Categories

Nokia employs as many people to develop its smartphone software as Apple does to develop all its products

In a recent post I pointed out that Apple’s R&D was about 2.2% of sales in the last quarter. Bernstein took a look at the R&D for Nokia and presented a chart showing the difference between the mobile industry players in terms of total expenditure on R&D.

I took inspiration from that to plot the Devices R&D for both Nokia and Apple over the entire 2010 period. I also compared that with sales and computed the ratio between R&D and sales.

The result is shown in the chart on the left.

Bottom line: Nokia spent 10.2% of phone sales in 2010 on phone R&D while Apple spent 2.5%.

Bernstein goes on to argue that at least for Devices,

Nokia spent $3.9bn in R&D in 2010, almost 3x the average of its peers, 31% of the industry’s R&D total spending, for an output that we can qualify as visibly disappointing.

To relate the $3.9 billion for Devices into head count, they estimate that Symbian projects employ 6,200 people; MeeGo and Qt 1,800; Services 1,800; and S40 1,800. Hardware headcount is assumed to be 4,700 and 900 more for fundamental research.

So Nokia’s total software headcount adds up to 11,600 people. Nokia smartphone headcount adds up to about 8,000.

Applying a similar formula ($240k/employee[1]) to Apple’s estimated iPhone R&D (from Bernstein’s chart) yields a headcount of about 3,200 and a total Apple company R&D headcount of 8,200.

So Nokia employs about the same number of engineers[2] for its smartphone software platforms as Apple does for all its product lines[3].

In fact, Symbian alone may cost twice as much to develop than the iPhone (including the hardware).

Notes:

  1. The cost per employee is obviously different between the companies as they operate in different countries. However there are enough sources of error in these estimates that they probably overwhelm the cost of living/tax differentials.
  2. I know that engineers are not all that R&D headcount covers, but let’s assume that the mix of overhead is similar for the two companies.
  3. The total for Symbian + MeeGo is 8000 and the total for Apple overall is 8,200, however one can comfortably allocate 200 out of 900 researchers out of Nokia research to their mobile software effort.
  • Pingback: Nokia vs. Apple vs. Facebook: On Bloat and Leverage | Technoagita

  • MattRichman

    It's amazing how efficient Apple is in absolutely everything they do.

    • dchu220

      They are even efficient at getting me to open up my wallet. I'm a notorious late adopter who lined up for the iPad on the first day.

    • Evan

      not in everything, their Apple TV is only a modest success, their social efforts are a waste of time and their cloud efforts are pretty poor especially mobileme. Might change in the future.

      • asymco

        Up until it's a success, everything starts out as a waste of time.

      • Evan

        we will see, not everything a company starts succeeds. Social networking/cloud computing requires different sorts of skills/mindset/DNA than selling the best user-experience packaged with the best software/hardware integration. Of course Apple could simply buy up twitter or something, but I doubt any innovative company is up for sale. Social networking/cloud computing is device agnostic. And besides the rivals in cloud computing, social networking are extremely innovative unlike Apple's rivals in consumer electronics Sony and Nokia which were very poor

      • Coward_the_Anonymous

        Social networking/cloud computing brings you poor revenue until you turn them into adware. So far. Apple didn't focus then in that area, but they keep hand on pulse. They' ve got even their own ad platform, so they probably plan to use it for ad-cloud (me.com) when it matures and proper time come.
        As for AppleTV. It wil work out if deals with content providers made, it concerns iTunes in cloud. It seems the media companies intentionally slow Apple as they got scared almost monopoly in mp3 distribution.
        Apple is poor when executes in environment it does not have ful control. Any way i don't think cloud is what they WANT to focus on. It probably gets its place similar to iTunes, add-on (and teaser) to hardware sold.

      • Synth

        "their Apple TV is only a modest success, their social efforts are a waste of time and their cloud efforts are pretty poor especially mobileme"
        Except that Apple, unlike Nokia, didn't waste billions on any of those efforts. All those initiatives simply reuse or leverage existing Apple technologies and software so minimal investment and R&D is involved. Yet they all serve various strategic or potentially strategic markets for Apple

      • Evan

        as I said, one has to look at opportunity cost, In cloud, microsoft is dominant, google is pretty good and in social facebook is dominant. The more they delay they more entrenched microsoft, google and facebook become. I am sure building up a social site is easy, but building up a 600 million userbase is difficult. You have to show a different form of social networking to attract away facebook users for instance. And facebook unlike Sony, Motorola, Nokia is very innovative company.

      • http://twitter.com/aegisdesign @aegisdesign

        I would think challenging Facebook directly is impossible today. You need a completely different model to challenge them. Maybe something like Disapora – https://joindiaspora.com/ which distributes social networking across many servers owned by individuals, not one company.

      • gslusher

        We know that Apple sold well over 1 million Apple TVs in the last quarter–actually, in less than one quarter, as the Apple TV didn't go on sale until late September (I can't find an exact date) and reached the 1 million mark on Dec 27th. To put this into context, Apple sold 1.23 million Mac desktops in that quarter. If one extrapolates the 1 million Apple TVs in about 2 months to a full quarter, the Apple TV is about as successful as Mac desktops.

  • Martin

    One new phone per year versus, what, 30? More? Annual refresh cycles are efficient. Only one production line setup, massive, massive economies of scale on components, one design, one testing cycle, and fewer variables for 3rd party developers to deal with.

    • r00tabega

      For 3rd parties, the sales potential (ie, millions of one unit vs. hundreds of thousands of dozens of units) provides a large addressable market for add-ons (like cases, headphones, etc).

      This provides a virtuous cycle for the product, it's accessories and the users.

  • kevin

    Because Apple only introduces one model per year aimed at the top end of the smartphone market, it can focus all its R&D energy on the technologies that matter in making the best even better, rather than also on making the average mid-tier phone a little bit more different. Plus, it very much leverages technologies (hardware and software) across all its product lines, including even the Mac.

    • asymco

      Well, perhaps. But note that Symbian is a platform product. It's one product. Symbian development is twice as expensive as iPhone development. iPhone is one product and it includes hardware, platform support and software.

      So saying Nokia has 30 phones does not answer the question of why their one OS is double or triple the cost to develop as iPhone's one OS.

      • Pieter

        The core of iOS is shared with Mac OS X, while both OS'es make use of open source packages, how much is saved by that? (Or, what part of 'Other Products' in the graph is actually helping iOS?)

        I have no idea if Symbian makes use of open source packages…
        (I know Symbian itself is open source, but that will not save money if the code is only (or mainly) developed by Nokia and not 'eyeballed' as in Linux).

      • Sander van der Wal

        Mac OS' low level stuff is also open source (Darwin), but Apple is also paying for that development.

        There are indeed a number of OS packages in Mac OS, but if those packages are not on Symbian OS, it makes it only worse for Symbian OS.

        I think that Symbian OS needs more people to work on for a number of reasons:

        1) The OS itself is harder to program, as it is more optimized for less powerful hardware
        2) The middleware is complicated for the same reason, making it also harder to maintain.
        3) The UI layer is very bloated. Basically there are two UI's, the generic old psion-based one and S60. The Psion based stuff is almost never used but has to be maintained too.
        4) There is a lot of stuff in the UI to help customize it for operators.
        5) The UI is very hard to program for as it's API is very complex. Nokia's own apps are therefore harder to maintain.
        6) Symbian has it's own unique toolchain. That costs too.
        7) I suspect Apple's developers to be more experienced and better designers and programmers. Look at the API differences between S60 and UIKit.

      • http://www.sunsetlakesoftware.com Brad Larson

        While the leveraging of open source packages in a few places is of benefit, I believe that most of the efficiency Apple has in developing iOS is due to the elegance of the underlying Cocoa frameworks. Much of the core structure of these frameworks was put in place at NeXT in the early 1990's. From this solid foundation, they were able to build software that has targeted a diverse group of computing devices.

        In fact, I was stunned when I watched someone recently do a side-by-side comparison of developing for the NeXT and for the iPad and saw that the process was almost identical. This is 20-year-old design we're talking about at the core of leading-edge devices. Horace was right when he said earlier that Apple has been positioning their chess pieces for a while now.

        I'm a Cocoa developer for this reason: the frameworks are well-architected, easy to understand, and they let you build quality products quickly. I've had more fun on this platform than any other I've worked with, which is hard to explain to others when they speculate why Android or Symbian don't have more developers given their market share. If Android provided a better developer experience, I'd switch over to that, but it doesn't come close right now.

      • Evan

        nobody buys apps on symbian smartphones they were never positoined as appphones to begin with. As for android, their sales are exploding only now and there is fragmentation and nobody seems to pay yet for android apps, although I heard free apps download numbers are now comparable to iOS free apps download numbers. Yet things can change and will change and are changing, google now has an improved android market and no doubt will improve further. And as a developer, you should think of the opportunity cost, if you don't develop for android, somebody else will and make money. Android is now big enough. There are 5 million plus java developers and they need something to monetize their skills now that java has stopped growing in enterprise and android fits the bill. Java is a far easier language to learn than objective C.

      • http://twitter.com/aegisdesign @aegisdesign

        Nobody buys apps on symbian?

        Ovi store current download rate is just over 4 million downloads a day.

      • Evan

        cool, not 'nobody' but I guess 'most of people don't', I guess Apple gets 200 per second downloads, which translates to around 17.28 million per day with vastly lesser number of smartphones than nokia smartphones(installed base) also android gets around 30 percent of apple downloads, so android alone accounts for lets say equal to symbian smartphones.

        PS: Flurry estimates 60 million android devices in existence as per end of dec, http://blog.flurry.com/?Tag=Android

      • O.C.

        Why are people always talking about apps as thou they are the reason you buy a smartphone. A recent study pointed out that over a quarter of apps get started just once and never again. And a big chunk of the rest barely gets used too. The amount of useful apps – apps people really need and use a lot – is nowhere near the 300.000 plus. So if an appstore contains 1000 or 300.000 doesn't really matter considering most are useless anyway.

      • asymco

        Why are people always talking about TV programs as though they are the reason you buy a TV? Almost all TV shows are watched only once and never again. Many of the shows available to watch are not watched at all. So if a broadcast medium has thousands of programs, it doesn't matter. Most are useless anyway. I really don't get the point of TV.

      • Evan

        oh yeah, TV is dying out. especially among youngsters. TV like newspapers is on its way out.
        http://www.bnet.com/blog/high-tech/subscription-t

      • follower

        Posted on Super Bowl Sunday, no less.

      • asymco

        Not quite. Consumption of TV is increasing. It's broadcast business models are dying. To use your newspaper analogy, reading of news is increasing, but the selling of papers is declining.

      • Evan

        I suspect newspapers as a concept will die out both online and offline, simply because it is not personalized, it is information being stuffed from the top into our heads, if I want to read stuff about apple, I just read asymco or gruber blogs for instance, if I want to read some stuff about google, I read something else, if I want to read about VCs, I just subscribe to avc.com RSS feed for instance. Who wants to read what folks at newscorp think about where the world is headed, why should I be interested in Rupert Murdoch's world views? I don't want to read a slanted view, I don't want leftist, rightist or centrist interpretation of events as dictated by the board, I want events as it is, raw, I don't want to listen to the editorial board of newscorp(news corp is just an example), who are they really ? Egypt has shown how powerful the new social media is and how inadequate the newspapers are in reporting things, newspapers are trying desperately to push the agenda of neocons by telling us how Hosni Mubarak is good for egypt and the world because he is a moderate, because if he goes, egypt will come under the spell of muslim brotherhood which is a threat to world peace and Israel security. Newspapers have lost it, they are no longer the moral guardians/watchdog of this generation and by next generation they will be gone(just like email and TV subscriptions)

      • O.C.

        People were buying smartphones before there were apps. People weren't buying TV's before there were shows to watch.

        Whereas a smartphone without apps is still very useful. A TV without a broadcast medium or dvd's to watch shows on is pretty much useless!

        So there is a big difference.

      • http://twitter.com/aegisdesign @aegisdesign

        From what I've heard, Nokia develops the hardware and ports Symbian to each handset. There's effectively a version of Symbian for every handset model.

        Apple on the other hand seems to pull developers off one project to work on another which was why OSX 10.5 was late (Apple pulled engineers off OSX to develop iOS). 10.5 was a bit of a disaster. It was Apple's Vista IMHO. It's why they then released 10.6 supposedly as a fix for what went on in 10.5. 10.6 is a bit of a disaster too with some tech that was in 10.5 and prior chucked out. It's evident in their non-OS software too. iWork for Mac delayed because of an iPad port.

        So, IMHO Nokia is spending too much having duplicate teams duplicating effort on each handset but never having enough responsibility to push the platform forward as a whole.

        Apple is not spending enough, resulting in half finished software, missing features or long gaps between releases as the developers have been pulled off to another project.

      • O.C.

        Half finished software you pay a premium for to use.

  • Ardy

    Power of a strong platform strategy

    • Synth

      Also the power of NOT having 200 different devices and multiple OSes to power said devices.

      Cnet currently lists over 200 Nokia cell phones. That is simply ridiculous.

      • asymco

        Here's the question then: How many Samsung devices are in the market. My bet, a lot more. There was a time when Samsung launched over 100 new phones a year and were proud of the fact, aiming to do even more. There was also a time when Samsung had at least five different mobile OS platforms implemented in products in the market simultaneously.

        Samsung is being celebrated nowadays as a "winner".

      • Evan

        currently they have bada, android, wp7, they only let go of symbian because it was not selling well. I remember they had some LIMO phones as well lol

      • asymco

        Don't forget PalmOS.

      • unhinged

        The multiple OS argument may hold, but not the number of devices – unless somebody _really_ messed up on the design.

        Mind you, that's entirely possible with companies where hardware, software and research are all seen as somehow separate entities that should have firewalls between them. One of the great strengths of Apple is that their design is informed by what the various teams bring to the table, allowing better trade-off decisions to be made.

  • Michael

    So, effective at levarging their eco-system, partners and suppliers cmp with Nokia + more focused?

  • Pingback: Korta klipp – 04 Februari 2011

  • Evan

    the reason why android and iOS development is cheaper than symbian is because of unix and linux which are available under open source license and on top of which iOS and Android have been developer. What we are witnessing is just the battle between linux and unix which played out in enterprise servers more or less. It is open source 'factor' which has enabled these companies to innovate fast. This is hidden to all the financial analysts who simply go around blabbering that Nokia is less efficient. Open source is a huge disruptor, all of google's servers in their huge data centers run on a custom version of linux for example. It is unacknowledged by everybody in the mainstream press who have nothing but contempt for open source software and open source is going to be the secret sauce of the downfall of the microsoft empire. I wonder how much time microsoft would have saved, if they had gone in with linux or unix based OS and not been so late to the phone OS party. Blackberry also seems to have picked up it game via its QNX which is again based on linux. It is telling that google gives away OS for free, but charges the OEM for inclusion of android market. OS is simply a commodity. HP's webOS is also based on linux. If only Nokia had pushed meego faster and abandoned symbian faster.

    • Evan

      open source and of course Steve Jobs magic :) is responsible for Apple's success and it also telling that Larry Page and Sergey Brin were mentored by Steve Jobs

    • asymco

      The value of Unix is not lost on readers of this blog. http://www.asymco.com/2010/09/29/unixs-revenge/

      • Sander van der Wal

        Having programming for Symbian for 13 years, and 3 years for iOS, I do not believe it is due to the Unix bit that makes iOS easy to program and Symbian hard to program. In my experience it is the UI layer that determines much of the speed of programming. Symbian has much unix-like API's for unix-like stuff as is possible in C++, and iOS recommended layer of programming is two levels up, not at the Posix level, but at the Core Foundation level.

        On Symbian the UI layer is much harder to program for than the iOS UI layer. Especially S60 is very hard to program efficiently for. Partly because there always seem to be multiple ways to implement something, only to find out it won't work as it should so you have to implement it in one of the other ways. And partly because the Symbian UI layer is at a much lower level than UIKit, so you need to write more code go get the same effect.

      • http://twitter.com/aegisdesign @aegisdesign

        Nokia have addressed that however. Qt (in particular QML) makes it much much easier to do the UI work than even iOS. The old AVKON days are gone mostly.

    • SteveT

      Neither QNX nor iOS is based on unix or linux, iOS (and OSX) is based on Mach with a BSD compatibility layer providing the 'unix-ness', QNX is a proprietary real-time embedded OS with a POSIX library providing the 'unix-ness'.
      In all the cases, POSIX compatibility is what you're talking about and even Windows can have that… the success of the platform has nothing to with it, rather its the ease-of-use of the main UI frameworks that contribute to the platforms success (or lack of). iOS & Android are good in this respect whereas Symbian is notoriously bad.

      • Sander van der Wal

        Errr. BSD is one of the two epitomes of unixness, the other being System V. Unixness has never been defined by the design and implementation of the kernel, but by the system calls. Which became Posix at a time people were still uing character terminals.

      • http://twitter.com/aegisdesign @aegisdesign

        You can't define UNIX by Posix system calls. Both QNX and Windows have Posix compatible layers.

        OSX however definitely is UNIX – It's Certified even.

      • Sander van der Wal

        The argument is whether Android and IOS success is due on both systems being based on Unix. It is not, the unixy bits being hidden as far down as possible is what did not hinder their success.

      • asymco

        Unix nature does not mean just the bits, but the modular nature of the OS that Unix pioneered vs. the integrated nature that was the standard at the time (see: Multics). Modularity is what allowed the scaling of the OS.

      • Evan

        so are you saying iOS success happened inspite of unix background?

    • http://twitter.com/aegisdesign @aegisdesign

      QNX is not based on Linux.

      Symbian is open source.

      Kind of blows your argument out of the water if you don't understand the fundamentals.

      • Evan

        oops yes wrong about QNX:), but my point remains UNIX and LINUX are the hidden sauce for the fast innovation of Apple and Google, they are basically getting 'huge' foundational aspects of platform for free from the open source community.

      • Synth

        Hmm, they are one of main ingredients of the hidden sauce. Laser like focus is the other hidden ingredient at Apple. And letting the OEMs do all the hardware heavy lifting (and taking the risks) is the secret ingredient at Google. Also Google doesn't really need to worry about making money off the phones, only the internet access.
        Nokia doesn't have any secret sauces.

  • Yuri

    Those charts make me think that Symbian will be victim of the Feb 11 event. Nokia will probably announce that they will be phasing out Symbian in the next few years and replace it with Android (and maybe WP7) once they have Qt ports in place. I find it a little hard to believe MS will allow Qt on Windows Phone, because then Samsung will come and say they want the Bada APIs supported too…

    • arvleo

      might be for high end smartphones…but may still retain Symbian to develop feature/low-end phones…it would be interesting to know what aspects/feature development of Symbian is costing Nokia to bleed money i.e. features used in smartphones or feature/low end phones.

      • Atlant

        Nokia is bleeding money for a reason alluded to above: their development methodology requires an enormous duplication of effort. While it's not true that every phone requires a completely unique version of Symban, there is enough customization needed for each phone that many, many groups of engineers are charged simply with the creation of the many software branches (forks), the actual customization work, final polishing of the variant, back-porting/cross-porting of *SOME* of the inevitable bug fixes, and the like. All of this work is essentially useless to the users; they'd be just as happy with one *GREAT* version of Symbian that worked everywhere rather than ten or fifty versions of Symbian that are all different in trivial ways but none can be called great. The existence of these dozens of forks also practically guarantees that users won't see updated software for their minor devices and only see a few updates for their major devices; the required work (including the testing) is just too daunting.

        Compare that with iOS where there's basically one version extant at any moment in time and it runs on all iOS devices produced within the last two years.

    • asymco

      The foot has been lifted off the Symbian gas pedal for a while. The support of MeeGo is evidence of that. The question for the 11th is only if it will step on the brake.

      • http://twitter.com/aegisdesign @aegisdesign

        I'd highly doubt they'd announce a full stop on Symbian publicly as it would kill off any further sales.

        Back when Elop came in, one of the first things he did was kill off Symbian^4's binary break from Symbian^3 and announce that the UI redesign from 4 was being put into 3. I'd not have bought a Symbian^3 handset myself if S^4 was due a year later with a binary break in apps.

        They've also recently backported a lot of the S^3 UI paradigms into S60 5th (aka Symbian^1) removing double taps, adding kinetic scrolling etc. so S^1 and S^3 are more similar.

        MeeGo surely isn't ready to step in to the high end yet. Development relies on Qt and they're missing important aspects still such as Qt Components being finalised and Qt Mobility needs some work still. Close perhaps and if they wait for it to be perfect then they'll never ship. Still, the MeeGo job positions have been advertised thick and fast.

    • r00tabega

      Android? Perhaps it will replace Symbian at the low end, in the future at some time.
      I don't think there will be any mention of Android or Google at the Feb 11th event.

      • Evan

        android is going to take over all the anti-apple space, now how big that is I don't know, only time can tell and it also depends on how fast Google innovates and how quickly Apple responds, unless of course we have another black swan like moment(seems unlikely)

  • Investors Relations

    Where did you get Nokia Devices revenue? Devices and Services revenue is almost 30 billion euros. Is there a currency conversion error or do you exclude something?

    • asymco

      Yes, you're right, the currency conversion was missing. I corrected it. The correct percent of R&D for Nokia Devices and Services is 10.2%.

  • Waveney

    Horace, just a thought, but might there be some taxation advantage they are utilising that is particular only to Finland whereby a large R&D budget is tax free?. We do have some weird tax loopholes over here in Europe.

    • asymco

      Perhaps, but I don't think Nokia is happy with this level of spending.

  • Henk Ster

    OTOH the user experience of the iTunes/iStore software on my Windows machine is horrible – it takes up loads of resources even when Im not using it, and the functionality is just "adequate". They could assign a couple of extra developers there…

    • Kizedek

      There is a big explanation for this, though it is not one that people like to hear. You can look the details up.

      As I understand it, MS took several design decisions when creating Windows. You can variously look at the decisions as smart (given the state of technology at the time), or as ignorance of the way the issues would play out, or as a botched shortcut of some kind.

      But, apparently, a big issue with the efficiency of iTunes on Windows is that Windows is designed to bundle application processes up with the processes of other applications into system threads. What I understand is that applications under Windows are not permitted to spawn their own threads.

      There may be good reasons for this. Or, maybe, there WERE good reasons for this. Maybe it is the best approach they could take given other inadequacies and failings of the system, who knows. But apparently, bundling app processes into one system thread is efficient for the system, but not for the app you are trying to use: apparently, it can be considered a frowned-upon, or even archaic and limited, approach. But I don't claim to know all the ins and outs of system architecture design.

      Anyway, iTunes, and all Mac software, is designed with the other approach in mind — they like to spawn their own independent threads. And other OS X architecture design decisions make this a logical, efficient and safe choice.

      Now, I suppose you could argue that Apple should re-engineer iTunes from the ground up using MS', at least to Apple, clumsy approach, just to give Windows users a proper taste of what is on the other side of the fence. But, rightly or wrongly, I suppose Apple thinks that it has done what it can reasonably do; and that if Windows users were really that concerned about their system resources they would take a good hard look across the board, and question other things which degrade performance and experience, too: like why they have to run virus protection in the background, etc.

      Anyway, there is, apparently, a legitimate technical reason for why the performance of iTunes on Windows can be poor and why it uses loads of resources — Windows is designed to use and manage resources in a different way.

      As far as its actual functionality and feature-set: I am sure there are alternative media libraries out there that do some things better. But Apple do continue to work on iTunes all the time, and I frequently receive updates… I just heard about the new searching filters in the iPad app store which I will be very glad of. I don't think anyone imagines it is perfect.

      • Evan

        can you elaborate on this part "that applications under Windows are not permitted to spawn their own threads"

        here is what I get from a general search. http://msdn.microsoft.com/en-us/library/aa289523(

        " The application will spawn multiple threads,"

      • Kizedek

        I probably should have said something like "apps under Windows are not permitted to send the threads they spawn independently through the processor". The whole point is, Windows treats and manages these things differently and seems to prefer to work with processes. Generally, threads are lighter weight and have less start-up and shut-down "cost" than processes, making them much more efficient for some things; which can lead to performance gains. There is a different approach or philosophy at play here, and iTunes was created with the other approach in mind.

      • Evan

        thanks

      • r00tabega

        Do you have a reference for this? I get no hits on my various searches.

    • chad

      Seriously. Using Apple software on a PC is atrocious. Always has been. I think Steve turns a blind eye to it because he doesn't use PCs on a regular basis (ever?).

      They need to drop the idea of trying to force OS X UI chrome onto Windows users and just make it work. Or not, Google Chrome feels great on PC or Mac and it seems to have totally custom UI chrome.

    • http://twitter.com/aegisdesign @aegisdesign

      It's awful on the Mac too. Bloated, horrible UI design that dates back to MacOS 8. About 5 years ago it was passable and they've only made it worse over the years and seems to get used as an experimental playground for wacky new UI design.

      It's written using the old Carbon API so that it's cross platform compatible instead of the more modern Cocoa. Features get crammed into it so that Windows and Mac users suffer the same way instead of using the better Mac services like Sync Services (used by iSync and other apps).

      On the Mac for instance, to sync calendars and contacts to your non-Apple phone, you use iSync. Developers can write their own sync plugins for their apps or new phones. With an Apple Phone you use iTunes and Apple is the only one who can write plugins.

  • Ashu Joshi

    Horace – very good analysis. If this numbers are accurate even with +/- 5% to 7% range, the challenge is crystal clear. I think this is not merely a technology challenge, but a business challenge. Nokia, as a company is not profitable but they build low margin phones for the rest of the world, for the shareholder that may not be the best ROI but for consumers. I think the dilemma is where to focus – be like Apple – go after one segment with tremendous zeal. Or continue pursuing this multi-segment market?

  • Fonnt

    I think the difference is that Apple’s product development *is* R&D — they learn and innovate constantly. Also, Apple has never had a problem setting a long-term goal (OSX) and working towards it when others tend to follow trends (Microsoft WP7).

  • Pingback: Apple ’6x more efficient’ than Nokia at smartphone R&D | Smartphone Management in Canada

  • http://twitter.com/Carniphage @Carniphage

    I am perplexed by the number of phones Nokia produce.
    It has to be expensive to bring so many products to market. But there seems to be very little difference in the final products. This has to be one of the causes of inefficiency.
    I'd urge them to simplify down to two lines.
    A business communicator with the keyboard – and a consumer-facing touch screen.
    Everything else should go.

    C.

    • TomCF

      I bet Nokia's relationship with the carriers has something to do with that. Motorola is similar. The carriers are their customers, not mobile users. The carriers force odd choices on the handset makers. Apple was possibly the first phone manufacturer that refused to compromise their product. (Maybe RIM was first, but I think they made concessions as well.)

    • kevin

      Segmentation is a known marketing and sales tactic to appeal to multiple target groups through customization. All companies do it, even Apple on its iPod and Mac lines, but their criteria for creating segments differ. Some look at use cases. Some look at age groups. etc. Segmentation could also help by generating lots of products for distributors and thus occupy shelf space and squeeze out competing products.

      Nokia used segmentation in the past to great effect. It was a Nokia strength. But Apple has changed so much in this industry that it's unclear to me whether Nokia's segmentation ability is still a competitive advantage.

      • r00tabega

        Completely on-spot, kevin.

        Nokia's 30 phones may be mind boggling, but have you ever taken a look at how many different laptops are actively sold by companies like Acer or Asus at any given time? Hundreds. Mind boggling.

        For another complete level of crazy, try Cisco and it's 20000+ products. I hear it takes dozens of dozens of folks there just to keep the product metadata in line.

        It's not exactly easy for the customer.

    • http://twitter.com/aegisdesign @aegisdesign

      They do tend to re-use the same platform on different phones though. The C6-01, C7, N8, E7 all run with pretty much the same hardware inside and same OS. The N97, 5800, 5230, 5530, 5228, N97mini and probably a couple of others were pretty much the same too.

      • kevin

        Correct. Platforming is a way to achieve segmentation while minimizing the use of resources. The idea is to have a few major platforms (based on common hardware or common software) and then many variations off of each platform. Car companies are a good example of this, though it's done in almost every manufacturing industry, from airplanes to coffee-makers to single-use cameras.

  • Pingback: Nokia gasta mais com Symbian que Apple com iPhone | Apple

  • Guest

    a bit off topic, but ..
    Has this site ever been viewed as rendered by chrome ? What a disaster !

    • asymco

      What seems to be the problem? Looks good to me.

    • asymco

      For the record, Chrome is the second most popular browser used to access this site. Visits for the past 30 days:

      Safari 47.68%
      Chrome 18.12%
      Firefox 15.50%
      Mozilla Compatible Agent 9.94%
      Internet Explorer 7.22%

  • Guest

    Nokia employs many more people and spends many times more than Apple in almost every other department, including marketing, strategy, management, finance, IT and other corporate services. Nokia spends massively more than Apple on external consultants, strategists, analysts, advisers, outsourcing.

    Nokia is an oversized sluggish bureaucratic dinosaur.

    • Atlant

      Many (most?) of Nokia's woes could be fixed simply by running a weed-whacker through the management ranks; sacking about 80% of them sounds about right to me, especially if you started at the PowerPoint Palace in Espoo.

      The recent analysis in the Helsingin Sanomat was spot on: Nokia's management stands in the way of progress at nearly every turn; they do everything they can to suppress new ideas that don't fit into their pre-conceived notions of how things should be.

  • http://twitter.com/hurhurhurhurhuh @hurhurhurhurhuh

    Maybe it is because Apple is a notorious patent violator. You don't need a whole lot of R&D when you can just steal another company's innovations and stick them in a die cast brushed aluminium case and have people paying double or triple the price of similar devices.

    • asymco

      From what I've seen so far I don't think Apple uses die casting but milling on cold rolled Al blocks. See the video on the unibody MacBook. http://www.youtube.com/watch?v=3OOtuftGO1U

      Die casting is metal injection molding. See in contrast: http://en.wikipedia.org/wiki/Die_cast

      Prior to the current milling process I believe the Aluminum products were made using stamping press technology.

  • John

    I suspect that part of Apple’s efficiency is that they have been at it so long. As pointed out above, the roots go back to Next days. For over a decade OS X and it’s core technologies have been slowly polished and tested in the marketplace making Cocoa and Objective C powerful and easy to use.

    Competitors are playing catchup which is chaotic and inefficient.

    • Evan

      chaos v/s order, eternal debate which is good, I believe both have strong points in their favor. One should not be so blinded by one system, that one cannot see the good points of other system . Internet is chaotic seemingly under nobody's real control(although egypt has shown that countries hold kill switch), but it is the biggest platform on the world today.

      • dchu220

        A mixture of both is always needed. I think the critical factor is if the system is well aligned. Especially if you are building a platform. You need to balance the needs of multiple parties. Adwords CPC/Ranking was revolutionary because it gave users high quality relevant results and marketers reduced risk/measurable ROI.

  • John

    One other thing I just ran across (from last October). Nokia may be hobbling themselves with a terrible corporate structure.
    http://www.theregister.co.uk/2010/10/14/nokia_dil

  • noogie60

    Microsoft could also take the prize for spending heaps on R&D with little to show for it.
    Speaking of Symbian and its issues, the roots of it go back to the Psion days when their engineers were proud they came from "hard" science backgounds and were thus good at wringing every last bit of performance but were never strong on usability http://www.theregister.co.uk/2010/11/09/symbian_d

    • Psi

      @noogie60, the quote that you re-worded from TheRegister is saying that Psion's engineers would put performance ahead of usability *for programmers* – ie that things in the OS like "active objects" were hard to learn to *program* with.
      They weren't saying that the Psion platform was low on conventional "Usability" (ie from a user's point of view) – its Usability was half the reason that the world's phone industry came together and standardised on it..
      (Sure, it didn't take the Nokia's of the world many years to destroy that Usability, but that's hardly Psion's fault.)

      • noogie60

        True EPOC wasn't that hard to use for the consumer.
        However its history meant that it and thus Symbian were always tricky to develop for – which I'm sure doesn't help Nokia in maintaining it or to try to develop it.

  • http://www.moretora.com/ Tora

    This difference would be so important if Nokia was making great products. The fact is they are not and it appears that they are on the decline. It is difficult to back out what these companies have in the R&D pipeline just from looking at the numbers. What it does point out is how relatively lean Apple is and that more spending by Nokia on R&D will not likely improve Nokia’s short-term situation situation.

  • mitodna

    I remembered I read that an ex-engineer trying to get 256MB of RAM to N97, but all he got is that make the software work at 128MB.

  • O.C.

    Well the fact Apple invest very modestly on R&D is not that hard to figure out. Apple products aren't sold on the merits of its specs. They rarely if ever come out with something we haven't seen years ago in a phone.

    As every new smartphone a competitor brings out gets labeled a potential iPhone killer because its specs are better. And there are plenty of smartphones that have more technology packed inside. Using dated technology like multi tasking and facetime – with a spin – to make it sound fresh and new requires very little investment.

    Apple does a great job at selling technology we've pretty much seen before but with a new spin.

    • asymco

      Historically, HTC have had the highest specs in the industry. They spend a lot less than Apple.

    • unhinged

      "It just works" is spin?

      What use is having a spectacular feature if only a small subset of users can figure out how (or bother to expend the effort) to use it? You might as well have not spent the R&D and manufacturing resources, because the majority of your customers are not using the feature (although I grant that some might be using it as a purchasing decision).

      Don't discount the value of improving the user uptake of a given technology.

    • dchu220

      The truth is that there rarely any truly new ideas in the world. Usually the victory goes to the person who works hard enough to find out how to make it work right. There were tons of search engines before Google, but Google was the first to do it right.

      Figuring out how to do it right can be very expensive as well. There isn't a direct correlation between building on previous ideas and R&D efficiency. Now, the skill of prioritizing which technologies are likely to have the biggest impact on improving your products is heavily correlated to with R&D efficiency.

  • Atlant

    Knowing *WHEN* to bring your products to market is just as important as knowing *WHAT* to bring to market. Nokia has very often brought a killer idea to market but a bit too soon for the market to be ready for it. Then, rather than being willing to support that idea in the marketplace, they abandon it. Later, someone else (often Apple) brings that idea to market when it's "more timely" and the idea is then successful. Certainly touch-screen phones could be cited as an example of this.

    • http://twitter.com/aegisdesign @aegisdesign

      I would suggest N-Gage as another example. Online gaming market? Phones with GPUs in? That would be 2003 and Nokia. All the rage now of course, just not with Nokia.

      I still reckon there's life in N-Gage yet with the S^3 models and "joining an ecosystem" such as Nintendo's ecosystem.

  • Roger

    Well, one thing Apple don't seem to spend much R&D on is power efficiency to provide an acceptable battery life :-)

  • Pingback: Cassandra – Monday Review: It will soon be Friday

  • John

    The problem with a blog analyzing Apple is, that if it does it well, it'll bring out the Apple-phobes. Acceptable battery life? Are you kidding me? iPhone 4 and iPad have astonishing battery life. Macbook Air is pretty respectable for a laptop not much bigger than an iPad.

    I won't bother replying to O.C. We've heard that canard often enough.

    It's amazing how often Apple turns the world on its head, and then afterwards when the ideas/technologies are ubiquitous and taken for granted, people will say "Well, anybody could have done that with good marketing/enough fanboys/IP theft/<insert random stupid thought here>."

    I hate pithy lines like "it is what it is," but sometimes you just gotta throw up your hands and say "haters gonna hate."

  • http://martijnlinssen.com MartijnLinssen

    Horace,

    where did you get these figures? Apple 2010 "iPhone and related products and services" are $25.2 billion – where an unknown portion of that "Includes revenue recognized from iPhone sales, carrier agreements, services, and Apple-branded and third-party iPhone accessories"

    So making that $30 billion is a gross overstatement – and may I ask where you get headcount involved in R&D? None of that in the annual report, let alone that being subdivided into products

    • asymco

      The data I have is from 10Q statements: $ million (calendar Q1 through Q4 2010):
      5445, 5334, 8822, 10468.
      The sum is $30b. Let me know if you see an error. (I might suggest your figure comes from the fiscal year?)

      The derivation of R&D figures and headcount is explained in the post.

      Note that the comparable figure from Nokia includes Devices _and_ Services.

      • http://twitter.com/MartijnLinssen @MartijnLinssen

        Hi @Asymco, I do. Apple has a broken bookyear, and ends its fiscal year 2010 in September 25 – not December

        So you should take the figures from:
        Q1 (ended December 26): 5,578
        Q2 (ended March 27): 5,445
        Q3 (ended June 26): 5,334
        Q4 (ended September 25): 8,822
        ______________________________
        25,179

        You might want to check out http://www.martijnlinssen.com/2011/01/real-profit… for further revenue, cost, profit and R&D breakdown per employee for Apple, Google, Microsoft and many others. All full years so safe

        HTH

      • asymco

        I am not comparing fiscal years, but calendar years. "… plot the Devices R&D for both Nokia and Apple over the entire 2010 period". Why would I use fiscal year total for Apple when I'm comparing data with Nokia which is matching the calendar year?

  • grandma

    There are few default reasons for the R&D expenditure difference:
    1. Apple hires very very few people for each engineering job. Nokia throws 100 engineers at it like all large corporations.
    2. Apple benefits from established technology plateau that has been reached by the sweat of thousands of engineers like the ones in Nokia. Nokia were key players in developing GSM, UMTS, HSPA over the years. Apple contributed zil to that. We must remember Apple is an integrator – and they are great at it!
    3. R&D expenditures often include much more than just "research" or "advanced concepts" like plain "boring" development. Development is a bulky budget and quite often you have to serve legacy products and platforms. Isn't this the classic event of established successful corporations falling victim to their own weight?
    No reason to raise eyebrows reading this article other than just agree and confirm once again the disruptive change that Apple brings to all of us.

  • Pingback: Nokia ‘considering move to Silicon Valley’ | Smartphone Management in Canada

  • Pingback: Two turkeys don’t make an eagle, but no penguin will ever soar. | asymco