Robert J. Pera's weblog

Over the past decade, electronics manufacturing has largely shifted into China because of cost-savings.  However, these cost-savings can often be misleading as manufacturing in China also comes with the risk of IP theft that can potentially lead to financial damages far in excess of the capital saved in manufacturing there.

As a hi-volume hardware manufacturing shipping millions of devices per year, Ubiquiti Networks partners with contract manufactures based in China.  For entrepreneurs looking at manufacturing in China, I want to give some advice in hopes of preparing others for the potential dangers they may face.

Before I start, I want to point out something that is often not communicated enough to the Western World regarding business ethics and the general integrity of the justice system in China.  Looking at the country of China as whole and making generalities about the business culture is somewhat unfair.  It is well known that China has the largest population of any country in the world, but it is not as well known that it is very much still a country with a great range of internal variance (probably more so than any country in the World) when it comes to the evolution of business ethics and intellectual property protection amongst its 23 provinces.  (to be exact, 23 provinces, 4 directly administered cities, 2 special-administered regions (HK/Macau), and 5 self-administered regions by ethnic minorities such as Tibet and Inner Mongolia)

In fact, a friend of mine in China summarized the variance best when he broke down 3 regions

 

-In Shanghai the attitude can best be expressed by the phrase: 上面说行才行 – meaning don’t do it unless the authorities say it’s ok to do it

 

-In Beijing, it’s 上面没说不行就行 – if the authorities didn’t say don’t do it, it is ok to do it

 

-And in Shenzhen, 上面说不行也行 – do it even if the authorities say you can’t do it

 

Unfortunately for Western technology companies, a large part of manufacturing today takes place in the Shenzhen region.

With that, here are my 3 pieces of advice

1. Take what your contract manufacturing partner tells you with a grain of salt: When you first engage with a manufacturing partner in China, they will likely tell you “We will sign an NDA” or  “we have a very strict information policy” or “we have very tight security”, etc. in an effort to make you feel comfortable that your IP can be protected in their hands.  Reality often deviates from words. The motivation for factory personnel to leak IP or take their manufacturing and project experience to lucrative counterfeiter operations is very high.  This is especially true if they are based in the Shenzhen region.  It is therefore very important to do reality checks with your partners to make sure they have effective internal IP protection processes and are in fact doing their best to protect your IP.  If they are not, it is time to change partners.
2. Apply for IP registration in China in advance:  Registering trademarks and patents in China is a long process that can take years.  You must start this process immediately and ideally well in advance of any plans to manufacture there.  Enforcing this IP in certain parts of the country can be very difficult, but there is a more important reason for filing.  There will be others in China that will attempt to fraudulently file your IP once you show up on their radar and will look for ways to use it to profit at your expense.
3. Use manufacturing IP to protect your product IP:  This is the most valuable piece of information I can give.  If you can protect your product IP at the source (manufacturing level), the potential agents of IP theft can be effectively neutralized.  At Ubiquiti, we developed a powerful system of product counterfeit protection that is virtually unbreakable.  It consists of (a) a security IC with a “a unique water mark signature”, (b) a software application running on a test station that loads the binary firmware image on to the hardware, and (c) a cloud based server which authorizes only stations with approved IP and MAC addresses, authenticates loading instances based on cryptography involving several dynamic values as well as the security IC signature; and logs all activity.  The system also uses passwords that must be changed on a frequent basis at the manufacturing line.  Using an anti-counterfeit mechanism such as the one described is your best protection against the future threat of rogue counterfeit manufacturing facilities.

 

Last weekend I had the pleasure of having dinner with Pitt Hyde, the Founder of AutoZone (NYSE:AZO) and just an overall impressive, humble, and great person.  Perhaps my favorite story of his was when he recounted his time serving on Sam Walton’s Walmart Board of Directors in the early days of his company.  According to Pitt, Sam would pick him up from the airport in a run-down car and take him back to the Walmart offices where they would prop up a fold out table and chairs to have their board meeting.

Apparently when Sam Walton was first starting his business, he had a unique vision where he could serve the underserved communities – areas in which larger store chains did not want to enter as they found it impossible to make profitable business cases for doing so.  Not only did Walton succeed in his strategy, he was able to make his business immensely profitable.

“Once committed to discounting, Walton began a crusade that lasted the rest of his life: to drive costs out of the merchandising system wherever they lay—in stores, in the manufacturers’ profit margins and with the middleman—all in the service of driving prices down, down, down,”

–John Huey, 1998 Fortune Magazine(also co-authored Walton’s autobiography)

Walton started from humbling beginnings and was seen as an underdog with an unorthodox strategy, but through the years, Walton was not only able to apply his strategy effectively to underserved markets, but eventually would grow to dominate the whole country en route to becoming the largest revenue company in the entire World.

Today when people ask me to compare Ubiquiti Networks strategy, business model, and story to another company, I like to believe it’s most similar to Sam Walton and his early days in starting Walmart.  We have found a way to deliver powerful and sophisticated plug and play hi-tech connectivity solutions to underserved areas of the world while making a profitable business out of it.  Unlike other companies in our space competing for existing market share, Ubiquiti has created new markets and opportunities that didn’t exist before in places such as Africa, South America, Middle East, India, and China.  Although we are in the very early stages of the company and predominately represented in underserved areas of the world, I also aspire to drive Ubiquiti’s transformation from David to Goliath one day just as Walton did.

 

I want to preface this post by saying I have very little knowledge about the business side of television.  The below analysis is the result of some surface investigation combined with my own experience in building new technology platforms, attacking distribution inefficiencies, and disrupting markets.

This weekend, I decided to finally take a look at the HBO series Game of Thrones.  After watching a few episodes on iTunes, I took to the Internet to do some deeper research on the characters and the creative process in making the series.  But, I unexpectedly found something else that was very interesting…

Game of Thrones stands to become the most pirated TV program of 2012.  In fact, piracy for Season 2 has reached beyond 25 million downloads already.  Now the interesting piece – a significant amount of these pirated downloads are originating from viewers that would gladly pay to view the show, but cannot.

How could this be?  Well, HBO owns the show and the only way to view the current season is to be an HBO subscriber through Cable or Satellite. HBO does offer the “HBO Go” streaming Internet service, but this is only available to existing HBO cable/satellite subscribers.  Why would HBO allow this piracy to become so rampant when they could distribute Game of Thrones through iTunes/Netflix and monetize it more efficiently?

To find the answer, let’s take a step back and look at the distribution of content.  HBO funded the creation of Game of Thrones and owns it rights.  HBO is a network that has deep relationships with the cable and satellite companies, which it uses for distribution.  Together, this network/distributor partnership has created a viewership base that it can control and monetize.

Unlike the pre-iTunes music Industry, the traditional television industry and the network/distributor partnership has still managed to maintain profitability even with the advent of pervasive broadband Internet because viewers still spend a lot of time watching cable and satellite TV.  However, the Game of Thrones piracy situation has exposed the traditional model as being incredibly inefficient by today’s standards.

I believe this traditional model will change.  The television Industry will inevitably undergo a massive change where traditional distribution mediums like cable and satellite will become increasingly primitive; and eventually completely replaced by Internet distribution.  In this future, there will be no television channels or traditional remote controls.  Everything will converge to the ideal user experience – the ability to choose the content you want to buy and view it from any web-connected device.

However, this change is not necessarily in the long-term financial best interest of the network/provider partnership.  And that is the reason why I believe Game of Thrones is not available for Internet distribution today.  The traditional players are in a difficult position as they know that the more viewership is shifted to Internet platforms, the more the financial viability of their traditional model becomes exposed.  Naturally, they are defensive about enabling a new model that will compromise their main leverage position (their viewership base).

In the end, the traditional model will fade away, but how fast this happens will depend on who can enter the market with a disruptive Internet distribution platform – one that quickly can own the viewership base.  Once this happens, I believe traditional networks will be reduced to older content licensing companies and content creators will have a newfound control over their financial destiny.

Although Netflix, iTunes, and Hulu have made significant progress, I believe for the television Industry to be changed overnight, a more dominant platform must be established that has a mission of aggressively buying the viewership base.  With that in mind, let’s see what Apple has in store for their upcoming iTV platform.

This week I was in Beijing working on a PR campaign to build awareness of the Ubiquiti Networks brand in China.  Beijing’s development pace has been astounding and the quality standards of their airport, shopping districts, hotels, and business centers now rival those of Dubai or even Tokyo.  The city is very alive and I can feel the people carry a collective motivation, pride, and sense of urgency to continue in its advancement.  The other thing I noticed is the proliferation of Apple devices has skyrocketed overnight.  There were countless people using MacBooks, ipads, and iPhones, where as during my first visit to Beijing a couple years ago, I would only see them on rare occasions.

Also, in some areas of the city, there is also a lot of international diversity; much more than I expected.  I would not be surprised if in 30 years the demographic diversity in Beijing starts to approach those of major cities in the United States and Europe.

One of the more interesting buildings I noticed in downtown Beijing was the China Central Television (CCTV) Headquarters.  Below is a photo of the building I took on the way to the Airport.   The building is a unique design that has a radical pretzel-like shape.  Apparently, it was a very sophisticated development process as a requirement of the design is seismic resistance to a Richter scale 8 earthquake.  The total project cost was said to be upwards of $1billion USD!  Although Dubai’s Burj Khalifa is the most impressive building I have seen to date, the CCTV building might take second place.

 

Unfortunately in the world we live in, there are the few that truly create value and then there are the many that prey off the value creation of these few.  One of my favorite movie scenes — from the classic “Hustle and Flow,” about a struggling Memphian (Djay) who enlists the help of a music producer (Keyes) to reinvent himself as a rapper, captures this idea pretty well.

People often ask how my company (Ubiquiti Networks) is able to achieve leading public market profitability and growth metrics while only employing upwards of 100 employees.  The answer is simple:  when you have a team of individuals who “walk the walk,” you can run circles around competitors with massive teams of employees who are resolved to “talk the talk.”

Although Ubiquiti’s operating expenses are far less than other technology companies, it is not because we are shy about spending.  In fact, on an individual basis, our compensation levels are often significantly higher than standard market ranges.  It is our ability to know where to spend which results in our unusually high return on R&D investment and unusually low operating expense metrics.

To illustrate this, below is a diagram showing distributions of engineers as a function of their value contribution.  For the engineer types I classify as “talkers,” I believe they often contribute negative value contribution to a company.   The “walkers” on the other hand, will always have at least some positive value contribution.  And, if they are “superstar” types, they can deliver phenomenal value contribution and return on investment.  At Ubiquiti, we focus on hiring these specific types while being as disciplined as possible in scaling our R&D teams.  I also believe it’s just important to avoid the wrong hire, as it is to get a great hire.

 

If you can find, recruit, and build small teams with a high concentration of these types of hands-on engineering “superstars”, the results can be awesome – including the potential to disrupt markets and challenge much larger competitors.

One of my favorite movie scenes is from the comedy Office Space about the miserable culture of Initech, a software development company.

When the company brings in two so-called “efficiency experts” to restructure its operations, the experts start by interviewing each employee to determine who essentially needs to be replaced.  One of their interviews focuses on Tom Smykowski, the long employed manager, who they expose as adding no value or even negative value to the company’s operations.

One of our earliest employees at Ubiquiti (and our head of hardware design) came up with a term I often use to describe these no value added positions.  He calls them “impedances.” And interestingly, they are everywhere.

At Ubiquiti, when we are in the planning phases of a new hardware design, we typically start with a selection process of integrated circuits vendors.  Usually we are self-sufficient enough to understand how to integrate vendor components into our designs; we just require basic application notes, pricing, and delivery information.  Unfortunately, when we inquire with a vendor for this information, our inquiry often sets-off a frustrating circus of middlemen — from marketing, sales people, and executives to 3^rd party distributors and representatives — wanting to meet and understand what we are trying to do.  We have little interest in these meetings; we only want delivery of the parts and pricing to proceed and get our product to market.  Sometimes these “Impedances” become so irritating, we will look for alternative vendors just to avoid them.

When it comes to our own sales, we take a polar opposite approach.  Instead of compartmentalizing sales with process, middle-men, and “information control”, we believe in complete transparency.  At Ubiquiti, we do not have sales people, marketing teams, or a bureaucratic process for disseminating market information.  We use what I like to call “applied social networking” where  we have over the years fostered a community of end-users that we freely allow to interact directly to each other and with our engineering teams.  This community uses various open forums on the Internet as well as our own official forum at forum.ubnt.com

I believe there are several benefits to this transparency:

1. Operations become more efficient:  Without the overhead of sales and marketing, operating expenses are significantly lowered.
2. Engineers become empowered when given direct customer access:  When engineers feel like they “own” their projects, they feel a deeper sense of responsibility to deliver a great user experience to the customers.
3. Brand Loyalty can increase:  Contrary to traditional thought, if the company makes mistakes out in the open, it is not necessarily fatal. The mistakes just need to be addressed quickly and if you can show a history of resolving them and improving, then a transparent market approach will eventually build end-user trust and loyalty in the brand.
4. Leverage expands:  As the “open” community grows, it can become self-supporting and self-growing.  In our case, Ubiquiti customers provide support to other Ubiquiti customers.  New product launches are virally marketed throughout the community and beyond.  And, new ideas and feedback are provided back to our R&D teams to help further evolve our platforms.

The key to succeeding with a “transparent” market strategy is to recognize that the customer is irreplaceable.  If the customer can be convinced that you exist to serve them and loyalty can be established, then a transparent market approach will become predatory to competitors with traditional sales strategies and operating models.

Background:

In 2001, the IEEE published the initial 802.16d Wimax fixed broadband wireless technology standard hoping it would bridge the digital divide amongst the underserved world.  Intel heavily backed Wimax and envisioned an interoperable technology standard supported by a consortium of chipset and system manufacturing companies that would eventually drive hardware price points down to levels capable of supporting global WiMax network deployments on a massive scale.  In short, Intel envisioned the WiMax standard to bridge the digital divide just as WiFi had universally unwired local area networks.

However, WiMax was very slow to pickup momentum due to its challenging cost model.  The starting infrastructure investment alone would require $100,000’s in capital expenses and this would not include costs associated to license expensive private frequency bands.  Typically, only large carriers were in a position to build out this infrastructure and many times struggled to come up with business cases to justify the investment.  Nevertheless, the WiMax movement was backed by billions of dollars and a close relationship strategy between hardware vendors and well-financed carriers.  The proponents believed costs would eventually come down to drive ubiquitous WiMax deployments in the same fashion as the WiFi model.  But, the costs would never decline enough and the vision of WiMax as the catalyst to solve the digital divide would never happen.

In sharp contrast to the disappointment of the WiMax initiative, Ubiquiti’s own AirMax platform, powering millions of subscriber deployments worldwide, was quickly succeeding in many places where WiMax had failed.  But Ubiquiti did not have large financial backing, carrier relationships, and sales teams.  Instead, we used a powerful R&D strategy to breakthrough traditional market inefficiencies, disrupt convention, and achieve tremendous sales “pull” from our organically grown grass roots community of entrepreneurial operators.

Specifically, our successful R&D strategy can be broken down to revolve around 4 platform design principles:

1. Initial investment cost must be low enough to allow for anyone to deploy the solution
2. The user experience must be intuitive enough to allow for anyone to operate the solution
3. The performance of the solution must meet or exceed existing enterprise and carrier quality expectations for functionality, performance, and reliability
4. Focus on making the solution cool — concentrate on aesthetics and features that promote end-user attachment and confidence

People often talk about the world of “machine-to-machine communications” and its growth potential looking ahead. We decided to apply our R&D strategy and disrupt it.   The result is what we call “Machine Fidelity” or “mFi” for short.  It’s one of the initiatives we have been working on for the past couple years at Ubiquiti and are excited to release today.

When Ubiquiti released the NanoStation “customer premise equipment” (CPE) in 2008, it changed the “Wireless Internet Service Provider” (WISP) market.  WISP’s (much like Ubiquiti) typically started as bootstrapped entrepreneurial ventures that were cash flow constrained.  With the introduction of the disruptive cost/performance NanoStation CPE, Ubiquiti provided a near zero time return on capital investment to the WISP operators and correspondingly greatly accelerated their network deployments worldwide.

At this time, Ubiquiti’s business was booming and highly profitable, but I was also scared to death because the business was still not defensible.  Although our new system CPE business was technically much more sophisticated than our original radio module hardware business, it was still based on the 802.11 WiFi standard.  And therefore, it could potentially be replaced in these growing WISP networks by a more cost-effective solution from a competitor.

Making standards based hardware is a dangerous position for a company to be in.   A great example from the past is the story of 3COM  — the pioneer of the original 802.3 Ethernet standard.  In its early years, 3COM was a monster company commanding very high margins for their network interface cards (NIC’s), which were used in nearly every networked PC during the 1990’s.  When I was a teenager I had a business setting up PC networks and would also use hundreds of 3Com’s popular 3C509 PCI cards to install into computers.  However, once the Ethernet 802.3 standard became more popular, new competitors entered the market supported by higher levels of hardware integration, and drove 3Com’s NIC business profitability into the ground.

I knew that our 802.11 WiFi based NanoStation CPE business would follow the same cycle of prosperity to commoditization that 3Com’s NIC business experienced if we did not correct the course quickly.  Fortunately, in our case, there was one key shortcoming of using the 802.11 WiFi standard for outdoor long-distance wireless that we were able to take advantage of in creating a defensible business.

The 802.11 WiFi standard is based on a contention protocol referred to as “Carrier Sense Multiple Access, Collision Avoidance” (CSMA/CA).  This type of protocol is very effective in an indoor environment where all clients are located in close proximity of the Access Point.  However, the WISP networks were not indoor networks; they were outdoor networks typically spanning many miles using directional antennas.  In this scenario, the clients have a direct line of sight to the Access Point, but become isolated “hidden nodes” to each other.  Consequently, each isolated client, unaware of the presence of their neighbors, assumes they are in the clear to transmit and their transmissions will collide with the other isolated clients at the remotely located Access Point.  As the network scales in clients, the collisions become exponentially worse.  Additionally, these same remotely located clients will often falsely interpret the presence of local transmissions from competing networks as the traffic of their real neighbors and at times turn to an idle state.  In short, the 802.11 wifi standard breaks down in these outdoor long-range networks as client capacity scales and is incapable of consistently supporting quality of service levels for the next-generation Internet applications including streaming video, VoIP, and gaming.

18 months after releasing NanoStation, Ubiquiti would solve these problems with our new AirMax technology platform.  It consisted of Access Points, antennas, and a new NanoStation CPE.   Although the hardware was still WiFi based, there was one very important difference: the protocol no longer relied on the 802.11 WiFi standard.  Instead, the AirMax protocol was based on a “Time Division MultipleAccess” (TDMA) scheme much like those used in Cellular Networks.  Unlike the 802.11 WiFi standard, the AirMax protocol was deterministic, resolved the hidden node problem, and allowed for far greater scalability, noise mitigation, and quality of service performance when compared with 802.11 standard equipment deployed in a long-distance outdoor environment.

Not only did AirMax bring important performance improvements to the WISP market, it marked the beginning of Ubiquiti’s defensible business model.  Now, when WISP’s built AirMax based networks, they were “locked-in.”  And as the AirMax market expanded, we would reinforce the strength of the platform by adding additional software features, management applications, and advanced products.  Following the Apple model, Airmax put us in a great position of leverage as we now controlled our own technology platform, and through our community, we would also control the end-user relationship.

In Part 2, I talked about Ubiquiti Networks’ beginnings that started with a flawed business strategy and eventually evolved to define a market.  Before I tell the follow-on story, I would like to preface it with a discussion of how I see two kinds of product development models:  “predictive” and “reactive.”

A predictive development model requires a mastery of market understanding combined with a team that has a demonstrated history of applicable market execution.  The gold standard for the predictive development model is Apple, Inc.  The iPhone for example combined a mastery of user experience understanding with a world-class development team that knew exactly how to bring the polished technology platform to market.  Apple invested in a very specific direction and they bet big.

Using vessels as an analogy, predictive strategies are like “super tankers” with destinations locked in place.  They are backed by a strong commitment to a pre-determined plan, with significant investment, and large, focused teams. And once the ship is set in motion, it is difficult to change direction.  But, when done right, they pack a giant punch and can fundamentally change markets like Apple did when they redefined the mobile industry overnight with iPhone.

The predictive model is also the basis of the venture capital world.  However, there is a difference between Apple and startup companies.  Where Apple has accumulated years of empirical experience (including many failures) in achieving a mastery of market understanding and development execution, new companies by definition do not have market experience.

From this point of view, the lack of funding or VC influence is actually one of the great advantages of bootstrapping.  Their absence forces the adoption of a “reactive” model with the goal to get something to market as quickly as possible to generate cash flow first and then react to the market challenges and feedback in refining the product strategy.  If a predictive model is analogous to a super-tanker, a reactive model is more like a speed-boat with agility to change directions quickly.

A reactive model will succeed when it is lead by an effective entrepreneur. At the core, this individual must be a tireless, disciplined, and creative problem-solver who is adaptable to market challenges and who can become incrementally more skilled with experience.

When Ubiquiti was faced with instant commoditization from lower-cost Asia competitor clones shortly after our initial radio module release, I quickly was forced to change development directions.  I knew we could never compete with Asia competitors in a hardware only module business; if we were to survive, we had to go upstream and compete with our system customers.  This would mean pulling together development resources that could deliver a complete solution to market that included investment into system hardware design, mechanical design, antenna design, and firmware development.  It would also mean offending a lot of our existing system solution customer base.

At that time (2006), the outdoor hi-power WiFi radio module market we helped to create was an integral part of the Wireless Internet Service Provider (WISP) industry.  WISP’s used these modules inside BaseStations and Customer Premise Equipment (CPE) to deliver broadband Internet access over many miles in rural and underserved communities throughout the world.   The CPE’s when placed on the roofs of homes and offices would receive the wireless signal from the Basestations many miles away.  It was the CPE cost that largely determined how fast the operator could realize a return on investment from adding a new subscriber.  Consequently, WISP’s were very interested in reducing their CPE costs to become more profitable.  At that time, Asia competitor cloned hi-power radio modules were selling for about $30 and when included into a complete CPE system, the system cost was roughly $150.  Our new goal was to transition from the module business and compete in the CPE market by introducing a purpose-built integrated CPE for under $150.

After 15 months of challenging development, we finally launched the PowerStation in May of 2007 and it was a miserable failure.  Although it was cost compelling for the U.S. market, we soon discovered it was not competitive in the international market.  Furthermore, it was too heavy for economical shipment, had several hardware issues affecting field performance, was missing key fundamental software features, had a poor installation experience, and it was poorly designed for volume manufacturing.

Two months after PowerStation launch (July of 2007), we learned our lessons, regrouped, and changed directions quickly again.  Only six months later (January of 2008) we would launch an entirely revamped CPE solution called NanoStation.  Unlike the fatal flaws of PowerStation, NanoStation was compact, with a clever cost-effective design, higher performance, optimized for volume manufacturing, and came with a much richer user experience.  Most
importantly, at a sub $60 price point, it stood to change the economics of the WISP industry and became a smash hit. Ubiquiti would finish the 2008 calendar year with $50mm in revenue lead by NanoStation.

The story of how Ubiquiti’s product strategy would evolve into the NanoStation design is an example of a “reactive” development model.  We were not experts when we entered the market, but after surviving long enough to accumulate experience, we were able to adapt to the market challenges and find success.

Fast-forward four years later to today, Ubiquiti is now completing its first predictive development effort in AirFiber.  Whereas the predictive model would not have been possible for us before, it now has become a viable model after several years of experience competing in our market.  The AirFiber story is one that combines a mastery of market understanding with a radio engineering dream team who has a history of development execution.

When a technology product is brought to market, it is typically the result of several partners working together.  For instance, an Android phone is the result of companies like HTC or Samsung productizing Google’s underlying mobile operating system platform and partnering with mobile carriers such as AT&T and Verizon who use their networks and stores to proliferate product sales to end users.

Android now accounts for the majority of the sales for the smart phone market, but the various Android hardware companies combined have far less profitability compared to Apple. Where Apple has been able to improve their strong financial fundamentals and profitability along with their legendary revenue growth, the Android smart phone vendors in general have seen their gross margins become increasingly under pressure.

So, the question is how can Apple protect their profitability and financial fundamentals when Android hardware vendors cannot?  It is because embedded in Apple’s business strategy is great leverage and protection against commoditization.

First, let’s take a deeper look at the Android world.  Android has evolved into a powerful technology platform.  Unfortunately, for the hardware companies, none of them own the platform.  As a hardware vendor, this is a big problem because if everyone has access to the same technology, there is by definition no competitive advantage.  And the result is inevitable commoditization — where gross margins become pressured as competitors are forced to compete on price.

When hardware vendors become commoditized, the balance of power shifts to the sales enablers.  In this case, Android hardware vendors become increasingly at the mercy of their carrier partners.  From a carrier point of view, hardware commoditization is great as it provides leverage to pit hardware vendors against each other and allows the carriers to differentiate themselves on the basis of their own value, such as network quality.  It also allows the carrier to “own” the end user relationship.  “Owning” the end-user relationship can be powerful as it creates long-term brand loyalty.

In sharp contrast, what Apple has been able to do in the smart phone market is remarkable.  Not only have they been able to own their technology platform and create significant competitive advantage, they have also succeeded in “owning” the end user through great brand marketing and use of their stores.  Correspondingly, they have commoditized the carriers.  Apple’s strategy has not only allowed them to maintain incredible profitability in the smart phone market, but by controlling the end-user relationship, they have also built a tremendously loyal user base they can leverage for future new technology market introductions.

On a side note — Recently, after watching Prometheus, I went back and watched the original Alien movie.  At one point, the crew of the ship turns to the Android Ash for an explanation for what the Alien is.  In which he says:

Ash: “You still don’t understand what you’re dealing with, do you? A Perfect Organism.”

Google has Android, but Apple just might be the Alien.  In terms of defensibility and leverage, the business model is perfection.