Categories

When will the US reach smartphone saturation?

The latest comScore data places penetration at about 61%, up over 10 percentage points from the same time last year. The 50% last year is itself up nearly 14 percentage points from the same time in 2011. This progress in the diffusion of smartphones is shown the following graph:

Screen Shot 2013-10-07 at 10-7-5.32.48 PM

I set the time frame for the graph to start in 2005 because there was still a market at the time. We can estimate the number of users from data supplied by RIM. They had about 2 million subs in 2005, the majority of whom were in the US.

Given these data points we can draw a new penetration graph extrapolating to the complete date range.

Screen Shot 2013-10-07 at 10-7-5.48.56 PM

 

I added a few events that mark some of the product launched during this period.

The penetration curve shown happens to be defined as the “Logistic Function” whose equation is well known:

render

 p1  is the number of periods (in months) that it took for the market to reach 50% penetration and p2 is an index of which can be adjusted for this market.

Note that I also identified successive groups of consumers adopting smartphones as defined by Rogers in his Diffusion of Innovations.  This is easily done since the formula for penetration gives us the answer (as the integral):

  • The Innovators (first 2.5% of the market) were recruited by February 2007, one month after the iPhone was announced. They were almost all BlackBerry users
  • The Early Adopters (the 13.5% which followed the innovators) were all using smartphones by the beginning of 2010. They were served mainly by iPhone 3GS and BlackBerries
  • The Early Majority were on board by October 2012, just in time for iPhone 4, Droid, Galaxy.
  • We are now in the Late majority which will run out by November 2015. The iPhone 5S came out about one third of the way through this period.

The next few years are shown in the following extrapolation. Screen Shot 2013-10-07 at 10-7-6.20.24 PM

 

The laggards will adopt smartphones from late 2015 until late 2020.

As this is the diffusion curve for the US, most other large regions are going to be shifted to the right and the global total can probably be estimated.

What remains to be done is to estimate the way the platform colors will fill in over this time frame for the US and to do the same analysis on a global level.

  • robdk

    Sure looks like iPhone is set to take the Late Majority whilst android tails off to very little or no new gain.

    My guess is Apple have done their homework In the USA with the 5c perfectly suited for the Late Majority.

    I presume Android v2.3 will pick up the laggards, dirt cheap, nearly 5 years old, and still being pushed at the bottom.

    • ankushnarula

      Based on the 5s launch weekend, it seems like Americans are keen for the premium device – in this case the 5s. However, I would bet that most of these are upgraders or switchers. Regarding Android OS version, what prohibits Android 4.2 or greater from becoming the defacto Android OS particularly as we see sub $249 tablets using high-density displays, high end processors, and plenty of RAM?

  • http://rimu.posterous.com/ rimz

    this is fascinating from a few points – including context to grasp where we are in the overall saturation cycle. It also points to the vast growth, size and permanence of the carriers’ market, which makes we curious as to how this chart compares to both the carriers’ ARPU and total revenue over the same time frame.

    • Walt French

      WSJ has an interesting article today about 4G rollout in Europe. Carriers there have spent less than half as much as US carriers each of last several years, so are a couple of years behind in data-intensive mobile use. Vis-à-vis my and @handleym’s notes on candidate users, incomes may be a bit flatter but the tradition of mobile in forms such as autos is still not as strong.

      In the article, the European carriers are said to be bundling movies and other higher-data goodies instead of exploiting the US carriers’ tack of handset subsidies. Looks like they’re exploring how to get the lowest-cost alternative and still get the payback on their investments.

  • Brrriiiaaallliiiaaannnttt

    Yea, the ‘smartphone saturation meme’ has become universally accepted, while ignoring the hard math behind ‘product adoption cycles’. Great analysis, which will unfortunately be completely disregarded by the financial analyst community.

    In the countries that are ‘shifted to the right’, it will be interesting if Apple can continue to maintain / extend the carrier subsidized model, which is allowing them so much success in the US. How the ‘colors run’ will be dependent on this. Luckily, for China, the most important ‘shifted to the right’ country, the subsidy model is just as generous as the US (And China Mobile will necessarily have to match the subsidies of China Unicom / China Telecom).

    • KirkBurgess

      Do you have a good link to the Chinese subsidy system? From what I can gather most “subsidized” Chinese handsets actually require the user to pay the full upfront cost, which is then “refunded” over 24 months by way of a discount off the monthly plan cost. Although I’m not sure how widespread this is so would love more info on it.

      • Brrriiiaaallliiiaaannnttt

        @Kirk

        http://tech_fortune_cnn_com/2013/09/17/apple-iphone-china-prices/

        Substitute dots for _

        Or search ‘A first peek at the Chinese rate cards for Apple’s new iPhones’ by Philip Elmer Dewitt

      • Chaka10

        If I may refer you to an earlier post in which I offered an explanation:

        “In the US, the subsidies work as an upfront reduction of the cost of the phone with a 2 year contract (i.e., a promise, to make monthly payments under the plan). That doesn’t work in China — consumer credit is simply not developed to the point that a carrier can allow a customer to walk away with the phone based on future promises of payment. Instead, the Chinese carriers make the customer pay for the full phone price (so if they bolt, the carrier doesn’t lose), but they credit the subsidy against future monthly payments under the plan. Over the life of the plan the customer recovers the full price of the phone that they paid.
        Edit to add: For Apple, it gets paid for the phone, in either case, by the carrier.”

    • charly

      Apple is a success in North America and Japan. Even WP is beating them in Europe so the question if they big outside USA/JP is no. Carriers hate that model so it is doubtful that the European carriers will follow the carrier subsidized model is NO. Especially now in a time that smartphone improvements are minor.

      • source

        “Even WP is beating them in Europe so the question if they big outside USA/JP is no.”

        Hmm, source?

      • Brrriiiaaallliiiaaannnttt

        @Charly WP is doing well in Europe due to the European recession and low price point of WP…still not as successful as Apple, especially in the UK. (But to be fair, Europe is a positive sign for MSFT)

        Apple is ‘big in Japan’, hugely successful, and with NTTDoCoMo, Apple is guaranteed to surpass even Android. (Apple was close to Android in JPN before NTT DoCoMo, which is why NTT gave in)

  • Walt French

    I wonder why we should assume the penetration goes to 100%. There are two demographic groups who are unlikely buyers of smartphones: the very elderly and the very poor.

    I believe it was Pew that documented that half of today’s non-smartphone-using population is in the 65+ category; eyeballing Wikipedia’s data (Census seems to be a casualty of The Shutdown), there are about 39 million Americans in that bracket, roughly 12% of the population. While some of us (that’s my demo!) are active smartphone users, there are many who just aren’t active enough to benefit from a smartphone—a larger-screen, immobile system is better in all aspects. As a SWAG, that’s about 5% who will not be users until the alternatives become worse.

    Also from Wikipedia, recent data shows 35% of individuals, 28% of households, earning less than $25K/year. (Another 36%/23% fall in the bracket earning less than $50K/year.) Many of the people in the lowest quartile have essentially zero income, and only a few of the rest have either the wherewithal or the incentive to spend anywhere close to 5% of their income on anything beyond essential food and shelter. This could easily be another 10% of the population who will not own a smartphone+data plan.

    (And yes, I *did* see the article about street people clicking on ad links for a few pennies a day of “income.” Tragic that whoever perpetrates this fraud on advertisers also perpetrates the fraud on individuals who could be productive employees. In any case, it is not scalable.)

    So I am thinking that the penetration curve gets close to 100% much farther in the future.

    Further, one of my early analytical experiences was working with a fine economist at HP, who used a similar adoption curve but was also seasoned enough to know that once the rush of adoption gives way to steady state, economic cycles start appearing. (HP at the time primarily made electronic instrumentation; this was pre-PC.) Capital goods (equipment meant to expand production) and consumer durables (products meant to last a year or more) both predictably are purchased when incomes have been rising, and the outlook is good. Smartphones are likely to show this variability (typically in 4–5 year episodes) as the market saturates, too. (I eschew the common “cyclicality” because it suggests a phase-of-the-moon regularity that just ain’t there.)

    Net-net, I think it reasonable to forecast that in the US, we will see sharply slower increases in penetration from here; absent those 15% or more that I think will be non-consumers, we might only hit 75% penetration by your dividing line between late majority and laggards. Second, if the Job To Be Done unevenly expands to encompass more users, we might well see periods of relative flattening of the curve. Finally, if laggards are income-constrained, we might see either more price-conscious users. But if they’re interest-constrained (if they don’t need mobile), laggards might actually opt for higher-priced, brand-name-reliable devices.

    Interesting times.

    • Mani Ghasemlou

      Factoring in things like Moore’s Law and economies of scale, it is possible that feature phones will no longer exist a few years from now. Even the dumbest phones will be smartphones going by today’s definition. It’s not that the laggards will be swayed into choosing smartphones. It’s that they will only be presented with smartphones as buying options.

      Some analogies:
      – CRT displays -> LCD displays
      – Rotary phones -> button phones
      – Manual transmission -> automatic transmission (this is in progress)

      • Glaurung-Quena

        Smart phones will not take over completely so long as they continue to deliver markedly worse battery life than feature phones.

        One example of where feature phones are still necessary: in poor countries where significant numbers of people live without power in their homes. They own phones, but don’t have the ability to charge them more than once a week or once every couple of weeks.

      • KirkBurgess

        Surprisingly the opposite may be true – a smartphone is possibly more important to those in poor countries as their only source of communication/banking/news. Point taken regarding power usage, but in developing countries there are a lot of good battery life features phones that get used like smartphones in that they have simplistic text based browsers that can access the Internet.

      • Davel

        Yes. I agree. Get a village windmill or solar panel to charge your device. Google already has talked about providing service to Africa.

        The military is looking at non oil based operations because it saves lives. The military started the internet. They may push the electrical infrastructure to new technologies that are not mainstream to power communications and power their vehicles. This would have beneficial side effects for consumers.

        The problem is funding. The governments may not want their population to have easy access to information.

      • Walt French

        Moore’s Law isn’t moving fast enough to make up for the battery life and data network costs to budge, and a large fraction of the world’s population is challenged to afford a basic feature phone.

        Your other examples happened over decades—today’s automatic transmissions go back to the 1940s and 1960s for their major design components. I *still* wasn’t able to find an automatic with the performance / economy mix I wanted when I bought my current wheels.

        So yes, progress. But not tomorrow.

      • charly

        The price difference between a basic phone and a smartphone isn’t large so most people who can afford a basic phone can afford a smartphone. But battery is an issue.

        LCD TV didn’t happen over decades

      • Walt French

        LCDs were first patented in the 1920s (!) and practical devices were in place 50 years ago. TVs came in a rush once several key parameters all came together (speed, power, burn-in, contrast ratio, brightness, …).

        Arguably, even their great reduction in size vs CRTs haven’t really changed the job that TVs do… we gather around a large screen in the Living Room or den, maybe have a second one in the bedroom while some have little ones in the bathroom, etc. But that MO was pretty well-defined once TVs had moved to color and small sets were mass-produced in the 60s.

        So aside from the multiple monitors in sports bars, I’m thinking LCD TVs aren’t really that different a market from the technologies that preceded them—they do the same job, just better.

      • charly

        The communicator on the Enterprise is a smart phone and that was the 1960’s. The time between that you could buy a bigger than tiny lcd tv and market dominance was less than a decade

    • Glaurung-Quena

      Another group that won’t be getting smartphones in large numbers anytime soon: young children. Even if they’re old enough to need a mobile phone, they’re not going to be trusted with an expensive device.

      • DarwinPhish

        More importantly, by law, comScore does not survey anyone under 13. So even if a child has a smartphone, it will not show up in comScore’s numbers.

      • KirkBurgess

        There is no law that smartphones have to be expensive. Android smartphones are already available under $50 unsubsidised. Also, there are already a large number of children that already use expensive mobile items such as iPods, handheld game consoles, tablets, and yes, smartphones.

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

      100% on this graph is not meant to be US population. It’s US phone users. My assumption is that there will be no non-smart devices available for purchase in the US after about 2015. I’ll do another post where I’ll give an absolute figure for these graphs.

      • Glaurung-Quena

        “My assumption is that there will be no non-smart devices available for purchase in the US after about 2015.”

        Perhaps, but only if it becomes possible to buy a smart device without being forced to buy into a data plan.

        I think the carriers are going to get a lot of pushback if they try to force everyone to pay an extra $10-20 per month for data. For instance, parents buying phones for their young children, people on limited incomes, and old people, are all going to resist the idea of being forced to pay for a smart device and a smart plan that they don’t need or want.

      • Kizedek

        Is that how it works in the US? In NL, I don’t have to have a data plan — but data is still available to me on a regular call and text plan. Therefore, if I happen to get careless and inadvertently use data, then I am charged ad hoc at a much higher rate per MB than if I were to add it to my plan. Makes sense.

        I have opted for a non-data plan, so I simply switch the cellular data option to off, and rely on WIFI.

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

        My premise is based on the assumption that nobody will make non-smart phones by then. The trend is evident in the shelf space allocation of US phone retail.

      • bloftus

        You are leaving out those that own 2 smartphones. Comscore looks at primary phones. As a physician, I now commonly see drug reps with personal smartphones and corporate smartphones. My son – business analyst for large banking company has 2 as well. Whether this will represent just a few percent or a large percent I am not sure. A businessweek posting images dot businessweek dot com/slideshows/20110213/the-20-countries-with-the-highest-per-capita-cell-phone-use has #20 Italy at 147%, #11 Russia at163%.

      • Walt French

        That’s exactly what drove my wife to get an iPhone: her HMO employer insisted that personal activity not take place on her BlackBerry. A nuisance, but to her, a blessing in disguise.

        Now we’re coming full circle: if you look at the security features in iOS7 — e.g., the ability to require that corporate apps only start up if a password has been given and that they only work on the company VPN — the nuisance of a second phone is going away. I’ll personally guess that BlackBerry is no longer selling hardware in 3 months, meaning that companies will have to rely on Apple, Good or other third-party solutions—not just that the old BlackBerrys couldn’t do anything that might get into trouble—for security.

      • charly

        NotSmartphones have the advantage of very long standby times and cheapness. Ideal for emergency phones so it wont be 100%. But i think you are right for any phone that is used regularly

  • handleym

    I’m afraid I don’t agree with you Horace. Your assumption is that the end-game here is 100% smartphone penetration, and you’re being sloppy in how you define 100%/penetration.

    In the first place there are populations we expect never to have smartphones. (No babies, through to only a few sub-teens to only some teens. Likewise for those in prison.)

    In the second place, the gating factor is not really the cost of the device, it is the cost to operate it. If every TV required a separate cable subscription to operate, we’d also see quite a bit fewer of them.
    IF the subscription cost problem is solved we could see substantially higher smart phone penetration, but I don’t yet see any progress on that front. The current thinking is interested in segmenting by voice (poor people, low price) vs data.

    When the RF tech has reached its limits, we may see some original thinking regarding segmentation. But what we have right now is essentially idiotic. The CHEAP option in data is the non-LTE option — but this is crazy because it’s more prolific of spectrum. You are using price to encourage people to use spectrum (your most limited resource) in a less efficient way rather than a more efficient way.

    Far more sensible would be plans that segment by prioritization. All plans would run everyone on LTE-A, but “Business” plans would promise that, in the event of contention (at the airport, at the sports stadium, in downtown) they get a higher priority as slots are handed out, while lower end plans gets fewer slots under such conditions.

    But that’s a long term future. Point is, right now, we’re not going to get to even close to 100% penetration because of the cost of subscriptions. There are enough people in the US for which the minimal subscription cost is both a significant amount of money and does not offer enough benefits for that cost that this matters.
    (I expect that in countries with flatter income distributions, like Japan or Finland, the maximum penetration level will be rather higher.)

    • Glaurung-Quena

      “In the first place there are populations we expect never to have
      smartphones. (No babies, through to only a few sub-teens to only some
      teens. Likewise for those in prison.)”

      Horace said upthread he was only looking at phone users, so babies and prisoners are not included in his equation. However, I agree with you that there are phone users who are simply not going to ever upgrade to smart unless forced to by the carriers (and they will do so only under protest at being charged for something they don’t need or want). Especially children, the elderly, and people with limited incomes — groups like this will want inexpensive, non-fragile phones that don’t require a data plan, and that isn’t going to change much.

    • KirkBurgess

      Have you not heard of the prepay market? Millions of current US smartphone users do not have a subscription.

  • ankushnarula

    Guys – correct me if I’m wrong – I think 100% penetration in this case doesn’t mean 100% penetration of the U.S. populace. I think it implies that 100% is the share of the mobile market. You’re not part of the 100% of the mobile market if you can’t afford because you’re too poor or aren’t eligible because you’re a child. This isn’t a social judgement, it’s an analysis of a pool of people who buy phones.

    • DarwinPhish

      comScore is the source of Horace’s data. They claim the total number of smart phone users is 145 million and penatration is 60.8%. That implies a total population of about 240 million, which sounds about right since comScore excludes anyone under 13. They probably also exclude the prison population.

      Also, the US has already exceeded 100% total mobile subscribers.

      • ankushnarula

        Not entirely disagreeing with you – but 100% adult mobile usage seems way off to me. Here’s what I threw together using current U.S. Census, 2013-Sep comScore press release and 2013-May Pew Research U.S. mobile usage figures:

        comScore smartPhone Penetration = 60% (or 143.3M)

        U.S. Census claims:
        Total U.S. Population = 316.81M
        Total U.S. Population age 14 and over = 262.95 (or 83%)

        Pew Research Claims 91% of U.S. Adult using Mobile Device:
        Pew Total Adult Mobile Market = 91% of 262.95M (or 239.28M)

        So then if we take…
        Pew Total Adult Mobile Market = 239.28M
        comScore Smartphone % Penetration = 60%

        Result is…
        Smartphone Penetration = 60% of 239.28M (or 143.57M)
        Pretty close to comScore’s 143.3M

        http://www.comscore.com/Insights/Press_Releases/2013/9/comScore_Reports_July_2013_U.S._Smartphone_Subscriber_Market_Share

        http://pewinternet.org/Commentary/2012/February/Pew-Internet-Mobile.aspx <- page is updated with 2013-Sep data

      • DarwinPhish

        You might be right. I may have over estimated the under 13 population.

      • Secular_Investor

        240 million is too low.

        Having a mobile phone is almost a necessity, having a smartphone will become almost a necessity and, as Accent_Sweden has pointed out, there are quite a number of people who have two or three phones.

  • Accent_Sweden

    Of course, this discussion is assuming we will reach saturation without any disruptive event or innovation that could extend or restart the whole process again. For example, we are on the cusp of the next wave in wireless speeds with the new 802.11ac standard. I just installed my new router two days ago and I’m already mulling what new possibilities this will offer. The new router now gives me wireless speeds on par with my fiber optic 100Mbps home line (96 Mbps down, 87 up) without even having any new devices that can actually take advantage if its maximum speeds. Access to this kind of broadband will drive demand for higher wired speeds, which will in turn open up new possibilities for offloading mobile networks and create opportunities for solutions and business cases we can’t imagine.

    Our current limited available mobile bandwidth could be doubled or tripled with offloading solutions that are 5+ years down the road. Wifi is heading to cover larger areas with more connected clients than current mobile networks. Perhaps Apple won’t need mobile operators in 10 year’s time. Suddenly, saturation isn’t about 100% penetration; it’s 200%+ as mobile becomes wearable, people have multiple devices in cars, fridges, basketballs, whatever. This isn’t unprecedented. SIM card saturation has been well over 100% penetration in many countries as people have had multiple phones and multiple devices (security systems, monitoring systems, electrical meters and so on).

    Apple could pivot its iPhone business into an iEcosystem of devices that may little resemble the current iPhone, just as it did with its cash cow iPod line transitioning to the iPhone. Bandwidth begets business opportunities. Solving the mobile bandwidth dilemma may have more to do with new wired options combined with expanding wifi coverage and speeds than trying to overcome the physics of our existing mobile bandwidth.

    So while we are projecting into the future about saturation, maybe we should go ahead and project about what will keep this train going after our current definition of saturation becomes obsolete.

    • Walt French

      While it’s *fun* to speculate about that here, you *know* that people in many companies — Apple, Microsoft, Google, Intel, to name a few — are actively working on these future products.

      One of the amazing things to me is how Apple intro’d such an expensive, barely-useful device in the face of a deteriorating economy, eventually the worst recession since the ’30s. Most smart companies would have tried to avoid this.

      And then, in 2010, they did it again, with the industry weak and IT spending at their lowest levels, Apple started its assault on the Enterprise market.

      • Accent_Sweden

        Exactly. Based on previous events, we should assume Apple will disrupt the current status quo and redefine the market. After all, when what currently makes sense isn’t useful, Apple has reformulated the question to catch the competition off guard and capture the imagination of consumers. That’s part of Apple’s secret sauce. I suggest that saturation worries are low on Apple’s list of concerns. They are surely aware of it, but are focused on what comes next that will make the idea of saturation irrelevant. Settling into a mature market isn’t in their nature or history since the days of milking the Apple II for 16 years after introduction, or at least since the iMac was introduced. They will take advantage of a mature market to make as much money as possible, but they are thinking past it.

        While the model can give us a sense of how much market share the iPhone could potentially grab in a certain time frame, I argue that discussing saturation is not yet possible since the market will likely expand greater than we can imagine with the introduction of new technologies and new definitions of what an iPhone is.

      • Secular_Investor

        And I argue that saturation is almost irrelevant as far as Apple is concerned, because they are showing that they can and are gaining market share and increasing sales in supposedly mature and approaching saturation markets such as the US, UK, France, Spain, Italy, Australia, Japan Mexico etc (accruing to Kantar’s latest surveys of users buying decisions)

      • charly

        Gaining market share compared to last year or compared to the quarter that they had an old iPhone

      • Secular_Investor

        The gain in market share shown in Kantar was Year On Year for the same 12 weeks ended August 2013.

        For example, in the US, iPhone market share increased YonY from 33.9% to 39.3% while Android share dropped from 60.7% to 55.1%

        Similarly in the UK iOS market share increased from 21.4% to 27.5%, while Android share dropped from 62.7% to 56.3%

        In a chart Comscore shows much higher iPhone US market share of sales than Kantar. They also show a three month gain of US iPhone share and a remarkably steep decline of Android market share in the 3 moths to August 2013.

        tech.fortune.cnn(DOT)com/2013/10/05/apple-ios-android-comscore/

        There are also a number of surveys which show that amongst existing users, that for every one user that moves from iPhone to Android, two or three Android users move to iPhones. This correlates with user surveys which show much higher user satisfaction and retention rates for Apple products than any of their competitors.

      • charly

        Flash memory was near the point that mobile phones would kill the iPod market. Apple had no choice than to go into the phone market.

        Assault on the Enterprise market? You mean BYOD?

      • Walt French

        “Assault on the Enterprise market? You mean BYOD?”

        No, I mean iPad—a device that directly replaces Windows laptops for many corporate functions.

        I remember acutely the 1997(?) Jobs speech in which he said the PC wars were over, Apple had lost and it was time to go after the Next Big Thing. Microsoft hadn’t bothered to block the door to the consumer, and Apple targeted it. But by 2010, Microsoft had been napping for a decade and Apple was ready to initiate a frontal assault.

        Look at Apple’s pages on iOS in Business, especially its focus on manageability and security, and you see a product that trumps the Surface on many fronts — it’s not just the breadth of the apps library that makes it attractive in businesses.

      • charly

        Storage for mp3 isn’t really dependent on device

    • Secular_Investor

      You make some interesting, though provoking points. It will be very interesting to see how these technologies develop. I am sure Apple will be at the forefront.

      BTW you are right about people having multiple phones, a number of my friends and acquaintances do.

    • AChicagoLad

      Brilliant note.

    • Chaka10

      “Of course, this discussion is assuming we will reach saturation without any disruptive event or innovation that could extend or restart the whole process again. ”

      Yes, exactly. There likely was a similar adoption curve with the original mobile phone (1G-2G). Then “true” mobile broadband and other technologies (battery, screen, etc) made the touch screen smartphone possible, which restarted the whole process. That really began — with widespread 3G, broader proliferation of WiFi and increasing home/office last-mile broadband — in the second half of the 2000’s for the USA. This is why I continue to focus, in looking at smartphone (and iPhone) market share, on the fact that not all geographic markets are at the same stage of development in terms of mobile broadband. As I’ve asked before, why would smartphone markets be the same in countries with 3G penetration in the twenties (China; single digits in India) as in countries with 70%+ 3G penetration? Why is it surprising that smartphone markets are different in countries that are well into 4G adoption from those in countries where that is just beginning or is facing structural impediments?

      The other important detail this broad adoption curve analysis doesn’t take into account is that penetration happens at different stages for different segments of the market — I expect penetration moves from the high to the mid segments of the smartphone market. Or put differently, the high-end of the market will mature sooner than the low end.

  • http://aaplmodel.blogspot.com/ Daniel Tello

    Why is there a dip in the logistic function right around now? I see a subtler one also around the 50% penetration level. The function should be smooth throughout.

  • http://twitter.com/LunaticSX Lun Esex

    What about that bottom, blue, Android-only curve? It’s leveling off. It looks like it could also match a logistic function. Is there a saturation point for Android penetration in the U.S.?

    • Secular_Investor

      Android is not just levelling off, but according to Kantar it is LOSING US market share, while the iPhone is GAINING market share in the US, UK, France, Italy, Spain, Australia, Mexico and Japan i.e. all supposedly mature “saturated” markets according to WS analysts and talking heads.

      The above also happen to be by far the most important markets for Google and for advertisers and marketeers.

      Where Android is gaining market share is mostly in less mature markets with developing economies. However, most of those gains are using forked Android versions on “white box” devices, which Google cannot either data mine or place adverts on and whose users cannot use Google Play and cannot buy or download Android Apps.

      It is therefore totally misleading to compare Apple’s iPhone market share with Android’s global market share.

      To make matters worse for Google, around 80% of their ad revenue comes from iOS devices.

      • 程肯

        “Where Android is gaining market share is mostly in less mature markets with developing economies. ”

        Which is interesting, as both Nokia and RIM started to tout their gains in developing markets, when their growth in mature markets slowed. We know how that ended up.

      • Secular_Investor

        Things could be looking really good for Apple in China, where the iPhone and iPad are seen as prestige products.

        I recently read a report on AppleInsider that there is a lot of discussion on Weibo and elsewhere about just how bad the “white box” devices are, that like Samsung, they all cheat on bench marks, that they are poorly built using the cheapest components, leaving their owners very dissatisfied.

        Is that true?

      • 程肯

        Hard to say really. Sina Weibo is just a slice of the picture, just like Twitter is in the US.

        The problem I see is that most people don’t know any better, and assume ALL mfrs are the same and cheat on benchmarks and make poorly designed products. Not everyone can handle an Apple product long enough to appreciate the differences, and even of those that do, many don’t value those differences.

      • bloftus

        You are both right. For the last 3 months Android gained 1 million uses but lost market share. Apple gained 3.7 million and gained share. For the last 12 months – Android gained 13.5 million users and lost 1% market share! Apple gained 17.5 million users and over 6% market share. I expect android to go negative growth in February 2014 but we will see.

  • David W

    There are a few issues with this.

    First, it’s likely there may be a small core group of users who will never switch. They just want a plain ‘o phone and that’s that! My Mom and my sister don’t want a smart phone and don’t want a data plan. There are users who have federally subsidized phones and aren’t qualified for a data plan. (Cell phone service is cheaper than land line now). This may be somewhere between 5% to 10% of the market. In that case, saturation is really at 90% usage.

    However, there’s also the possibility that carriers will simply force users to have data plans and smartphones. You already see this today. You go to Verizon or AT&T, and they only have Android and iPhones on display. If you ask for the cheap phone, the carrier give you some obsolete Android device and charge you for a data plan whether you want one or not. If you can no longer get non-data plans, smartphone market penetration will be 100%. That might be the case by the end of 2015.

    Which brings us to a very interesting question — How many people who have an Android handset are actually “smartphone” users? You might be looking at the wrong statistic. Many people who have an Android phone are not “smartphone” users. They got the default phone. Now, when they decide to get an actual smartphone, they are choosing to go with something like the Samsung Galaxy or the iPhone. This maybe why the Android share of the market is slowing and may even be shrinking. People with those cheap Android phones are just now deciding to enter the smartphone market.

    In that case, you need to look at the premium handset share to see the actual smartphone market penetration. See how many 4.3/4.4 versions of Android phone, iPhones, and Windows 8 phones are being sold. This is the group that’s conscientiously selecting a smartphone. That would be more interesting to see.

    • http://lucasarruda.com/ Lucas Arruda

      But, at some point, like in 5-6 years, maybe, all phones will be smartphones or smartphones-like. The entry lines will have maybe a simpler browser, lacking one or two features we have on top-notch phones today.

      I don’t see we having “brick” phones (those brick shaped nokia models) if even the most basic ones already have quite good color display and some smartphones features.

      Probably the worst mobile processor (aprox. 16 times faster on 2019 according to Moore law), will be lot faster than current top mobile processors and using much less energy. So those “dumb” phones that have them will probably be smartphones by there.

      Another point also, is that those old operating systems Nokia, Samsung, etc have will at some point no be commercially viable when Android/MS/etc scale and become wide-used and runs on each and every kind of phones (even the “slow” ones).

      • Walt French

        A basic featurephone for $50 can sell profitably in some parts of the world, so the question is what additional is required for a smartphone. Arguably, a much bigger screen, much more memory, a real CPU, OS, much more battery, stronger radios, additional licenses.

        Between technological advances and economies of scale, these have been getting cheaper by the day, but not exponentially so; I just can’t see how you’d establish a brand with features worse than today’s market leaders in any geography. Users would wait or make a stretch purchase.

        PS: Moore’s Law was about # of transistors, which are put to service of speed, but are not speed per se.

      • meliorist

        $50 is the price of a basic smartphone. A basic feature phone is more like $10.

      • Walt French

        Thanks.

        I’ll reiterate a note I made in an earlier post: In purchasing-equivalent income, 90% of Indians are below the mark of the bottom 10% of Americans. (And much of the world is less well off.) There are many, many millions of people in the world for whom a $50 computerphone is out of their reach, but a $10 talk+SMS phone is a doable stretch.

      • meliorist

        I don’t think that makes any difference. People in the third world have more reason to buy a smartphone than people in the developed world do since, for many of them, it would be their only computing device and their only internet access device. Practically everyone who can stretch to such a device will buy one.

      • http://lucasarruda.com/ Lucas Arruda

        That’s true. I live in a 3rd country world and I can pretty much say that every one that can buy a tablet/smartphone, will buy one, and many are buying one instead of a desktop/laptop.

        The number of Android phones compared to iPhones is huge. While on NYC subway you will see mostly iPhones and some Galaxys, on a bus in a development country you will see 1 or 2 iPhones and 20 Androids.

      • meliorist
      • TexanPatriot2

        With contract, right?

      • http://lucasarruda.com/ Lucas Arruda

        I think for that price you can already buy a basic smartphone, as @meliorist:disqus said.

        Some phones from China have 2 sims slot, TV, etc, have Android and you can buy for that price.

        It’s very hard to believe that Android and similar OSes will not dominate market, even in “low-end” phones which will me much more powerful in 5~6 years.

        P.S.: I know Moore’s Law is about transistors. But that can be extrapolated to speed approximately like I said on top.

        P.S. 2: Some recent data on iPhone – shown on the launch of the 5S – can support that:

      • Walt French

        Yes, I took that 40X number last night and calculated it’s an 85% per year speed increase over 6 years.

        I’m pretty sure that’s waaay faster than CPU capabilities in general; it reflects multiple cores which are not always usable and more silicon, smarter software. We shouldn’t extrapolate that rate.

      • http://lucasarruda.com/ Lucas Arruda

        I think until smartphones are faster enough to catch desktop architecture, that growth will sustain.

        It’s not only about cores. It’s about smaller and more transistors, which are faster and use less energy than before. If it were by cores, it could only have doubled at max, since iPhone still have “only” 2 (and that doesn’t guarantee double speed).

        Just compare a top phone from 2004 (9~10 years ago) which has 104Mhz processor (Nokia N-gage) and much older ARM 11 arch and today’s iPhone 5s.

        If you do this analogy with Desktop we can probably try to predict mobile. Even the cheapest Desktop you can buy (probably $299 nowadays – 2013) can run latest OS, so why that wouldn’t happen on mobile at some point?

      • charly

        The latest Samsung chromebook uses the same processor as their smart phones so i think they have catched up with the desktop

      • http://lucasarruda.com/ Lucas Arruda

        @charly yes, that’s true. For small tasks / basic usage and a stripped Linux, it works just fine with a mobile ARM processor.

        Imagine that in 5 to 6 years. It will be amazing the kind of computing power we will have at out hands.

        Batteries will pretty much be the limit.

      • charly

        But will you do anything but smaal task / basic usage on a phone?

      • http://lucasarruda.com/ Lucas Arruda

        You can definitely do. Like playing Infinity Blade II, which is a heavy game.

        But, I guess most people won’t and that’s exactly the point. And that is why I think most people will have basic smartphones instead of dumbphones in near future.

      • Hyptiotes

        Unless a smartphone can load 4500 10 MB photos to peruse and touch-up in photo editing software, I’d say smartphones aren’t anywhere near a desktop in function. Can they sort through one of my emails from 2000 out of the 100,000? Can they run statistics in the background while extrapolating 800 different dynamic state variables while building climate models? I like to make my computer CPU cry and I’m not even a gamer. A desktop is nothing like a smartphone. The first makes stuff. The second consumes it…if it doesn’t choke on the data stream.

  • Secular_Investor

    My key take away from these charts is that Apple still has huge headroom to grow market share and sales even in a mature, supposedly mature “saturated” market like the US.

    This is shown by Kantar’s survey of users actual buying decisions, which is far more reliable than the “shipping” numbers used by Strategy Analytics, IDC and others who rely on fantasy, channel stuffing numbers from the OEMs marketing departments.

    Kantar shows that the iPhone has GAINED market share Year On Year in the three months ending August 2013, in the US, UK, France, italy, Spain, Australia, Mexico and Japan.

    What makes this even more remarkable is that Apple has achieved these market share gains with record June quarter iPhone sales with the nearly year old iPhone 5, nealry 2 year old iPhone 4S and nearly 3 year old iPhone4 competing against Samsung’s latest Galaxy and other competitors latest models.

    Also very encouraging confirmation that the iPhone may accelerate its market share gains comes from Canaccord Genuity’s survey of the 4 largest US carriers, which shows that in just the last 10 selling days of September the iPhone 5S and 5C were the TWO top sellers and BOTH OUTSOLD the Galaxy S4 which was for sale the entire 30 days of September.

    • bloftus

      Early adopters are mostly current users and last year market share movements did not occur until the new year after there was supply/demand balance. Most non-iOS users will want feel the phone before buying. Also a lot of androids were bought for Christmas 2011 and will not be off contract until January.

  • Walt French

    Historically, this type of saturation analysis applies to moves that take place over many years, even decades. The little window of market-share data is based on an incredibly compressed set of technological changes and market responses.

    One feature that some analysts (e.g., stratechery.com) are watching is the Job To Be Done. Apple is pushing the envelope pretty hard here, what with its 40X speedups, I don’t think many Americans would even use an original iPhone if it were given to them.

    Thanks to economies of scale and technological advances, the cost of a phone stays fixed. While I expect advances to continue, increasingly those advances will target brand new features — authentication/identity and location are the two standouts in iPhone5s — more than raw power. (Although my 5 no longer seems amazingly fast; LTE must be a lot slower these days.)

    If these moves are smart (eg, Apple eschewing Blu-Ray and NFC as ahead of their time / not right for when the time came), the iPhone X of 2018 (assuming Apple will still be making an iPhone) will barely resemble today’s devices. I really can’t think of the analogy, but think that the ubiquity of friendly data devices/PDAs/pocket PCs/phones will hasten their own death.

    Exciting times.

    • Chaka10

      “LTE must be a lot slower these days.” I’ve noticed that too. I initially noticed it on ATT, so I moved to VZ for my iPad, but now that’s worse than ATT. I take this as the result of usage and congestion — I.e., real life experience with mobile broadband being a scarce resource, which supports its hefty pricing.

  • Gene Grush

    Interesting article on market penetration, but have you ever address turnover rates and how they will change in the future. As an Apple investor that is one of my biggest worries going forward and how will affect the revenue. To date I have held onto my IPhone for 1 to 2 years on average. Currently, I plan to hold onto My IPhone 5 for two years. Going forward, will people hold onto their phones for 2-5 years or will it stay at 1-2 years. Mobile phones are so heavily used and abused that the turn over may stay in the 1-2 years for some time, but eventually it will go to 2-5 years or higher. Do you have any data to expand on this and what would the affect be on the expected yearly sell rates?

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

      Since mobile phones have been invented the turnover has remained pretty constant at 2 years. Phones physically wear out, are lost, break, screen scratch and batteries wear out. I don’t see that changing. It’s the nature of personal technology that it wears out. You need to think of them more like items of apparel that are worn every day than as a computer.

      • Gene Grush

        I agree that the lost/break/screen scratch/batteries may keep the usage rate at 2 years or last for the foreseeable future, but Apple’s business model is to build a high quality product that last a long time. I now keep my computers 4-6 years if not longer, but before I was replacing them every 2-3 years. Doing some Web search, americans seem to hold there cell phones for 16-24 months on average. If this was to go to 30 months, that would drop the buy rates by 50% or more. Some countries the usage rate was 5 years or greater. If the hardware additions become less compelling and the cell companies start giving lower rates after the two year contract, people will start holding their phones longer. This is probably 5 years out, but it should start affecting the purchase rates within the next 5 years. Everyone assumes that the PCs are selling less because of the tablets, but how much of the drop in sales is people holding onto their PCs longer.

      • TexanPatriot2

        I get rid of my phones and other devices right before the resale prices is less than half. That way it partially funds the replacement. No sense in keeping them longer. I’ve sold phones which paid for up to 75% of the replacement phone.

        By the way, I’m with a carrier without a phone subsidy. (T-mobile)

      • TexanPatriot2

        It’s a wearable computer. It’s more powerful than 10,000 Commodore 64s my iPhone 5S.

  • StevenDrost

    I’m not so sure the future will lead to anywhere near 100% penetration of smartphones. The smartphone when created changed the device from a phone to more of a data device and that trend is like to continue. Voice is only a small part of what most people do on their phones and over half the bill. Its used less and less everyday and the function can be replace with a variety of VOIP services and messaging Apps. Seeing as most people have WiFi at home this function could be replaced for free and data service extends the capabilities everywhere.

    The dynamics of the U.S. market are such smartphones are not offered as data only devices, though it is possible. That leads to the potential of the tablet stealing market share from the smartphone, especially for the budget minded consumers. There are plenty of “pocket-able” tablets, like the nexus 7 or the IPad Mini. These could fill the voice role, though not as well, as headsets or speaker phone would be required or risk looking like you have an 80’s boombox next to your ear.

    The smartphone movement may very well become the smartdevice movement and smartphones may actually see reduced market share.

  • RichardinMelbourne

    I suspect total smartphone penetration may top out around 75% (see Bass 1969 for evidence that 100% total market size as an end point is not a sure thing), leaving a bunch of people who will never have smartphones. If 2013 is 61%, as Horace reports ( 2012 50%, 2011 36%) then it won’t be long to see who is right…

    If smartphone users pass their old phones to non-users, that could be a way to penetrate the last 25%. I have a hand-me-down iPhone 3GS from my wife, for instance. I have used it for 2 1/2 years now… But my 80+ year old Dad has a feature phone. If my wife upgrades from iPhone 4S to 5S or 6, then I might get her 4S and my Dad might get my 3GS. Maybe… But there is a reasonable chance he won’t want it… And my wife is holding off upgrading because she finds IOS7 hard to read… (sigh)…

  • TexanPatriot2

    Eventually EVERY phone sold will be a smart phone. Once that number hits 95%, you can pretty much say that.

    Even the “non-smartphones” are approaching smartphones in capability with texts, media, and web features.

    There are few simple “just phones” left. Most of them are “given away” as Obamaphones by TracPhone and others, many are bought by the elderly or parents not wanting their kids to have that kind of phone.

    Remember the Firefly phone? It only had buttons to call Mom and Dad and maybe 2 others. That’s long gone now. Those same kids have smartphones.

  • Dieter Moreno

    I don’t think smart phone penetration will ever reach 100%. I have no intention of ever upgrading to a smart phone because I just don’t see the need when I already have desktop computers and laptop computers in my house. I suspect that even though I’m in the minority, there will be others like me.

    Well at least, not upgrading to a smart phone BY CHOICE.

    There may come a time when carriers only sell 4G phones and those 4G phones are only smartphones, and then when they shut down the 2G and 3G networks, us non-smartphone users will have to choose between either giving in to conform or giving up and not using a cell phone at all and bringing back home phones.

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

      You won’t have a choice. There will be no incentives to continue making dumb phones. The cost of making non-smartphones is likely to soon be higher than making smartphones. See also black-and-white TVs and CRT TVs. The older technology, even though “good enough” for many was simply not economical as scale of the new technology swamped it. The phone market, unlike other appliances and consumer electronics has a far shorter life span due to battery degradation and mechanical wear and tear. They are effectively consumables and therefore they will need to be replaced.

      • Dieter Moreno

        Interesting comparison. History has had a tendency to repeat itself.

        We can look at color TV adoption, which as of 1999 in the chart shown reached 98% adoption.

        http://www.edn.com/electronics-news/4362842/Lessons-from-history-The-adoption-of-color-TV

        Looking at the Diffusion of innovations Wikipedia…

        http://en.wikipedia.org/wiki/Diffusion_of_innovations

        1954-1965: Innovators adopt color TV

        1965-1970: Early adopters adopt color TV

        1970: Color TV reaches the Early Majority

        1970-1975: Early Majority adopts color TV

        1975: Color TV reaches the Late Majority

        1975-1980: Late Majority adopt color TV

        1980: Color TV adoption reaches the Laggard adopter stage

        1980-1999: Laggard adopter stage

        1999: 98% color TV adoption

        Color TV was designed to be backwards compatible with Black & White TV, but LTE is not backwards compatible with GSM or 1xRTT CDMA.

        So the laggard adoption stage for smart phone adoption is going to be much shorter than the laggard adoption stage for color TV adoption.

        Likely, approaching the 2G and 3G shutdown we will reach the Leapfrogger stage.

        “The phenomenon when resistors upgrade they will often need to skip several generations in order to reach the most recent technologies.”

        I think we kind of had the Leapfrogger stage when analog TV was shutdown in 2009, as most people who were still using a CRT upgraded to a flat screen TV just to continue to watch OTA TV.

        However, the Leapfrogger stage was not as big with OTA TV as it will be with smart phones, because you can use a converter box.

        I indeed am using a converter box right now to watch TV on a CRT.

        So I suppose I am a laggard for OTA TV who never was forced to leapfrog.

        You will not be able to use a converter box for cell phones.

      • Merckel

        Hardware innovation that brought Plasma and LCD panels stimulated new demand well past the adoption of Color TVs, so saturation is a bit illusory in the context of potential sales growth.

    • Jason Smith

      So you have a smart phone, but it’s a bottom of the range crappy smart phone… If your phone has a camera, touch screen, web browser, qwerty keyboard…