I make it a practice not to spend a lot of time in these pages on my personal interests and pursuits. More than occasionally, something related to the general brief covered by Going Private will overlap with my overly broad spectrum of interests, but I don't spend a lot of time on such things in the abstract. Be this as it may, occasionally, I can't help myself.
All of this is a round about way of saying that I have finally grown to understand one of my guilty pleasures: Apple. In doing so, I have managed to focus some attention on some structural issues that have been, at least subconsciously, bothering me for some time.
I admit, grudgingly, to owing an iPhone. I bought it in time (that is to say foolish-early enough) to have Mr. Jobs give me one of those $100 "Sorry I screwed you on the price" rebates. I also have to admit to owning a top of the line 15" MacBook Pro (I bought after the latest upgrades- watching a colleague run Excel on a Windows XP instance isolated from the rest of his system and displayed in a window on his OS X desktop was the last straw). Before that, I had two white, Intel MacBooks. This is unusual, because three and a half years ago, I never would have bought Apple. This last, that new Apple owners almost find themselves surprised to own an Apple, seems a common trait among the, admittedly small, sample of Apple owners I have encountered. I've watched four or five people who swore they would never own an iPhone give in, buy them and proclaim, in such similar tones one wonders if The Amazing Alexander works for Apple now ("I loved it. It's much better than PC. I am going to buy it again, and again and again..."), that it is the best phone they have ever owned. And this is where I began to wonder, why the near epiphany in reaction? Now I think I know.
Why this reluctantly amazed reaction among iPhone buyers?
Because the definition of "phone" created by hardware and network providers today is so limited.
Why have so many Macintosh buyers had the same reaction?
Because the conventional definition of "laptop" or "operating system" or "computer" created by hardware and software providers today is so limited.
Because these companies hate consumers, hate their desires, hate their needs and, consequently, make sure that the conventional definition of, e.g., "laptop," or "phone" is very limited.
Why is this? I blame Michael Eugene Porter. Not that Porter is a dipstick, (well not only that) but because the majority of his modern adherents certainly are.
The eager and almost rabid application of Porter's "Five Forces" (Supplier Power, Customer Power, Threat of New Entrants, Threat of Substitute Products, Industry Rivalry) to technology products and services has bred an entire generation of MBAs in marketing positions dedicated to developing and maintaining closed systems and closed hardware platforms. This is particularly egregious in the case of business models that are effectively based on distribution channels. In conventional analysis there is nothing wrong with making your living on distribution channels. Remember, that in 1979, when Porter developed the Five Forces framework, distribution channels were highly expensive to create and maintain and, owing to these costs, constructing them effectively presented a significant barrier to entry. Your product didn't even have to be particularly good, because the threat of substitutes was reduced via the difficulty and expense of the competition actually getting those substitutes (however good they might be) to your customers. Suppliers, if they wanted access to your customer base as a proxy to sell their raw materials, had to go through you. New entrants had to build an entirely new distribution channel. Customers were stuck. You owned the market. But you had to guard this distribution channel carefully. And you had to make sure you hadn't forgotten something simple and critical. That's not part of a conventional Porter analysis. But why would it be? Conventional distribution channels are quite physical, antique and boring.
If you want an example of how dependent firms have become on distribution channel dominance as a strategy (and how lethal this reliance can be) consider the example of Blockbuster. All Blockbuster does today is provide, at great expense, an elaborate distribution channel to deliver very cheap plastic discs with expensive data on them to every neighborhood in every major city and town in the country. As soon as DVDs became the predominant data format, Blockbuster became nothing more than a highly expensive, slow, ultra-high latency internet with a data warehouse limited by inventory practicalities. Of course, DVDs did eliminate the hated "rewind charge," but that's another story. All Blockbuster is really doing is delivering digital data. Poorly.
But this isn't a story about how Blockbuster is going to be destroyed by delivering video data over the internet. It is much simpler than that, and it is simpler because Blockbuster doesn't actually add much value. In fact, beyond their delivery channel they provide no real value at all. Oh, sure, you can get microwave popcorn, Raisinets and Ju-Ju Bees at the counter after the third fight with your significant other in as many minutes about which chick flick not to get because last time you watched The Break Up and the first 5 episodes of Season 2 of Sex in the City without complaining even once (she won't be impressed no matter how quiet you were, trust me), but that's about it.
The thing is, on the strength of a Porter analysis, Blockbuster was a complete no-brainer in 1996. Hollywood Video, the only competition, was a distant second which primarily distinguished itself from Blockbuster by carrying adult videos in contrast to Blockbuster's "family experience" focus (I will leave it to you to decide which was more likely to pay dividends in the long term). The Blockbuster brand was a huge barrier to entry, and the fragmented mom and pop video stores were getting consumed rapidly. Supplier power was low because if you wanted that post-release revenue (which, increasingly, is a huge part of movie production) you had to deal with Blockbuster. Building something like Blockbuster's store network was next to impossible to do from scratch. If you were a firm looking to enter the market for video rental in the late 1990s, and you had an MBA with any authority on your management team, you would have picked another business instead because he would have performed a Porter analysis and told you to just forget about it.
Marc Randolph and Reed Hastings didn't, I suspect, have such an MBA on their team, or they saw how worthless he was, and instead thought about the problem hard enough to realize that, while they couldn't build an expensive distribution network, they didn't have to. The United States already had in the form of the United States Postal Service. In addition, Blockbuster never had to be customer centric. They never had to develop much of a product, or pricing strategy owing to their distribution dominance. Randolph and Hastings' creation, Netflix, could easily beat Blockbuster for customer satisfaction (this was a low bar), and they could do so cheaper (Blockbuster spends $1.5 billion annually on its stores. Netflix spends around $300 million on postage). They also quickly realized what the eventual pricing fate of distribution based business models would be and applied it almost immediately: commodity pricing. In distribution networks, this means flat-fee revenue models. Period.
The result? Blockbuster's revenue and Gross profits are basically the same today same as they were in 2002.
You find a lot of these twisted business models in distribution-only businesses. Hollywood video was eventually bought by Movie Gallery, which, I suspect as a result of some student of distribution power analysis, also decided to put tanning beds in some locations. This is backwards in more than one way. Aside from the obvious (where is the overlap in these services?) the most likely tanning consumers (women) are the least likely to have a lot of free time at the video store. (I recently read a study that indicates that women spend 400% more time browsing in video stores than men do). Even the most obtuse analysis tells us this is a dumb idea. Perhaps one of the dumber ideas... ever. This, dear friends, is the logic of distribution analysis. Blockbuster explicitly banned their franchisees from following suit, and the amazing part of that fact is that franchisees actually wanted to.
Movie Gallery emerged from Chapter 11 recently, crippled by the debt assumed from buying Hollywood video. The Deal summarizes:
With profits in the DVD rental sector harder to come by over the last couple of years, Movie Gallery boldly went ahead and purchased Hollywood Entertainment in 2005 for $1.2 billion, which included the assumption of $350 million in debt. The acquisition catapulted Movie Gallery to the No. 2 slot among movie rental companies in the U.S., behind Blockbuster Inc. But stiff competition for the entertainment dollar from rivals and cable proved too much for Movie Gallery.
This desperate, last minute clawing to realize revenue out of expensive and, increasingly, less profitable retail distribution networks should set off loud alarm bells in typically astute Going Private readers whenever it appears. This is, of course, distinguished from the strong overlapping synergies of a strong distribution leveraging strategy. FedEx-Kinkos is an example of the later, Blockbuster-Circuit City, the former.
Run the way-forward machine (sounds like a Tory convention platform, no?) ahead 30 years and distribution channels for products like software and the like are infinitely cheaper. Suddenly, one has to find other ways to reduce customer power, the threat of substitutes and the threat of new entrants. Like create a real product that consumers actually like. Imagine that.
Coming full circle, we arrive at the business model for every telecom company on the planet. The amazing part to me is that, almost De Beers like, wireless providers have managed to somehow maintain the fiction that they maintain massive proprietary networks, and price data transmission accordingly.
$0.10 per SMS at 160 characters per message works out to around $1,000 per megabyte. This is easily 100x the revenue extracted from shipping TCP/IP data over "cellular networks," with the most expensive providers, and yet, functionally, SMS is vastly inferior to TCP/IP instant messaging services on almost all counts. SMS is limited by message size, media type, and it is high-latency. This triumph of premium pricing over a total mismatch in supply-demand dynamics for the functionality makes SMS one of the great marketing accomplishments of the century (hence my De Beers reference). Of course, the distinction between "great marketing accomplishment" and "consumer fraud" can be quite thin, but, then, this is the essence of Porter and shades of interpretation at the FTC.
The hallmark of (over)extended Porter analysis in distribution-centric business models is this kind of restrictive, closed structure, and, as the distribution network's dominance is increasingly threatened, the advent of "use restrictions." These two approaches, technical restrictions and contractual use restrictions, represent the two last-ditch steps before total commodity pricing of distribution takes hold.
A classic example of this progression can be seen in the ongoing attempts of internet service providers to maintain price differentiation between business and residential users. This first took the form of "aDSL/sDSL" (asynchronous v. synchronous digital subscriber line) which originally had its origin in the longer loop length aDSL (and therefore the larger consumer base) the technology could provide. That advantage has long since reversed and aDSL now is effectively an artificial technical restriction, keeping upstream bandwidth speeds low and allowing internet service providers to charge business users more for sDSL, which is, in effect, the cheaper, better technology. To some extent this differential has been eroded, forcing providers to provide the same technology solutions and employ "use restrictions" based on "terms of service" agreements.
Comcast, historically, is a severe offender here, and their approach characterizes the efforts of large internet service providers. In a few words: Box in the customer. Consider their service agreement text:
You agree that the Services and the Comcast Equipment will be used only by you and the members of your immediate household living with you at the same address and only for personal, residential, non-commercial purposes, unless otherwise specifically authorized by us in writing.
The "non-commercial purposes" requirement would, at first blush, be violated by anyone with a home office. As is typical, Comcast adds more restrictions in their "acceptable use" agreement:
[You may not] use or run dedicated, stand-alone equipment or servers from the Premises that provide network content or any other services to anyone outside of your Premises local area network ("Premises LAN"), also commonly referred to as public services or servers. Examples of prohibited equipment and servers include, but are not limited to, e-mail, Web hosting, file sharing, and proxy services and servers;
Therefore, Comcast reserves the right to suspend or terminate Service accounts where bandwidth consumption is not characteristic of a typical residential user of the Service as determined by the company in its sole discretion.
Read: "You can have all the bandwidth you paid for, until we say you can't."
I doubt anyone really believes that a single resident customer with 3.0 or 6.0 Megabits of bandwidth is going to swamp Comcast's network. Nor, frankly, are all the customers maxing their bandwidth 24/7. Bandwidth is cheap. VERY cheap. This is an artificial technical restriction. But, Comcast also does much more. They do a lot of "bandwidth shaping," that is, evaluating your traffic and if it looks as if it is, for instance, related to BitTorrent (legitimate use or otherwise) choking it down to the point where the service is unusable.
Again, no one contends that BitTorrent use is going to swamp Comcast and degrade their network (if so, stop selling 6.0 megabit packages to residential customers- oh, you can't because the market would kill you? Gee, that sucks), nor does it seem likely, given the extensive indemnification clauses and the current state of communications jurisprudence, that Comcast is likely to be hit with liability for clients that, for instance, break copyright laws on Comcast networks. (In fact, by exercising bandwidth shaping, they start to appear to be examining customer traffic and exercising editorial control, which means that they are inching closer to liability for customer activities). That would make sense if these restrictions were about any of that. But, of course, they aren't. They are, instead, the culmination of the struggle to stave off commodity pricing of a distribution network via extended Porter analysis. And this is another clue to the astute Going Private reader: The explanations for the technical and use restrictions don't stand up to even moderate scrutiny.
Like any distribution-centric business, as wireless provider, you can restrict use in the same two ways. First, close your hardware platform, forbid extensive third party software development and lock users into your applications, and therefore, your built-in use restrictions (cast as technical restrictions). Of course, there is a balance here, playing the importance to consumers of hardware design innovation in cellphones against the resistance of the telecoms to opening use any further than they dare. If you license third party hardware, you do so in the context of very strict limitations on agreements, up to and including final form factor design approval, feature approval, application approval or outright development restrictions that forbid the hardware designers from writing code not approved by the network provider. This was Nokia's model for years. Sprint was even worse. Second, you impose contractual use restrictions. Verizon experimented with these for a time when they started offering "unlimited data," but then using using "bandwidth consumption" clauses to terminate customers without warning for, e.g., downloading video. (The New York Attorney General sued, Verizon settled, and has since started the much more open "any app, any device" as a business model- I suspect the iPhone had some impact here). The problem is that, in the end, you are just shipping bits. It is just distribution. And it is, in effect, very cheap.
Apple, I believe, has evaluated these markets, and, being a highly consumer-focused organization, endeavored to break the cycle and give consumers what they really want. The success of the iPod and iTunes is based more on breaking arbitrary restrictions on consumers than anything else. Jobs has almost done for music what Netflix did for video rental. The process is simple. Admit that commodity pricing is on the horizon and, rather than cling to old models, simply implement it. We aren't quite at the point where iTunes is a flat-fee per month for unlimited downloads, but a fixed per-song price is an amazing advance considering the artificial technical restrictions the music industry has imposed. And at least here we are paying for content, not for distribution. I don't have a good dollar figure on the negative revenue effect to the industry implicit in finally permitting a consumer to buy a single song for $0.99 instead of $14.99 for the entire album, but I suspect it to be somewhat significant.
Digital Rights Management, Trusted Computing Platforms and the like are artificial technical restrictions designed to prevent the commodity pricing model in music. Suing your customers pursuant to use agreements is not much better. We'll see how those work. I have my guesses.
Enter the iPhone. Apple aligned with AT&T, I suspect, because they provided the cheapest network solution Apple could negotiate quickly. They certainly did not have the best network (EDGE is no 3G), or the best coverage, but I personally doubt Apple intended to stick with AT&T for long, and it was a good "proof of concept" deal to see how the iPhone sold before committing to a more committed strategy. Apple, unlike most firms, actually gets Porter, understands it as a guide to offensive as opposed to defensive market strategy, i.e., build a base in hardware, users loyalty, and then use that to push around network providers. And that is just one aspect of supplier power that Apple understands to the core. This is the same strategy that has made iTunes a powerhouse and able to mostly dictate pricing terms to the big labels. AT&T was used to dictating terms to hardware providers, including design criteria. (Nokia and AT&T had any number of bitter battles, none of which Nokia won). Some of the terms reportedly included: No uncertified third party applications, locked access to AT&T networks only, no user-modifiable hardware or firmware, and reasonable efforts by Apple not to permit modification of the iPhone by users. Clever Going Private readers will immediately recognize AT&T's justification "we need to protect our network" as an absurd rationalization for these restrictions and therefore pattern behavior in the genre of distribution protectionism.
Personally, and as a matter of opinion, I find it hard to imagine that Apple has done much beyond the minimum required by contract to prevent third party modification and development. Further, and again, just one girl's opinion, I believe this quite intentional. Several examples leap to mind:
1. The iPhone is not particularly secure from end users as a platform (if you caught the irony in the the use of the phrase "secure from end users" in the context of a consumer product, you are likely a regular reader). At least in the most recent versions, all applications run on the operating system as "root."
2. Though the iPhone's root password is changed with each firmware update, it consistently violates every precept of strong password generation in existence (simple six letter, all lowercase words that I will not reproduce here). While the root password is not the end-all be-all of iPhone security, it certainly reflects a particular security complacency from a firm that absolutely knows better. Why is that?
3. It would be trivial for Apple to identify "jailbroken" iPhones and "brick" them. Apple has made it easy for sophisticated users to mask their naughty behavior and still get updates to iPhone firmware.
4. While the iPhone ships locked to AT&T's network, it is trivial to unlock the device. As a side note, to the extent Going Private readers place any credence at all in equity analyst reports, it might be a useful exercise to go back and find those analysts that expressed dismay and confusion that the number of AT&T activations of iPhones lagged Apple's reported sales of the devices by 1.5 million or so. Any number of explanations (hoarding by AT&T, fraudulent inflation of sales figures by Apple) offered by the best and the brightest equity analysis had to offer could have been immediately disabused with a 60 minute commitment to standing in front of an Apple store and simply asking iPhone buyers what they were doing with the devices. Those few (new) Going Private readers that had any faith in equity analysts will now be reevaluating their confidence therein.
But all this is very much like Apple. Those users that really want to fill their iPhone with useful applications (once I discovered you could put an HP 12c on your iPhone, my fate was sealed) can get away with it without much effort and without blowing up their phone. Suddenly, rather than needing a network provider to subsidize the hardware costs because consumers won't even pay $100 for a chunk of Nokia fertilizer, <s>$600</s>$500 out of your own pocket doesn't seem crazy for a phone. It was only crazy back when you thought you were just buying a phone. Or, should I say, what passed for a phone.
If Apple follows the experiment to its logical conclusion, we should see them buying a wireless provider one of these days. Or building one. Flat fee for (true) unlimited data couldn't be far behind. Could it?
Why do I love Apple, despite all peer pressure and conventional wisdom? They understand that superior products kill three porter birds with one stone. They eliminate substitutes, increase development costs (and therefore reduce new entry threats) and decrease supplier power. Locking consumers into a closed system only gets you one, and pisses your revenue source off at the same time.
Why do I love Apple? They intend to make money because of my desires, not despite them.