office suite, dlna client, dlna server, any format video player, lan client, ftp client and server, pc remote control, offline map of all the cities i want…

      • Anonymous

        You do like to just make stuff up, don’t you?

        – How many iPhone users do you think have bought SD cards? And I suspect damn few have bought “extra sync/charge cables”, not least because they can (within limits – there have been minor changes along the way that limit complete backwards compatibility) use the sync cables and chargers they used for their Apple devices four years ago.

        – Few people in the Apple world want to get into the hurt that is DLNA — they use AirPlay. If they want to play other video formats, there are a dozen free video players on the Apple Store. I’ve no idea who wants to use an FTP server (or even client) on their phone — Apple users live in 2011 not in 1995. There are free VNC clients on the App Store. I’ve no idea what a “LAN client” is — it sounds like something from 1984 that you bought from Novell as an Add-On to DOS 3.1. etc etc
        You are AWARE of the App Store? Have you ever actually LOOKED there? It’s populated by a MASSIVE amount of free stuff, and a MASSIVE amount of ridiculously cheap stuff.

        Please list, with numbers, the software that one NEEDS to buy (can’t find a free version) on iOS that matches what you get for free with Android.

      • Anonymous

        - everyone i know who can use sd cards does (at about 30 euros for 32GB). Apple customers can’t, and Apple charges 100 euros for 16 or 32 GB. Non swappable.
        – Apple does require more cables, and those are more expensive. Good for you if you already bought them…. Or not.
        – the dlna rant, and the rest from then on, really makes me think you’re a paid shill. Even the most rabid fanboy couldn’t fail to see how open standards are better than proprietary Apple stuff , especially since dlna seems to be working fine everywhere i go ? As opposed to Airplay, which works almost only with Apple stuff.
        – If you don’t know what accessing files over LAN means, you might want to look it up, network are this new thing that got invented recently… It’s like iTunes, but fast,free, and compatible with all contents and devices !

      • Anonymous

        How can breaking a phone not materially affect its TCO ?

      • Anonymous

        Oh for crying out loud. This is elementary mathematics. Let’s suppose the breakage rate of iPhone 4s is TWICE that of other phones — 2% in the one case, 1% in the other. These numbers are rounded but are not far off.
        Then the MEAN cost of an iPhone 4 is 1.02x the STICKER cost, and the MEAN cost of another phone is 1.01x the STICKER cost. That’s just not that much of a difference.

        In practice, the difference is even smaller. Apple does not have an official policy regarding breakage (I suspect because such a policy lends itself to abuse), but I know that, as a real-world matter, store managers have discretion to replace broken phones without cost, and they do so when they consider breakage to be “unfair” or “unexpected”. I suspect Samsung, HTC et al do not have such policies (and if they do, good luck co-ordinating between them and the Best Buy or VZW store or whatever where you bought the phone), so that, as a practical real world matter, the MEAN cost of an iPhone 4 (taking breakage into account) is below that of alternative phones.

      • Anonymous

        “‘m interested in your source of accidental breakage data. You seem very certain it’s only twice…”

        I was doing the numbers from memory, but I remembered things slightly incorrectly — or perhaps I was remembering a different study reporting something else.
        Regardless, I looked for the latest numbers and, sadly the numbers are worse for your point than my remembered numbers…

        Look here:
        http://www.squaretrade.com/pages/cell-phone-comparison-study-nov-10

        The important points are
        – If you care about OVERALL failure for any reason, over one year, we range from iPhone 3GS at 11.7% to iPhone 4 at 15.9% to “other smartphones” at 16.9%
        – If you care about ACCIDENTS (on the theory that other hardware failures are a hassle but will be covered by warranty) we range from Blackberry at 6.7% to iPhone 3GS at 9.4% to Moto and HTC at 12.2% to iPhone 4S at 13.8%

      • davel

        I would argue that Apple should stick to its knitting. Focus on excellent products and keep it simple.

        Samsung can have success building many devices for all types, but it also shows a lack of vision. Not that they cannot make money doing so, but I would think it is inefficient coming out with something new every few weeks because you want to change something. It also splits your products. It makes it harder for your third parties to support you when there are too many types of devices out there. You lose your ecosystem.

  • Anonymous

    It will be really interesting to see next quarter’s numbers. How much of a spike does Apple get on the iPhone-4S? How much does Nokia get from the Lumia? Does the Nexus drive Samsung even further? Can Sony turn that little uptick into a sustained recovery?

    • http://www.asymco.com Horace Dediu

      HTC today warned that their revenues will be up to 23% lower than expected. Indicated zero growth y/y and down sequentially for the first time in two years (that freefall was due to their reliance on a Windows Mobile portfolio.)

    • kevin

      Apple also gets to even its year out and “catch up” with the rest, via a14th week in this quarter. (Apple’s quarters are exactly 91 days, so every 6 years, it needs to add a week to a quarter. Its peers use 90-91 (leap year), 91, 92, and 92 day quarters per year.) The extra week should add another 2m+ iPhone units.

      HTC’s warning indicates it will have shipped about 10m units, about 2 to 3m less than it forecast just 3 weeks ago, which was already 2 weeks after the launch of iPhone 4S. (HTC shipped 13.2 units last quarter.) Analysts suspect that two of HTC’s top customers, Sprint and Verizon, have seen its customers continue to choose iPhone 4/4S.

      • Gregg Thurman

        “Apple also gets to even its year out and “catch up” with the rest, via a14th week in this quarter. (Apple’s quarters are exactly 91 days, so every 6 years, it needs to add a week to a quarter. Its peers use 90-91 (leap year), 91, 92, and 92 day quarters per year.) The extra week should add another 2m+ iPhone units.”

        That extra week isn’t going to help any, because of the delay in launching the iPhone 4S (OCT 14). When launched there were only 77 days remaining in the quarter.

        More important than the length of the quarter is the aggressive manor in which Apple added countries and carriers in the iPhone 4S launch quarter.

      • Anonymous

        I suspect a far more important issue is how fast Apple can get on top of their problems.

        At the “local” level, we still have unexplained battery drainage (not to mention that part of the iOS 5.01 solution, of simply killing the frontmost app when the screen goes dark is a real PITA for apps that take a long time to launch).
        At the “global” level, we’ve seen
        – way too much server overload on the release of iOS5, followed by
        – way too much server overload on the release of iTunes Match
        – occasional shutdowns of services — and what’s worse, no decent feedback on the phone about why something has stopped working. Just this morning Siri appeared to be out again, but the phone gave no useful indication of that fact — just really long “think times” followed by nothing happening.
        Yesterday there was period when traffic on maps appeared to be down (not clear if that was Apple’s or Google’s fault)

        A company at the top of the world has the problem that any sequence of mistakes can and will be turned by the press into a tipping point — “Apple has become so large/rich/greedy that they no longer care about doing what they used to do so well, blah blah”. I think Apple HAS to come up with solutions to both these issues (power under iOS5, server outages and overload) SOON, before they enter this dynamic with the press, and that these matter far more than one more week in the quarter, or when shipments of iPhone5 to Vietnam start.

      • Anonymous

        Yeah, well, my wife has been grousing a bit about her iPhone 4s. I said I’d trade her across fair and square for my iPhone 4. As I reached out for her phone, she grabbed it back. No trade.

        Everyone knows there are some minor software issues that need ironing out. Biggie. That’s the nature of new stuff. Apple always solves these issues, and usually far faster than their competitors solve theirs.

        What Apple needs to do is what it’s always done: stay on the cutting edge and put the customer first.

      • Guest

        “I said I’d trade her across fair and square for my iPhone 4.”

        Slightly off topic: And you are surprised that she was mad? I know marriage is not easy but trading your wife for a iPhone 4 (not even 5) …

  • http://www.facebook.com/james.scariati James Scariati

    In the first set of graphs, there are five lines in each but only four labels. It looks like the purple line is unlabeled? What is it?

    • http://www.asymco.com Horace Dediu

      The purple represents the average of the four.

  • http://twitter.com/bennomatic bennomatic

    Looking at the Samsung spike, it appears to me that this is not based on units sold to consumers, but rather shipped into the channel. Of course for Apple that is the same number and Samsung doesn’t report end-consumer sales. But with that being said, shouldn’t the description say “shipped” versus “sold” for clarity?

    • http://www.affenstunde.com James Barnes

      If this is the case might it be, in some way, indicative of better carrier relationships?

      • http://twitter.com/bennomatic bennomatic

        I would certainly agree that there would be a correlation with good carrier relationships. That is to say, I’m sure that all the carriers are thrilled that they don’t ever have to sit on stock long enough that it gathers dust on their shelves.

    • gbonzo

      Apple does not report end-consumer sales. As everyone else, Apple reports devices shipped into the channel. That is what they get paid for.

      The late Palm was the only one that ever gave two numbers. So in addition to devices shipped into the channel they also gave an estimate of sell-through.

      • kevin

        Apple often does provide the change in iPhone channel inventory during their conference call. That allows one to get a fix on end-consumer sales.

      • http://pulse.yahoo.com/_KEVQDCIMSOI6AFXXFFS5ZFFWBA PeterK

        For Apple and only Apple: Units shipped = Units sold +/- 3%…

      • Chandra2

        Google announces Android activations

      • http://twitter.com/bennomatic bennomatic

        Right, but Apple has repeatedly indicated that they are basically selling every device they can make, so those numbers effectively sold-to-consumer.

    • http://www.asymco.com Horace Dediu

      Samsung does not report shipments either. The numbers here are estimates which were discussed in previous posts.

    • Anonymous

      I did not go anywhere near Black Friday sales; but it has been widely reported that Amazon was selling Android OS phones for $1. As I am pretty sure there would have been a massive demand, could a significant percentage of these have been from Samsung and might that have accounted for their spike in sales? It is pretty clear that HTC was not a major benefactor.

      (Sheeple, wanting to be iTard)

  • TwangisKhan

    It’s more and more clear Samsung is the only clear winner with Android. Their control of the hardware stack gives them a clear competitive advantage over the rest. I must say I’m surprised in took such a short time for HTC to falter. How is Android going to grow long term without supporting companies making any money?

    It is conceivable Samsung could corner the hardware market for Android. Google doesn’t know enough about hardware to save Motorola and it’s not like they can give Motorola a pricing advantage with the software.

    Once Samsung corners Android, what will prevent them from forking?

    Boys and girls this is nothing like the PC wars.

  • Andeas

    Why are the sparklines unordered?
    Please order them from highest to lowest score so that the ranking becomes apparent.

    Otherwise: Great, thx.

  • Anonymous

    I’m interested in your source of accidental breakage data. You seem very certain it’s only twice ?
    As for why it’s harder to coordinate samsung and best buy than apple and best buy,… Why again ?