The end of phone vendor tiers

It’s time to review the mobile phone market at the end of the first quarter of 2011. Before I begin, I’d like to remind that this analysis will span multiple posts and that many details will be published separately due to time and space constraints. Data about platforms, sales, profitability and pricing will be posted separately.

All data sets and chart data will be available for interaction and download through Asymco Interactive when complete. You can purchase a license to Asymco Interactive anytime and you will have access to any new data sets for next 90 days so don’t hesitate to pre-order the report.

The overall phone market grew at a compound rate of 9% over three years. The pattern of entrants focused on smartphones and mobile computing growing faster than the incumbents continues. The compound three year unit growth for the tracked vendors in descending order is:

  1. Apple 122%
  2. HTC 55%
  3. RIM 50%
  4. Samsung 15%
  5. Other 14%
  6. LG 0%
  7. Nokia -2%
  8. Sony Ericsson -29%
  9. Motorola -30%

To give an idea of the split between smart and non smart, Nokia’s smart business grew at 18% compounded while its non-smart units contracted at -6% rate.

In terms of y/y growth the market grew at 26% and the vendor ranking is:

  1. HTC 194%
  2. Apple 113%
  3. Other 103%
  4. ZTE 75%
  5. RIM 42%
  6. Motorola 13%
  7. Samsung 9%
  8. Nokia 1% (smart: 13%, non-smart -2.3%)
  9. LG -10%
  10. Sony Ericsson -23%

Sequentially, the growth rankings are:

  1. Other 28%
  2. Apple 15%
  3. HTC 7%
  4. RIM 5%
  5. ZTE -10%
  6. Nokia -12%
  7. Samsung -13%
  8. Motorola -18%
  9. LG -20%
  10. Sony Ericsson -28%

In terms of market share, the field is becoming more crowded.

The ranking of units shows a continuing relegation of Sony Ericsson and the increase of HTC to 7th spot. Apple reclaimed its fourth spot from ZTE.

As the number of vendors increases and the share of incumbents decreases it becomes harder to identify who is a “top tier” vendor or, in fact, what the tiers are.

It’s pretty clear that on a unit basis there are nine vendors selling significant volumes. The other important observation is that vendors like ZTE are graduating from “other” while “other” is still growing, now at 25% of the market.

  • newtonrj

    "The overall phone market grew at a compound rate of 9% over three years" Is there mfg capacity potential? -OR- At what market growth rate will mfg cease to accommodate? I'm trying to understand if the market is at equilibrium growth curve between consumer demand/mfg capacity or if there is still potential for exponential growth yet above the compound 9%. -RJ

    • asymco

      I don't think the overall phone market is supply constrained. There are sub-segments of it that are.

  • What is the criteria to "graduate"?

    • asymco

      Being cited in top x rankings like this one. This is partly driven by publishing your own performance and being noted for it.

  • Alan

    The first chart and explanation are a little confusing since the growth reports are over a 3 year period but the chart spans 4 years.

    • asymco

      True. The reason is that I don't have data for some vendors from 2007.

  • The first graph would express growth more clearly if it were biased by the average compound growth rate so that any vendor growing at that rate appeared as a flat line.

  • ewan

    Great data, thank you. Do we know the timeline for the "Change in Unit Market Share Ranking among eight vendors" ? (Q207-Q410 ?)

    It would seem the differentiated product (read: AAPL, and perhaps Microkia) would benefit from the high level of competition from the Asian producers since the Asian producers are selling essentially a commodity product. Market forces dictate that they will eventually have to compete on price, driving out the profits.

    It'll be ironic if Apple is able to continue to pull all the profit out of the market, while continuing to "lose market share to Android". Apple gets the profit, Google gets the Ads, we get better phones, and the Asian producers get slim margins.

    • asymco

      The time frame is quarterly starting with Q2 2007.

  • Also, I'd be interested in seeing graphs that were based on customer expenditure instead of units sold or dollars destined for the vendor. Customer expenditure would include phone cost, taxes, monthly fees, and data plan.

  • Sandeep


    Any plan on looking at the data to see if there is a region bias? Specifically, it would be good to have some hard data of sales across regions rather than hand waving you see from other analysts on the subject.

    PS: Going to subscribe to Asymco interactive

  • debugger

    Do we know what goes in the "Other" category and why it has grown so fast, most notably in 2010. Are these mostly featured phone makers?
    Thanks for the charts. Especially the last one, very interesting..

    • asymco

      Other includes every other phone vendors than those listed. The main growth has come from so-called "grey market" producers who are building products sometimes without certification or licenses.

  • Dave Fischer

    I dislike the stacked percentage charts. They work at cross purposes with visual communication. At first glance, they read as line / series charts, but they're actually area graphs. So they confuse the changes of an individual data series with the changes of all the data series below them. And since the order of stacking is arbitrary, what is conveyed for an individual company is arbitrarily dependent on whatever "strata" it happens to be upon.

    Simply: even though e.g. Apple might be growing in marketshare, it appears to be in the decline due to downward motions of the other data it's stacked on top of. The chart requires the reader to interpret bar width that is changing position over the course of the graph.

    I'm not sure what a better presentation method would be, but I'm increasingly convinced this is a poor type of graph.

    • davel

      I too have an issue with them as you clearly state.

      But also it is hard to follow one grouping and get a sense of increase/decrease if the change is not large enough.

    • berult

      The first chart is peak and valley sedimentation.

    • asymco

      There is no perfect chart for every situation. Anyone who subscribes to the Interactive version of this will be able to try their hand at looking at the data in different ways. Nothing would please me more than to see different visualizations.

  • CndnRschr

    Agreed. I've had this problem in showing data in this format as well. One way to mitigate the depressed neighbour effect is to arrange the stack in order of growth over the time span (rather than initial percentage of market). Even this gets messy when a vendor both increases and decreases over time. It's not so easy to plot in Numbers either (anyone know how?). There is also the problem of too many vendors vs colours available. Maybe a rainbow graph?

    • berult

      The second is pure fluid dynamics.

  • Ajs

    Re: stacked percentage charts comment from Dave Fischer. I agree that these charts are hard to read. If you're going to use them, a better algorithm for stacking would be:
    Put the item with smallest absolute change on the bottom of the stack 
    For each of the remaining items: Take the item that when placed minimizes the change indicated by all the currently stacked items.

    The general idea is that with each placement, you try to keep the top line of the graph as close to horizontal as you can. (For simplicity, with each placement, just minimize the slope of the straight line between the start and end points of the range).

    This doesn't eliminate the problem with these charts but it would be better.

    • berult

      The third, an ECG.

    • berult

      All speak with allegorical iridescence of the underlying symmetry of measurable phenomena, while being faithful carriers of facts in all their precision-centric blandness.

  • chandra

    My shortest response so far:
    I'm speechless Horace!
    The more things change, the more they stay the same.

  • in the first chat, the use of 2 similar green lines which cut across (guess is they are Apple and SE?) confused me for some time till i looked at the numbers.

    i would love to have a different color major if possible 😀

  • in the first chat, the use of 2 similar green lines which cut across (guess is they are Apple and SE?) confused me for some time till i looked at the numbers.

    i would use a different set of colors if possible 😀