Measuring US Mobile Platform Shares: Kantar vs. comScore

The latest comScore US smartphone install base data is in and there are few surprises. iPhone has reached a new record high penetration (39.2%) and user base (54.3 million). Android has reached a new high in user base (72 million) but share at 52% is below the peak reached in November 2012.

Screen Shot 2013-06-05 at 6-5-3.48.35 PM

This pattern of gradual iPhone share gain in the US has been consistent for over two years even while Android has catapulted into an overall lead. The surprising thing is how Android seems to have peaked in share. There are still 95 million non-smartphone users and there seems to be headroom for growth even though the other platforms have been tapped out. But it does not seem that Android phones have any particular advantage over iPhone. My hypothesis remains that as price is taken out as a differentiation, the adoption of iOS is slightly higher than Android.

Another measure of market performance is the implied net platform user gains which is shown below:

Screen Shot 2013-06-05 at 6-5-3.23.55 PM

It shows that iOS added more users in the last few months than Android.

The problem is that Kantar Worldpanel measures shipments and their share data shows a seemingly different picture.

Screen Shot 2013-06-05 at 6-5-3.22.23 PM

In their data Android is shown as selling more units during January through April while Apple sold more during October through December. Of course we don’t have the absolute number of units so can’t see the effect of higher holiday overall sales volume. Nevertheless, the balance of growth seems to be disproportionately in favor of Android relative to the data from comScore.

To look at the situation more closely I measured the differential in user adds for comScore and the differential in market share for Kantar’s data. Then I overlaid the two differentials so that months are matching, as shown below:

Screen Shot 2013-06-05 at 6-5-3.51.57 PM

There is a similarity to the frequency of oscillation with comScore data showing a delay (as would be expected since their data is sampling over a three month period). However, we are still facing a vertical offset where there is apparently more growth bias for iPhone in the comScore data.

Possible factors which might be explanatory:

  • Methodologies used. comScore data excludes ages below 13 and non-personal devices (business expensed phones.) 
  • Replacement sales are invisible in comScore data since the user base does not change when phones are replaced (and old ones are discarded.)
  • Missing data from Kantar. Given their survey methods it’s possible that their panels miss some market segments.
  • obarthelemy

    Can’t you, under the protection of an NDA, get a few broad-based neutral sites (amazon, CNN, TMZ…) to give you browser stats ? These seem in the end much more reliable than obscure stats from providers with an agenda ? If not for actual levels (there’s bound to be a bit of bias) at least for trends….
    I’m not sure why sites keep that info close to their chests. Les Numeriques in France publish it monthly, it makes for a fun thread :-p

    • I believe in open data and refuse to look at non-public information. The data is there but if it’s not shared it’s of little use to the world. But it’s valuable to some. Google built a business on closed data.

      • obarthelemy

        défine “open” ? Unless the two outfits you use disclose their methodology including technical details, sampling techniques, and sample sessions, there’s nothing “open” about their data. Public maybe, open certainly not. And unauditable, which is always stinky. I sometimes get into panels and surveys just to get a view from the other side…. there are horrendous methodological, logistical, training, bias,… issues on all surveys I’ve been subjected to.
        I’ll take one large site’s raw data over a mysteriously compiled monstrosity anytime.

      • Bruce_Mc

        Are you suggesting that data collection practices should be audited before data is used at Asymco? Can you name any publishers of data that would allow this?

        I think it is well understand that the companies publishing this kind of data are not trying to advance the sum total of human knowledge or win the Nobel prize; they are trying to make a buck. So take the data with a grain or two of salt.

        Using publicly available data means that the work done on this website can be checked, it means that other people can come up with their own interpretations of the data. It attracts people who want to learn to the website. Using private data attracts people looking for an investment edge: hustlers and fools.

    • Mark Jones

      “I’m not sure why sites keep that info close to their chests.”
      They sell the data for big bucks to carriers, vendors, analysts, etc. The public data is to garner industry awareness/prestige in order to drive those sales.

    • I find StatCounter to be the best source of this type of data and it is published under Creative Commons. It is based on an aggregate of 3 million+ web pages but they do not disclose their users. I suspect it is million of smaller sites VS really big ones but it is still drawing from a very very large sample.

      What is nice is the amount of control you have over the query. You can control country/region/resolution/mobile/OS/MobileOS/desktop/vendor/date (yearly, quarterly, monthly, weekly, daily) and many others. You can export the data to CSV for off-line work. NOTE: Screen Resolution queries can take some time and will frequently time out.

      Is an example of looking at the US Web Usage of the iPhone 5 VS the HTC One/Galaxy S4. Because each group had a unique screen resolution, it was possible to tweak out the data. The iPhone 5 has the tall 320X568 while the other two are the first generally available 1080P phones available in the US.

    • One problem with relying on browser stats to measure smartphone share/usage is that almost all provide a total iOS number which combines iPads and iPhones. Yes, you can try to tease out some of the difference with screen resolutions, as another commenter did, but that gives you only a small segment of the iPhone population.

      • StatCounter does not group iPads with iPhones/iPods Touches. The iPad is listed separately within the “DeskTop OS” option. Using screen resolution is nice because the iPhone is almost unique in using 320X480 and is fully unique in the 320X568 resolutions in the mobile space. There may be a small number of nock-offs but those numbers are very small.

      • JohnDoey

        When all Android is combined (even different manufacturers, even feature phones) why would you want to remove iPad from iOS numbers?

        Android comes on various screen sizes — so does iOS.

  • Mark Jones

    Kantar says this about their data: “The data is derived from Kantar Worldpanel ComTech USA’s consumer panel, which is the largest continuous consumer research mobile phone panel of its kind in the world, conducting more than 240,000 interviews per year in the U.S. alone. ComTech tracks mobile phone behavior and the customer journey, including purchasing of phones, mobile phone bills/airtime, and source of purchase and phone usage. This data is exclusively focused on the sales within this 3 month period rather than market share figures.”

    So Kantar claims their data is sales, not shipments; and taken also over a 3 month period (just like Comscore). I can’t tell how many of the panel is polled about purchases every quarter; maybe it’s everyone but maybe it’s just a third. The interviews could be for gathering a deeper set of data.

    • Chaka10

      Sales can mean sell in (to channel inventory) or sell through (to end customer). Usage is a bit loose, but I think the former is what’s meant by “shipment”.

      • Mark Jones

        Agreed, but Kantar is using a consumer panel, who would reveal nothing about sell in.

  • Shipments of iPhones fall quickly before a new model launch, see Kantar share data around july 2012, while sales stay higher due to inventory sales, see Comscore data in the same period.

    The latest data show the begin of the same trend, shipments going down while users are going up.

    Perhaps it is a sign of a stop in shipments by apple in view of the launch of the next iPhone?
    In 2012 the direction split between yellow and blue lines span about three months, september will be the launch month this year? It seems so.

    Horace do you how many months of sales are stuffed in channels for iPhone? Perhaps it matches the three months decrease in shipments of 2012.

    • Mark Jones

      Apple reported channel inventory of 11.6m iPhones at end of Mar 2013, and 10.6m at end of Dec 2012, both of which are within their 4-6 week target. It was 9.1m at end of Sep 2012 (right after iPhone 5 intro), and that was short of their 4-6 week target.

      I wouldn’t consider any of those numbers as anywhere near “stuffed”, especially since we expect iPhone 5 and maybe 4S to continue to be sold after new iPhone(s) are introduced.

      • Yes, but a decrease in shipments could be done just lowering the 4-6 weeks span to say 1-2 weeks, this can be done shipping less phones every month, for 5 and 4s models, while the shipments for the 4 model are reduced progressively to 0, always progressively lowering each month shipments, for three months.

  • Mark Jones

    Also, thanks for attempting the compare. I tend to find these data sets don’t fit well with Apple’s and the carriers reported data, thus limiting their usefulness to qoq or yoy compares within each data set. For example, I think comscore overstates repeat iOS purchases, but couldn’t gather enough data from other sources to definitively show it.

    Kantar also provides US data on big 4 carrier share, and platform share for Verizon/AT&T. In the text and another article I saw yesterday, Kantar also said new iOS purchases over Feb-Apr consist of the following: 31% from featurephone (it was 9% a year ago), 20% from Android, and 38% from iPhone. New WP purchases (a much smaller volume) consist of the following: 42% from featurephone, 25% from Android, and 25% from Windows.

  • bobfet1

    If iOS users are more engaged with their devices and user them more/visit more sites, would that account for the difference? If comScore’s data is based on usage, perhaps that’s why iOS looks better than Kantar, which might show if Android devices are selling but not being used as heavily.

  • Chaka10

    It seems the oscillation (overlay) lines up better if the Kantar graph is shifted to the right by 2 months – I gather that’s what you mean by the comScore data showing a delay. Would seem to me that’s what you might expect if Kantar data were showing sell-in (to channel inventory) and comScore showing sell-through (roughly).

  • Walt French

    Is anybody able to triangulate between this data and the just-updated info from Good Technologies about their Enterprise support software, which shows iOS capturing approximately 3/4 of users who register w the Enterprise?

    My hypothesis is that WP is primarily going into Enterprise situations, but that even there, in part thanks to BYOD, iOS is dominant; the fraction of devices that are somehow tied to non-end-user situations would be insightful in looking at future shares of smart devices.

    • JohnDoey

      iOS in enterprise is not just BYOD. Apps are really important there and Apple is so far ahead in apps. I worked for a company who did their first iOS app and it cost less to build and deploy than their Web apps and they started pushing iOS standardization across their userbase. Also they were really unhappy with user training costs on app platforms except iOS. And really unhappy with anti-virus costs on all platforms except iOS.

    • Secular_Investor

      We can triangulate this: Samsung spend 10 times as much as Apple on advertising, marketing & promotion – Kantar is a subsidiary of WPP which is the ad agency for Samsung and other Android makers – Kantar’s data favours Android and Samsung

      We can also triangulate this: Android and Samsung don’t give sales numbers – analysts talk of fantasy, unverifiable “shipping” numbers which show Android ships more smartphones to the US than Apple – but verifiable numbers from AT&T, Verizon and Sprint show that the iPhone outsells Android.

  • bloftus

    Comscore is looking at net change in users. Therefore Comscore is influenced by what the user was using previously and whether their old phone went into hand me down status or into the trash. Apple’s first quarter acceleration was probably related to Android’s relatively high increase in new users from Christmas 2010 who were now off contract and eligible to switch to iPhone. Those that switch from Android to iPhone also depress the Comscore market share during Q1 resulting in a declining market share despite new smartphone users.

    Looking closely at the data ATT Q4 2010 – 7.4 million smartphones, 4.1 million iphones, How many of those 3.3 million non-iphone turned into iphones at upgrade time? I don’t know but it is most of them I think as ATT Q1 2013 – 6 million smartphones sold and 4.8 million were iPhones.

    Remember, Verizon Q4 2010 – 0 iphones, Q1 2011 – 7.2 million smartphones, 4 million iPhones.

  • kankerot

    Apples sharp and possible illegal business practices puts into perspective how subsidies are distorting the market. The end of subsidies would lead to lower prices for the consumer.

    • handleym

      You might want to be a little more skeptical when you read this sort of article.

      The REAL complaint here is that telcos, which have been doing this sort of thing for twenty year (forcing the phone companies to do their bidding through things like telco specific models, crippled features, or suppressing services) are now pissed off that Apple is doing the same thing back to them.

      Let’s lay off the “possibly illegal business practices” nonsense. If this were ever to go to court, Apple would have a grand time giving the details of one sordid story after another of how the telcos have done exactly the same thing to the phone companies…

      Just because someone whines in a newspaper doesn’t mean it’s actually true… Especially when the whine comes from “according to a former senior executive at a major European operator.”

      • kankerot

        You might want to lay off the Apple Kool Aid. The European Commission doesn’t just ask for information if they don’t think their is a possible issue. You can dismiss the issue without any further evidence – looks like you who needs to show some sense.

        Also mistakes in the past do not justify current mistakes.

      • Jessica Darko

        First off, consider the source. Your source is not reputable.

        Secondly, the European Commission is a joke. Worse, they are a criminal organization that punishes some businesses to the benefit of others and the consumers detriment.

        That they may be lining up apple for a shake down racket isn’t surprising. Nor is it damning to apple, on the contrary, means they’re doing things right.

      • kankerot

        Your deluded second paragraph is just laughable.

        Again no response to the issue of subsidies.

        I take it you were against the EU actions against Google and MS?

    • Joe Slow

      And how much does Samsung pay you?

    • Mark Jones

      Subsidies probably do distort the market, though the consumer could exploit the use of subsidies as well. Twice I took advantage when AT&T allowed early 12 month full-subsidy upgrades, and combined it with reselling my iPhone for a small profit.

      Subsidies were already commonplace before Apple entered the mobile market. Apple didn’t use subsidies for the original 2007 iPhone, but instead negotiated to receive a share of AT&T’s $20 data plan fee (similar, but not exactly the same as RIM). They subsequently reworked the Apple contract to use subsidies.

    • “according to a former senior executive at a major European operator” Unnamed.

      “If Apple is found to be a dominant player, this kind of preferential treatment could be challenged as breaching competition law.”

      We hear all the time how Apple is far from the dominate market player with Android having 75% market share. The advantage of not being the dominate market player is you have much more room to negotiate.

      “Proving Apple’s dominance could be difficult. The world’s most valuable technology company had a 22% share of the smartphone market in Europe last year, according to research firm IDC – a far cry from the 40% usually considered a threshold for dominance.”

      Oh, they got to that point in the article. Good. Of course you could define the market as smartphones with touch interfaces costing over $600 or EVEN better, smartphones running iOS. By George, we got it. Apple is dominate in the iPhone space!!!

      “Apple’s terms were not illegal individually, the operator source said. ‘If you looked at them one by one they were OK, but if you consider the overall impact of these provisions they could hinder competition. This most favoured position they had in the contract was very useful for them. They stalled competition in the market.'”

      Or you could look at smartphone designs BEFORE the iPhone and AFTER the iPhone and see what the iPhone did to spur competition. Nahh… That would be logical and not give the EU a way to crete fines for the sake of it.

      So, to recap: Apple is not a dominate player. Subsidies are not illegal. Apple, being a minority player, wants to hinder the competitive advantage of its competition. That is the definition of competition BTW. Apple’s terms are not illegal. Apple’s competitors don’t like Apple’s contracts (but that is not illegal).

    • Jessica Darko

      You seem to forget that Apple tried to end the subsidy business model with the original iphone, and people attacked it as being too expensive. Said there was “no chance” they could get significant market share. So they went to the subsidy model.

      Whether something is illegal on the socialist continent increasingly has little to do with whether it is moral or not.

    • JaneDoe12

      Subsidies probably do distort markets, but in the case of smartphones I think it’s beneficial. It allows more people to buy them, and the phones enhance their lives.

      Do you remember the following comment from Horace? (Link here) The subject was luxury cars.

      “The capitalist achievement does not typically consist in providing more silk stockings for queens but in bringing them within the reach of factory girls in return for steadily decreasing amounts of effort.”
      Joseph Schumpeter, Capitalism, Socialism and Democracy (1950, 3rd ed.), part II, chapter VII, p. 82.

      I would say it as, any fool can sell luxury to the rich. Finding a way to sell it to the poor requires genius.

      I think this applies to iPhones because they’re a luxury.

      The end of subsidies would lead to lower prices for the consumer.

      I’m not sure the manufacturers would lower smartphone prices below the 2-year contract price.

  • Jessica Darko

    Are any of these android numbers from directly measured sales? We’re talking about statistics here and methodologies are quite important. Otherwise you could draw completely wrong conclusions.

    Chikita web request stats may not be a perfect measure of market share, for instance, but we know the methodology behind them.

  • Anon

    Folks, whats up with Phil Schiller referring to the “designed by engineers” in CA? *twice* today.

  • Pingback: SmartPhones or Smart Phones - or just Smart Usage - Beyond Bridges()

  • Pingback: Can Google Become a Force in Enterprise Computing? »()