The symmetry of share shifts in mobile phones

Thanks to a reader’s suggestion, I took another look at the profit and market share data from the mobile phone market overview posted last quarter. This time to compare the vendors on two dimensions at the same time.

I plotted the vendors on share of market and share of profit axes in two different points in time: Q2 2007 and Q2 2010. I further broke the chart into four quadrants as shown:

I chose to label the quadrants to indicate the possible categorization of the companies.

  • Dominant: large market share and large profit share
  • Star: low market share with large profit share
  • Fading: high market share with low profit
  • Marginal: low market share and low profit

It’s interesting to note the shift over the time period of the eight vendors. Whereas Nokia moved from Dominant to Fading, only Apple rose from Marginal to Star. The Marginal players dispersed their placement with Samsung moving laterally toward Nokia while RIM followed in the Apple trajectory and increased profit share while maintaining relatively low share.

Thinking about motion around these two dimensions could be reduced to a vector model. If each vendor is represented by an arrow showing direction and distance moved, we can compare each vendor vs. a symmetric competitor which might account for the loss or gain of share.

The vectors can be paired as follows:

Although the data does not contain all market participants, the shifts in share are quite closely conserved with the vector magnitude (arrow length) and angle (direction) paired. Note how Nokia and Apple swapped Profit without too much impact in Units volume. Similarly, RIM and Sony Ericsson traded profits and volumes quite symmetrically. Samsung also took volume while Motorola lost it. HTC and LG did not change positions dramatically.

Finally, moving all vectors to the same origin leads to this view of share gain/loss over three years in two dimensions:

It becomes quite easy to see who the winners and losers are and by how much.

This model is only an observation of the shifts in market power and is not conclusive about where share was gained or lost, however there is compelling symmetry which also matches the intuition of  observers.

  • Sälli

    Beautiful, just beautiful. Data visualization truly is a form of art, and you are a master at it.

  • I am discounter, so I feel Nokia's pain. I however, do not understand why they would make so little profit.

    In discounting the most important measure after margin is stock turn. So, if a product is a low margin product, what we then try to influence is its turn. I am wondering how you treat this Horace.

    If their margin is low (given) and their turn is also low (don't know), why are they bothering?

    Apple should be a case study in all business school. They are doing something very different, totally unconventional.

    • asymco

      This is not measuring profit margin, but total profit share. That means how much profit did Nokia retain out of all the profit that was present in the entire industry. I exclude losses. When an entrant takes the profits away from an incumbent, it's usually a sign of a disruption. (see also the linked articles)

  • Billy

    Surely those manufacturers with near zero volume and near zero profit (mostly Android supporters) are not simply marginal, they're totally fucked.

    • This chart is misleading. HTC being near the origin of the axis doesn't mean they are having a very bad profit. It means they simply have lower profit than Nokia or whoelse

      If company A makes 1 billion revenues and 200 million of that is profits, while company B makes 10 billion revenue with 2 billion profits, does that mean that company A is doing very badly??

      Of course not. They have 20% profit which is the same as company B, yet they only have 10% "profit share" compared to company B.

      Do you understand now how these charts work?

  • They are gaining market share on the low end where they make no money. They are very upbeat though about the sales of Galaxy S. Apparently it is doing very well for them. I however do not think they have a bright future, too many lines and no focus. Apparently Bada is as important as Android, they are a seriously confused shop.

    I am wondering when Apple will dump them as a supplier. If they manufacture iPhone/iPad parts that surely gives them a window to Apples strategy and that cannot be good.

  • AlleyGator

    Great graph, Horace.

    In the context of Android, what's interesting to me is that Android is turning some winners into losers and vice versa. Motorola has had a great run with the Droid, so even while their unit sales are way down, their profit is way up (positive now, finally!)

    In essence, when I look at this graph I see proof of what I'm noticing: that Android has created a product that can compete with Nokia on volume, price, and carrier selection by letting talentless software designers at manufacturing companies stand on google's shoulders, and thus give them a shot at selling lots of phones to the 80-90% of people using bad phones.

    In my opinion, Android isn't really going to be much competition with Apple on ANY carrier that sells both phones. Look at AT&T, look at the European carriers. Find any carrier whose iPhone sales lag their Android sales.

    When Apple busts out of AT&T exclusivity, they better have their supply chain ready to explode. There was a lot of talk about how Apple went after a license to use LiquidMetal technology, and after reading the interview with the FoxConn CEO on how difficult it was to make the iPhone 4 case, my guess is that they'd need LiquidMetal just to produce them fast enough to meet widespread demand.

  • Billy, Yes you are right.
    It reflects Google's strategy; make the handset (profitless) "free" and run suite of Google applications.
    The handset makers have little choice;
    (a) Adopt Android and have chance against Apple, or
    (b) role their own (ex. Palm, Nokia), or
    (c) get out of smartphone market (get our handset market altogether).

    Handset market will end up like the PC industry ( Intel and MS taking +80% of industry profits), a very few players will take majority of profits.


  • Narayanan

    The summary of all summaries! The vector plots show the trends clearly.

  • Pingback: Mobile phones: Who's up, who's down and who's moving sideways | All in one Articles | Submit free articles()

  • Gandhi

    Great graphs that succintly summarize the market players. Only suggestion I would have in the vector graphs is to show different colored arrows (may be red) for total market growth/units shipped and total profit growth over the same time period, just to see how the overall smartphone market did compared to each of the individual players.

  • rossor

    It will get worse for the Android makers.

    While Apple has a distinct brand and unique set of benefits, Android manufacturers have little to set one apart from the other. When the pie stops growing, they'll have to compete against each other on price. That's a path going nowhere.

  • Joe_Winfield_IL

    I know they are a distant 3rd or 4th in mindshare, but it is interesting to see RIM succeeding along with Apple. Obviously their star doesn't burn as bright, but they are the only other manufacturer with a truly differentiated product and ecosystem (BBM, hosted email, etc.). I wonder what this portends for WebOS when HP finally decides to start making product.

    Also, MOT has probably made their entire upward move on the profit share axis in the last 9 months. I bet this graph would look different on a shorter timeline. I for one would love to see this as a regular (quarterly or annual) feature on Asymco. My guess is that Apple will continue on its trajectory but the others will continue to shift around quite a bit.

    • asymco

      I will update this data on a quarterly basis.

  • Is is just me or Android caring companies are not very profitable. Seams the only thing that makes money for them is the paid licensing of the "Droid" word from Lucasilm by Verizon wireless.

  • Tom


  • Christopher Brown

    Love this article and its graphs. I just wish it was required reading alongside the dung-pile of "Android rules the mobile space" articles that are churned out ad nauseum.

  • Pingback: Mobile phones: Who's up, who's down and who's moving sideways | Mobile Press()

  • very interesting …apple=low market share but apple = high profilts…

  • RobDK

    A picture speaks a thousand words.

    Absolutely fantastic visual synthesis of the data previously presented by Asymco!

  • Pingback: Winners and Losers of the Smartphone Industry: Flow Chart - Phone-stuff()

  • Pingback: Winners and Losers of the Smartphone Industry - Phone-stuff()

  • roberto

    no mention of motorola at all. Did all those flat flip phones die?

    • asymco

      Motorola was mostly unprofitable over the period studied. It also lost significant share. The vector representing it is certainly shown in the article labeled MOT.

  • ziggystarduck

    Excellent analysis. I’d like to see a comparison of profit share for developers in the app stores/marketplaces though I guess a lot of that data isn’t published.

  • "there is compelling symmetry"

    See also: ley lines.

  • Pingback: Winners and Losers of the Smartphone Industry | iPhone World Blog()

  • Mark Hernandez

    The design of the Galaxy Tab and even the commercial for it takes many cues from Apple's products and marketing. We've been wondering when a company was going to try that. And I wonder if Samsung is feeling brave right about now because they supply Apple with the A4 processor. Very interesting. We all know imitation is the sincerest form of flattery, but when you see the recent graphs of who is suing who in mobile here and here well, it's hard to figure out but fun to watch.

  • Nate

    It seems that financially the only companies to survive in this market are those who produce their own software and hardware – Apple, RIM, Nokia, HP/Palm. I'm not sure what that means for Android and Win7. It will be interesting to see how this graph changes in the next year with the maturity of Android and the introduction of Win7. We'll also see Nokia's new OS and HP/Palm will finally release a new phone.

    My personal take is that manufacturers will have to either choose Android or Win7 if they want to survive. Apple, RIM, and Nokia will become the dominant players although I really hope that HP makes some headway. Conclusion – Android and Win7 are going to be the bain of most companies here in a couple years.

    Oh, and the manufacturers are really screwed because the winners of this industry are not going to be competing on hardware specs to win customers (that is now taken for granted by consumers), the real money is in the services offered to differentiate the products. That is the real reason the Apple and RIM are doing well – the services behind the phones.

  • Peter

    Very, very nice indeed. Keep up the good work!

    One small niggle, for the final chart you say (emphasis mine) "Finally, moving all vectors to the same origin leads to this view of share gain/loss over TWO years in two dimensions:" but the chart axes are labelled for "3yr."

  • Pingback: Process for the Enterprise » Blog Archive » Profitshare and Marketshare()

  • This article seems to say RIM has a much larger share than represented here:
    I feel like I'm misunderstanding something – platform v. hardware share?
    Thanks in advance for any clarification.

    • asymco

      This is not a graph of share but a graph of share growth in a three year period starting with the launch of the iPhone (2Q2007 to 2Q2010). The market being measured is all mobile phones sold globally by the top eight vendors.

    • Chris

      Also I believe the Nielsen report is for smart phones, where as I believe these graphs are for all phones.

      • asymco

        And the Nielsen report is for the US and for recent purchases.

  • normlewis

    Very compelling and the visualisations are excellent. However, it wasn't necessary to have such great graphics to see the writing on the wall a few years back. It would be great to plot the axes but with the operators as the subject (vs the OTP players like Google, Amazon etc).

  • asymco
  • Pingback: Smartphone market-share vs profit-share visualized | Best Smartphone Blogging()

  • bsacramento

    Very interesting graph. I dont know if you have the data on this but any thoughts about plotting a 3rd dimension to visualize? Namely this data show trend over a 3 year period but the market itself hasn’t stood still during that time. It has grown over the last 3 yrs with segments within it varying in growth rate and profitability (e.g. smartphone), and some of the players were already established vs. newcomers in the period represented. It would be interesting to see not only profit+mktshare trends but also which players are winning in new market growth vs. exchanging existing market. Don’t know if the data exists for that or if result would end up visualizing additional trends or not, but from a thought-experiment at least it seems like it could be interesting.

  • Pingback: Smartphone market-share vs profit-share visualized | Ipad & Iphone Tech | Ipad & Iphone Technology Blog | Ipad & Iphone Tech()

  • kevin

    Great visualizations, one issue with the validity of the representation: because the data for apple starts at zero zero in Q2 2007, they only have one way to move (up and out). I think more realistic proportions for the vectors might result after an initial period for apple to establish itself in the market share. like starting the plot in Q2 2008 for example.

    • asymco

      Q207 was the first quarter of iPhone sales, so yes, they did not have anything before, but I'm measuring share change, so although they only had 0% share, they gained x% share and NOT increased by x%. In other words, to say they had zero one quarter and 10% at a later time means they gained 10% of share, not that their share was 10% bigger. The articles linked describe the raw numbers in much more detail. This visualization is only a way to summarize this rather complex set of data.

  • I don't mean to be the lone skeptic, but does any one believe that such outsized profit as the one now on the iPphone could possibly be a temporary phenomenon due to well-know factors?

    Windows 7 Mobile deluge will add to the price pressure as its OEM approach would mean even less s/w work for OEMs and more lower-priced devices.

    My own take is that 34% net margins on iPhone are more likely to settle to something close to 15%-20% range over time (still well above all other players, but much lower then now), although it could be a while.

    • asymco

      I am forecasting lower ASP but not lower margins. For some reason which I can only speculate on Apple chose a high-end only strategy for the iPhone vs. a portfolio approach as they took with the iPod (and may take with the iPad). The high-end strategy may be due entirely to the unique distribution problem the product has: that operators are in the way–an issue that does not plague any of the other products Apple sells.

      For this reason maybe Apple is content with a skimming strategy for cellular while going after volume with non-cellular/non-regulatory-fettered products.

  • Pingback: Smartphone market-share vs profit-share visualized | iPhoneBizBlog()

  • Niilo

    This post is a triumph of style over substance. It is not necessary to invoke vectors etc. in order to talk about winners and losers in the handset market.

    You have made some really good points about this space on other posts, but here you have just overegged the analytical pudding with wacky data-mining (why would the vectors symmetry show who has stolen share from whom? huh?).

    A rare self indulgent departure from your usual excellence

    • asymco

      I concluded: "This model is only an observation of the shifts in market power and is not conclusive about where share was gained or lost"

      Visualizing multi-variate data may be an indulgence but so far it has attracted more viewers than prior versions of the data so it has some value beyond my gratification.

  • Why does the axis need to be symmetrical? Nokia'sprofit or Apple Margins seems to be outliers. Therefore, the first to charts are misleading. I suggest the axis scale to be reference to median industry profit / margin and that way you can visualize who is better / worse of.

    • asymco

      Not sure what you are suggesting. If I scale the to the median then I'd have to discard Apple and Nokia from the graphic. And why can't you see who is better and worse off? The vector diagram at the end is pretty clear who moved how much and where.

  • Pingback: Smartphone market-share vs profit-share visualized | iPhone Development News()

  • Pingback: Mobile phones: Who's up, who's down and who's moving sideways | All in one Articles | Submit free articles()

  • Lee Penick

    Vector investing. interesting.

    Have you looked if there is a correlation between the vector length and the p/e (or other valuation metric)?

    • asymco

      That might be an interesting analysis. It stands to reason that the profit vector length would be correlated to P/E (since length indicates growth over time and hence earnings acceleration $/time/time). The share measure might not be correlated to much financially.

  • Pingback: Парочка интересных графиков |

  • Pingback: Vem stämmer vem i den mobila världen | MaxiMac()

  • Gavin Costello

    You succeed at Visualisations that it is almost impossible to pick holes in any of your work 🙂
    Looking at the first two graphs however, I was struck by how much better they would be if the movements were plotted against each other in one graph.
    Your follow on graphs partly solved that, and the single source graph at the end pretty much removed the need for my requirement (though I'd still like to see it!)
    Wrt the data, one of the interesting things I would like to see (and you do it so well in other stories) is comparing Apple's trajectory with the first three years or so of other previous stars in related spaces. And especially what happened to them overtime.
    I'm sure Apple are covering their bases by keeping their eye on this sort of knowledge, and if they keep churning out in demand, usable products like they have been, they'll continue to break forecasting moulds, but I't would be an interesting exercise all the same.
    Keep up the good work.

  • Old One

    Very nice use of the BCG and SPACE matrices concept. Shows how robust they are.

  • Pingback: Quem está ganhando a guerra dos smartphones? | Loucos Por Celular Brasil()

  • Incorect. They are fading because mose of their phone and smartphone business is NOT about Android. If 95% of their business is non-Android but losing money, do you really think 5% of the rest with Android will make that much of a difference to compensate? Galaxy S is their first real Android phone and it's also their most successful and profitable product ever, but it's just not enough to compensate for the other phones.

    To be more profitable now, they had to start selling Android phones earlier. Just look at HTC which sold Android from the beginning and is growing 20% per quarter, and already has 2/3 profits of Nokia, while being 3x smaller in revenue than them.

  • Horace your charts are very misleading, just like your other article. You should update your article and say what your charts really represent because I don't think most people here understand what they mean.

    Just because one company's total profit is only 5 or 10% of the profit of another company doesn't mean it's doing badly. People look at your chart, see HTC so close to the origin of the chart, and they think their performance is very poor – when it's completely FALSE. HTC has huge profits and is growing 20% per quarter.

    If company A makes 1 billion revenues and 200 million of that is profit, while company B makes 10 billion revenue with 2 billion profits, does that mean that company A is doing very badly?

    They both have 20% profit, yet A has only 10% total profit amount compared to B.

    If you really want to be fair, update your post and say what the charts actually mean so people are not mislead anymore.

    • asymco

      Lucian, the charts are not mis-labeled, so I'm not sure what I can do to improve them. Regarding HTC, having low share of profit and units does not make them a poor performer, but the last chart showing the change over three years in these two shares indicates how the company has evolved its position relative to the peer group.

      Some people care about absolute performance, like those who own shares or work there. Some people care about relative performance, like competitors and analysts looking at the whole industry.

      Readers can see the absolute figures in the original data sets that are linked. This article is about relative performance over time.

  • Pingback: Who Is Really Winning The Smartphone Race? | Gizmodo Australia()

  • Pingback: Una imagen vale más que mil palabras | Comentarios - Blog()

  • Excellent chart and analysis! I've long wanted a view showing smartphone growth and capture of profits and you've taken these and created a very helpful graphic. Thanks!

  • Pingback: La guerra degli smartphone: una visuale d’insieme | bruno trani dot info()

  • Pingback: » La guerra degli smartphone: una visuale d’insieme()

  • Pingback: Who Is Really Winning the Smartphone Race? | Noob Progs()

  • asymco

    Point taken, no offense taken.

  • Pingback: Sony Ericsson market share shift doesn’t make for pretty reading | Xperia X10 Blog()

  • Pingback: Nokia ‘fading’ in battle of the mobiles | Portable Gadget()

  • Pingback: Windows Phone 7 launch: liveblog from Londonblog! |

  • Pingback: asymco | Escape from license-land: Measuring phone vendor commitment to licensed mobile platforms()

  • Pingback: asymco | What really matters? Market Share vs Profit Share()

  • Pingback: Windows Phone 7 launch: liveblog from Londonblog!()

  • Pingback: Smartphone market-share vs profit-share visualized -

  • Pingback: Windows Phone 7 launch: liveblog from Londonblog! | Tech2Crave()