« The grid land | Main | The MySpace mirror »

Complementary innovation

June 01, 2006

In a recent post, I talked about the central role that complements are playing in shaping the economics of the web. (Complements are products that tend to be consumed together, like popcorn and movies or blogs and Google ads.) In Complementary Genius, my latest column on innovation for Strategy & Business, I take a broader look at complements and the different ways companies can use complementary innovation as a strategic weapon.

Here's how it begins:

What were André and Edouard Michelin thinking? In 1900, shortly after the two brothers took control of their family’s venerable rubber business in Clermont-Ferrand, France, they suddenly decided to publish a guidebook for tourists. Their Michelin Guide provided information on gas stations, hotels, restaurants, and roadside attractions along with various maps and driving tips. The brothers printed 35,000 copies of the first edition - and gave them away free.

According to our contemporary notions of business logic, the move seems hard to justify. After all, book publishing has little to do with rubber processing. Management gurus, if they had existed then, might have chided the brothers for losing sight of their “core business” and expanding beyond the scope of their “organizational capabilities.” They might even have used the story as a case study on why family businesses should bring in professional managers.

But the diversification turned out to be an act of genius ...

Read on.


Your article was very nice. I had also put up, a while back on my blog, some economic complements and substitues in the IT industry.

Posted by: Nitn Goyal [TypeKey Profile Page] at June 2, 2006 04:24 AM

hmmm... not sure of this one. 1. I appreciate the innovative way you've just got me to register with www.strategy-business.com - but feel like its one step above a "pop-up" ad
2. although wiki and other sites of unbias facts -www.michelinman.com - state what you did about Andre' and Edoard's strategy of driving tire sales via the Michelin guide - its very easy to connect such genius via history. Now, if you were to tell me they sent Henry Ford a white-paper on moving-line assembly to produce cars for the average person - you'd have something. My assumption is, as heirs to a family business selling rubber (albeit a commodity) in early 1900s France, they were pretty well-off. They and their well heeled friends all owned "exotic" vehicles (things with engines not organs) but didn't know where to get gas, where to stay, etc... thus, as a compliment to their "class" they published a book with their friends and spread it around boosting their class status. Ego and luck drive most "genius". It just so happened that this was in the middle of the 2nd industrial revolution (1874-1914)and rubber (to no doing of their own) was needed to heel the horses of their time. Now, to survive as a company for 100 years is pure genius! That would be an interesting article as they have to be but a handful that have done it. With 20% marketshare, inacting the world standard for tires - PAX system, and offshore manufacturing in brazil and Nigeria ahead of their time - they are amazingly well run. However, I find it hard to connect agreeing with a Michelin five star restaurant rating to my purchase of tires (today's commodity). They will fight the Innovators dilemma on both michelin travel guides and tires in the next 100 years - and a spin-off of the guide is not unfathonable.

Posted by: chris mills [TypeKey Profile Page] at June 2, 2006 11:55 AM


No, I think you're wrong. It's pretty clear that the brothers published the guide with the specific goal of promoting driving (and tire buying) just as earlier they had sponsored a big bicycle race to promote bicycling (and tire buying). I don't think business owners would print 35,000 copies of a book to impress their class (though that may have been a secondary motivation - I wouldn't know). But you're right about Michelin's many achievements since then, not least of which was its willingness to kill its old business by pioneering radial tires.

As to your equation of my post with a pop-up ad, well, I don't have much to say there. People are rolling around in free content like pigs in shit, and they're still finding reasons to complain about it. "Hey, can't you make this free stuff even freer!?!" I guess it's to be expected. Show me a gift horse, and I'll show you a bunch of people looking into its mouth.


Posted by: Nick Carr [TypeKey Profile Page] at June 2, 2006 12:14 PM

Nick, I didn't like the signup step either. As far as the free stuff is concerned you should know very well
that nothing is free.
What I m' saying is that the people who come and read your blog are giving you sth of value (remember the attention economy?) Many of them (eg myself) are recommending your blog to others. Isn't this kind of reward good enough, or only money talks?

Posted by: pgrontas [TypeKey Profile Page] at June 3, 2006 10:24 AM


If you don't want to sign up, don't sign up. That's fine. I don't make any money whether you sign up or not. The signup requirement is Strategy & Business's, and I don't begrudge them it. You can always go to a newsstand and buy the magazine.


Posted by: Nick Carr [TypeKey Profile Page] at June 3, 2006 10:08 PM

Quote: Although many people traveled on the lines during weekday rush hours, commuting back and forth to work, few rode them at other times. The imbalance in demand undermined the profitability of the operators. They had to build their systems to accommodate the peaks in usage, but most of the time their expensive systems generated little revenue./EndQuote.

This is exactly the problem with delivering computational power via silicon chips, packaged inside the Dell box, which is itself packaged in a cardboard box. Individual businesses are facing the very same problem as the railway operators in the United States. You will witness periods of very intense usage of the computational power available on the silicon chip. On the other hand, there are very long off-peak times, when the silicon is doing nothing more than running a screen saver. The modern business cannot justify that equation. That is why Intel and Microsoft are both throwing away potential revenues at the moment, and leaving a gap for a new innovator to move right in. It is this rising and falling in peak usage that limits the business model employed by Wintel. It is based around the shipping of atoms, rather than shipping of bits, or pixels, which are the molecular equivalent of bits. Or Voxels, which are three dimensional pixels!

Quote: Examine all the products and services that are used in conjunction with your own goods. Are any of the complements in short supply? Are any expensive? Are any difficult for customers to find and use? It may be possible for your company to spur advances in the production or distribution of complements in order to make them cheaper and more plentiful./EndQuote.

Look at the notion of a rugged PC, from Intel, which fits quite neatly into that idea of complements.


Because the Intel business model is based on the shipping of atoms, you want those atoms to be rugged enough to last the journey into middle Africa right? While if you business model was more like that of Sun Microsystems, and based upon the shipping of bits, in the form of a utility service, you would not need to worry about your PCs being 'rugged' enough! Intel doesn't want customers in Africa, China or India, getting attached to a thin client! The tribes people should not turn to lightweight and robust handheld technology. But if you want a really good example of complements in the computer industry, from the past, just check out the quotations from Martin Campbell Kelly's book, I posted here:


At least IBM had the sense in those days to include a manual with its computers! Years later, DEC hadn't even learned that crucial lesson when they were shipping their mini computers.

Quote: Of course, such strategies can cut both ways. Your own core products are probably complements to some other company’s offerings. It would be wise, therefore, to think defensively as well as offensively. How vulnerable are you to innovations that would make your own products cheaper and more plentiful?/EndQuote.

That was always the old argument about Intel's Itanium and Xeon products. If you make Itanium cheaper and better, you are simply eating into the revenues from selling the Xeon product. AMD understood Intel's dilema here, and capitalised on it, by introducing the AMD 64 platform, which offered customers a stepping stone between older 32-bit applications and full 64-bit. To make it even worse, Intel had investigated that notion of a bridge to 64-bit, years ago. But rejected it, because it might threaten the Itanium product! Between, all Intel's efforts to protect the value of their products, it dropped the ball. It allowed AMD to slip in and grab a major chunk of the server market, and all Intel could do, was watch helplessly from the sidelines. So, trying to save your own products from your own products, is not good advice in my opinion. In this sense, IBM did make the right choice to move to Linux. Sure, the move cut into their own product and service revenues. But at least they prevented a window of opportunity appearing, like Intel allowed for AMD. IBM did what Intel didn't do, IBM offered a direct pathway for its customers to move to Linux. While Intel, too greedy, didn't give its customers a sustainable path to 64-bit applications, while staying within the Intel platform. Sure Intel have a kind of AMD64 technology nowadays, but the cat is already out of the bag. While I take your point Nick, that Intel were clever with Wifi branding, Intel did miss the much bigger branding opportunity that has now been taken by AMD. As far as customers care, AMD offered the bridge-way to 64-bit applications, Intel didn't.

The Centrino campaign was a repeat of the older 'Intel inside' campaign. Intel know it's Pentium brand is dying slowly on its feet. Intel understand the requirement to find a replacement for the aging Pentium brand. Intel are pumping money into the Centrino logo, coaxing markets away from 'Pentium' and towards 'Centrino'. From a technological point of view, Centrino is about as hollow and meaningless as it comes. But from a branding point of view, it is pure gold dust. Intel has always been very good at looking inside itself. Which is a positive attribute when you own most of the market share, and are really competing against yourself. But Intel does need to learn to look outside of itself.

Microsoft have the opposite problem entirely. Microsoft has always been good at looking outside of itself. But nowadays, it needs to looking inside. Looking inside is not something Microsoft likes to do. That is not part of it's cultural DNA. It has been too successful, by looking outside of itself. There is a very interesting book, written about this in relation to Digital Equipment Corporation, the mini computer maker. It is called 'DEC is Dead, Long Live DEC'. I suggest that Intel and Microsoft both read this book, and then 'bang their respective heads together' on this problem of cultural DNA. To be more than a Wintel merely in name, and to be Wintel in every sense.

You wrote about elegantly about this Nick in 'Does IT Matter'. About the two ways a company can perceive itself. That in reality, a company needs to look outside aswell as inside. All the famous eastern brands such as Yahama, Sony, Mitsubishi, Hitachi, Toshiba can be found on pianos, motorbikes, music synthesisers and CDROM drives. Every good product that Yahama sells is a good reflection of the strength and health of the parent organisation. In the West, a company like Johnson and Johnson will employ distinct, separate brands for each and every product they sell. That is closer to the Intel way, where you find sub brands, and campaigns within campaigns. Intel Inside, Pentium, Centrino, Mobile Centrino, Mobile Pentium, Pentium D, Celeron, Celeron Mobile, Celeron D, Itanium etc, etc. In the world of branding at the moment, it appears the East is adopting western techniques and visa versa. A very good reference book on this, is 'Re-inveting the Brand', by Jean-Noël Kapferer. It is a very easy book to read, consisting of about 20 essays, which can be read individually, at any time.

Brian O' Hanlon.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 4, 2006 01:49 PM

Read about Microsoft's problem in more detail here:


Brian O' Hanlon.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 4, 2006 01:59 PM

My comment here:

Quote. At least IBM had the sense in those days to include a manual with its computers! Years later, DEC hadn't even learned that crucial lesson when they were shipping their mini computers. /EndQuote.

... is really trying to say, that IBM understood the benefits of shipping out some few bits, in the form of printed manuals, along with the atoms, the computer hardware itself. Later on, because of a lawsuit against its monopoly of the computer industry, IBM unbundled the software from the hardware. The shipping of bits and atoms became separated and that has overshadowed the industry ever since. DEC was a company, that built its success entirely from the shipping of atoms in greater quantities and on time. As Dell is doing today. But like DEC in the past, I fear it will be too late by the time Dell, Intel and others realise the futility of shipping atoms on their own. The 'rugged' PC concept by Intel smacks of a crisis in the company to innovate in the right direction. Take a look at Apple computers, which designs its own software and hardware. The opposite to the 'unbundling' concept of IBM in 1970. Apple has enjoyed a strong market performance in recent years, because it does exactly what Dell has failed to do. To innovate and market a product, which is a combination of the bit and the atom. Which makes IBM's decision to invest in Linux, look very wise indeed. Or Microsoft's move into shipping XBOXs, also quite clever.

Brian O' Hanlon.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 4, 2006 02:13 PM

If you really think about it, the incident of the AMD64 technology backs up my observation, that Intel is too focussed on atoms, and not focuessed enough on bits. Only in recent years has Intel began to invest in software engineering talent! AMD are in the business of shipping atoms, in the form of silicon too. But AMD managed to focus more on the 'bit' side of the equation. They realise that bits were more important to customers, than atoms. That customers would purchase based on the knowledge their bit investment would be safe guarded. Not the other way around. A sudden transition to Itanium 64 bit technology, would mean a devastating loss of years worth of knowledge capital in the form of 32-bit applications. That was unacceptable.

I would trace Intel's expertise in dealing with the atomic side of the equation, back to their investment in fab plants all over the globe. Forget about what it lost making Centrino chipsets. Chipsets are only used to back-fill spare FAB capacity anyhow. Intel has billions of dollars tied up in atoms all over the world, in the form of chip fabs. Intel might as well experiment with those old fabs it has lying all over the world. Intel is using those older fabs to make Itanium for a decade or so. A time will come I am sure, when Intel reaps a reward from that activity. Then you will quickly see Itanium moved onto the cutting edge silicon process, and products like Pentium demoted to the older.

But the trouble is, Intel is too successful at looking at their atomic side of the equation. And doesn't appreciate fully an economy based on the manipulation of bits. Which could mean anything from a mobile phone network, to a web service. Apple's gutsy move in the music industry, is entirely informed by a compound of various expertise, in the skilled manipulation of bits and atoms. Dell etc, and the Wintel is just about shipping boxes I am afraid, and today, is hampered by their lop-sided knowledge.

Brian O' Hanlon.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 4, 2006 02:23 PM

There is only one ray of hope I can think of for Intel. Intel is now putting chips in Apple computers too. Intel has been excellent at pushing atoms around the globe, into various markets. But Intel has become one-sided in its perception. It has spent far too long, hanging around with its buddy Michael Dell. Michael is the experts expert, in pushing around atoms. Try working with Dell for any length of time and you will realise that.

Because Intel is now supplying chips to Apple computers, I only hope that some of the cultural DNA in Apple can rub off on Intel. Or in the same way, Sun and Microsoft announced a cooperation not too long ago, which will increase the contact between those two organisations. Perhaps Microsoft will learn something more about big iron, and visa versa, Sun can learn something about intelligent client machines.

Only in recent years has Intel began to invest in software engineering talent! How long will it take for the influence of that talent, to work its way into the cultural DNA of Intel as a company. That is the question.

Brian O' Hanlon.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 4, 2006 02:35 PM

Another common theme of posting here at this blog site, seems to be the hive mind, collective authorship, wikipedia, de-centralised surveillance and what not. It is just interesting, in that arena too, the best solution is a combo of individual and collective. Jaron Lanier makes that point well, here in this paragraph.

"Every authentic example of collective intelligence that I am aware of also shows how that collective was guided or inspired by well-meaning individuals. These people focused the collective and in some cases also corrected for some of the common hive mind failure modes. The balancing of influence between people and collectives is the heart of the design of democracies, scientific communities, and many other long-standing projects. There's a lot of experience out there to work with. A few of these old ideas provide interesting new ways to approach the question of how to best use the hive mind."

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 5, 2006 03:26 PM

Summary of points I discussed here this weekend would include:

individual versus collective.

company looking inside or outside itself.

eastern versus wester branding strategies.

atom versus the bit.

Enjoy your week everyone,
its been quite fun posting here.

Brian O' Hanlon.
June 2006.

Posted by: Brian O' Hanlon [TypeKey Profile Page] at June 5, 2006 03:29 PM

Post a comment

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Remember me?

carrshot5.jpg Subscribe to Rough Type

Now in paperback:
shallowspbk2.jpg Pulitzer Prize Finalist

"Riveting" -San Francisco Chronicle

"Rewarding" -Financial Times

"Revelatory" -Booklist

Order from Amazon

Visit The Shallows site

The Cloud, demystified: bigswitchcover2thumb.jpg "Future Shock for the web-apps era" -Fast Company

"Ominously prescient" -Kirkus Reviews

"Riveting stuff" -New York Post

Order from Amazon

Visit Big Switch site

Greatest hits

The amorality of Web 2.0

Twitter dot dash

The engine of serendipity

The editor and the crowd

Avatars consume as much electricity as Brazilians

The great unread

The love song of J. Alfred Prufrock's avatar

Flight of the wingless coffin fly

Sharecropping the long tail

The social graft

Steve's devices

MySpace's vacancy

The dingo stole my avatar

Excuse me while I blog

Other writing

Is Google Making Us Stupid?

The ignorance of crowds

The recorded life

The end of corporate computing

IT doesn't matter

The parasitic blogger

The sixth force



The limits of computers: Order from Amazon

Visit book site

Rough Type is:

Written and published by
Nicholas Carr

Designed by

JavaScript must be enabled to display this email address.