Monday, December 3, 2007


Great quote from the assistant dean of my business school:

"I think of my job description now as having two lines: one, to create structure for sustainability and growth and two, resist structure.”

Wednesday, November 28, 2007

What is this design thing, anyway?

I have been interested in design lately. For the last few years, I have thought of design as this subtle art of making things aesthetically pleasing. As in "interior design". Now, my interior design skills are limited to my recent realization that, one: I like walls that are painted a color other than white, and two: crown molding looks kind-of cool. Other than that, I am completely lost.

So when people talk about the importance of design in consumer products, I can definitely tell what I like, but I have no idea how to do it myself. I feel like this is important, but I feel helpless, so I feel frustrated. I look at products from Apple, and I agree that they are beautifully designed, but I cannot explain why. I read websites about design, and apparently the beauty is because this corner is rounded, and the other corner is square, and the bright-white power cord creates a cognitive dissonance. I really, really, don't get it.

Then I read this article in Fast Company about how all of Apple's products are flawed, and suddenly, I got it. I am not saying that their products are not flawed - I use several Apple products on a daily basis, and know exactly what I am getting and what I am giving up with this choice. I am saying that the description of what was wrong with the iPhone completely misses the point of design. In fact it missed the point in such an obvious way that it helped shine a light on exactly what good design is.

Design is not the art of making a corner round or square, or picking exactly the right color. Well, it is, but that is a small part of design.

Michael Fitzgerald thinks that the iPhone is a pretty box, but that it is missing a bunch of features, and that pretty boxes are easy to copy. "Fashionistas" (who apparently only care about the pretty box) will buy the iPhone for its looks, but the general population wants all those features. He lists a whole bunch of features that could be in the phone, but aren't, and labels them as fatal design flaws. And this is where design enters the picture.

A well designed product is not a bunch of features thrown into a box. For every feature available, there is a clear trend toward "better". For memory, battery life, network bandwidth, and free beer, more is always better. For weight and cost, less is generally better. We know this already. Just saying that there is a technology available that is better than the one we have is not the same as designing a better product.

Early in my career, a senior colleague told me that "Engineering is fundamentally about compromise". I didn't like this when I first heard it, but I quickly understood that he was right. Faster processors cost more money, consume more battery power, and generate more heat. Getting rid of the heat requires bigger heat-sinks (more size and weight), and / or more active cooling (more size, weight, and power). The bigger batteries demanded by the bigger processor and its cooling system take up size and weight, cost more, and require a bigger, more expensive charger. Aside from the functional requirements (what the gizmo being designed is supposed to do), a designer is typically working within power, weight, and physical envelope (size and shape) budgets.

Design is about making the trade-offs between faster this, slower that, and heavier overall package. I wouldn't be at all surprised to hear that Apple designers sweated over decisions that would have made the iPhone one or two millimeters thicker, or a few grams heavier.

The user interface and the aesthetics of the phone are part of these decisions. Apple didn't think that the iPhone virtual keyboard is better than a physical keyboard, they just thought that it was a worth while trade-off to be able to hide the keyboard when you didn't need it (and without making the phone twice as thick so it could slide). They could have made the iPhone bigger and uglier, and harder to use, so that they could fit in all the biggest, fastest features on the market, but lots of people already make that phone.

The definition of a disruptive technology is that it is inferior to the state-of-the art. Telephone calls made over the Internet (VOIP) were, and perhaps still are inferior to those made over traditional circuit switched networks. But we love them, because they add convenience, are low cost, and can integrate with other technologies much better than my desk phone can. Disruptive technologies are disruptive because they change the competition away from the old metrics (e.g. fastest network access or typing speed), and begin competing on new metrics (e.g. easiest to use, or tightest integration). Kodak went from being a company with 90+ years of industry dominance to nearly irrelevant when we removed film from our cameras. Digital cameras had inferior image quality compared to Kodak film, but when was the last time you bought a roll of film?

So what is design? It is about creating a complete solution to a customer's problem. It is about making the trade-offs between all the possible features, and making the selected features work perfectly together. Design is like cooking: The best apples don't guarantee the best apple pie, but a great chef can make a great desert, even with average ingredients.

Tuesday, November 20, 2007

Best paragraph ever

"A few years ago, the company I work for decided that I was such a good programmer that I had to stop doing it immediately. I was now to tell other programmers what to do, using all the social delicacy and interpersonal self-confidence I’d built up over two decades sitting in a dark room and staring at a monitor. I was being promoted into management."

Full posting

Tuesday, November 13, 2007

Hoping for better from Harvard

From Harvard Business School prof. John Quelch, talking about the iPhone launch:

First, the reviews were mixed. At that retail price and with that level of hype, the critics were going to be tough but a raft of concerns from delayed activation and sluggish email (AT&T's responsibility as the exclusive network provider) to feature shortfalls began to dampen marketplace enthusiasm.

I don't really care whether professor Quelch thinks the iPhone is a good phone or a bad phone, or even if it is good or bad for any particular use-case. Instead, I am pretty annoyed at his sloppy analysis of the iPhone launch. My first criticism is about language, and his use of the word "hype" regarding the marketing activities surrounding the launch. I am perfectly fine with his use of the word "promote" (earlier in his article), but nearly all of the definitions of "hype" suggest some form of trickery, deception, or dubious methods. When I hear "hype", I think "hyperbole", which is an "obvious and intentional exaggeration". I don't think I am picking nits here, because the conclusion of his article points directly to this:

Hype can hurt stock prices and investor confidence when expectations are not met.

Almost every product is advertised with this "obvious and intentional exaggeration", and I find it tiring, even though I understand the reasons. If I pay attention to an ad, I am looking for clues as to whether the product being advertised might be suitable for my needs; when hyperbole is routinely used, I cannot tell.

The striking thing about the iPhone ads was that they engaged in no hyperbole whatsoever. They showed someone using the phone to perform a series of tasks, in real time. There wasn't a cut, or a special effect, or a "your mileage may vary". In fact, any iPhone user could duplicate any of the ads on their own, with no special training. When you perform an experiment that anyone, anywhere can reproduce and get the same results, this is called science. To me, this is the antithesis of hype.

In the same spirit, and moving on the the rest of my complaint with Prof. Quelch, I don't think there was a single feature or capability that was promised that was not delivered. Anyone who was unhappy with their iPhone shouldn't have been surprised. We routinely expect companies to mislead us about the performance of their product - usually around measures that are hard to pin down, like gas mileage or battery life. But even in this area, I hear no one complaining that the iPhone doesn't live up to its published specifications. All the complaining I heard was that the iPhone didn't have a feature that the complainer wished it had.

Just to stay on track, my points so far are that 1: the iPhone was not hyped, and 2: the iPhone did not disappoint expectations.

As for the price cut - every cell phone ever made (that may be a bit of hyperbole) has gone through the "rapid skimming" marketing model, where it is launched at a high price to capture high margins from early adopters, and then is reduced in price to capture additional revenue farther down the demand curve. The only unusual thing about the iPhone was that the price cut came so soon after launch. This might mean that the initial price was set too high, or it might mean that Apple and AT&T sold so many phones into the early adoptor market in the first few weeks that they could now target more price sensitive buyers.

The other thing that was unusual about this situation is that normally, a cell phone manufacturer would come out with a new phone, which would push down the price of the previous market leader. If Apple had done this, people still might have been surprised by how fast it happened, but not surprised at all that the old phone had to be discounted in the face of the newer phone. The unusual thing is that the iPod Touch, which is not a cell phone, is still a competitor to the iPhone. Many customers who were considering a iPhone might switch to the new iPod Touch. Apple probably made some rational calculations about how much an iPhone customer was worth to them compared to an iPod Touch customer, and adjusted the price of the phone to match. When new products are released into the market, the demand curve for the old product shifts.

Apple stock was at $144 on September 4, the day before the launch of the iPod Touch. It did dip on the launch news, but by September 25, it was $153. That is a 6.25% increase in 20 days. Annualized, that is equivalent to a 114% increase. The NASDAQ rose less than 4% during that same period. I won't go into the calculation here, but even accounting for this increase, and Apple's sensitivity to the market (beta), this is a real gain by Apple.

If I were doing an event study on this, my conclusion would probably be that the market reacted to the news as if it were bad news, then over the next couple of weeks figured out that it was actually good news. Benjamin Graham used to say that in the short term, the stock market is a voting machine; but in the long term it is a weighing machine.

So let's revisit our good professor's conclusion:

Hype can hurt stock prices and investor confidence when expectations are not met.

My assessment is that Apple did not hype their product, did not fail to meet expectations, and except for a short-term blip (less than 20 days), did not have their stock price or investor confidence hurt. The truth is, I agree with professor Quelch - hype is a dangerous thing. I just think he picked a horrible example in Apple and the iPhone launch. And from Harvard, I was hoping for better.

Wednesday, November 7, 2007

10,000 hours

Apparently, it takes about 10,000 hours of study to master any complex subject. But, with that study, and a few friends, you can solve problems that are so complex that they border on impossible. In this video, Malcolm Gladwell argues that a dozen disciplined, smart people are more valuable today that one bona-fide genius.

In your job, you might be able to focus 6 hours per day on an important area of expertise, yielding 30 - 40 hours per week of focused effort. Outside of work, you might be able to devote 20 - 30 hours per week on a serious hobby. Of course this will vary for each person, but it seems likely that most people could devote between 20 and 40 hours per week to the study of a particular subject, and still be a normal person. At that rate, it takes between 5 and 10 years to really master a subject. (Mr. Gladwell makes the same estimate in his presentation.) If you could get a job that aligned with your interest, so you could spend a significant part of your work day learning more about it, and you spend some of your off-hours working on the same subject, you might get 50 hours per week of focused study - resulting in 10,000 hours of study after only 4 years.

4 or 5 or 10 years might seem like a long time. It is a long time. My 8 year-old little girl will be going to college in 10 years (eek!). But the time will pass, regardless. And at the end of that long 4 or 5 or 10 years, you can be a world-class master of something. Or you can look back at that long time and wonder where it went.

What are you going to master in your next 10,000 hours?

Thursday, November 1, 2007

It doesn't matter what you think

In an article for Fast Company, Zach Nelson (CEO of NetSuite), talks about the worderfullness of the iPhone. I don't know about you, but I followed the iPhone frenzy pretty closely since the January announcement, and I was surprised by all the strong opinions, and even more surprised about all the wrong opinions.

It seemed to me that the fastest way to get noticed in a blog, or in a magazine, or on TV, was to find something wrong with the iPhone. You didn't have to have experienced a problem (since almost no one had actually, you know, touched the new phone yet), and you could rule out entire classes of users (like people with jobs) based on how you imagined yet another group of people (like IT managers) would feel. For example, a common theme was that because the iPhone didn't work exactly like a RIM Blackberry, it could not, would not, and should not be used by any business person, ever. Or else. Don't say we didn't warn you.

Look, I used to be an IT manager, and I completely understand the desire to have things running in a nice, controlled, orderly way. I had lots of conversations with users about what technology they wanted to use, and whether that fit into our plan. The trouble was that every time I told someone they couldn't use something with our network, I was making my job easier, not theirs. And the trouble with that was that they were probably generating revenue for the company, and I wasn't.

This wasn't supposed to be an article about the iPhone, or about IT managers, or about bloggers who complain about technology they don't like. It was supposed to be about Zach's insight into the iPhone, which was really about new technology in general. So maybe I should get right to the point.

It doesn't matter what you think. It doesn't matter if you think the iPhone is good or bad for business users. Even if you are an IT manager. Even if you are the CIO. Even if you are the CEO. Not even if you are very well respected and widely read journalist.

If you still aren't convinced, just wait, and the decision will be made for you by your best and brightest new hires. Never lose sight of what the college students of today are accustomed to.
Zach Nelson, from the FC article.

Right now, college students might seem unimportant in the grander scheme of things. But next year, they (as a group) will be working in every segment of the economy. And after a couple of years, they will be doing most of the work. And a few years after that, they will be running the business. The communication tools and techniques they are getting used to now, will define how we work over the next decade.

The serious business phone - the Blackberry, is interesting because it does email so well. The Blackberry exploded in the late 1990s because of all the people who got hooked on communicating over the new medium of email in the early 1990s. The real power of the Blackberry is it's integration with the Microsoft Exchange corporate mail server. The iPhone doesn't do this integration very well; instead, it is much better at browsing the web than a Blackberry (or any other pocket device). Given that the biggest threat to Microsoft right now seems to be Google, and since Google's mail system uses a web browser, where do you think things will be in 10 years? Still not sure? Ask a newly hired college graduate which mail system he prefers.

Tuesday, October 2, 2007

Good Question

ars technica wrote an article yesterday that basically says Nintendo will not be able to ship its popular Wii game console fast enough to meet Christmas demand. John Gruber asks "What is wrong with Nintendo?"

This is a good question. In business, the problem generally is to find or create a product that customers are willing to pay more for than it costs you to make and deliver. Oh yeah, it also has to be better or cheaper than your competitor's product. In the simplest examples, we think about how much the parts costs, slap a little margin on for development costs, and maybe a little profit, and decide that is how much the product should cost.

In business school, we dig into this quite a bit, and find that there are lots of ways to optimize your business so it is more efficient, which translates into a lower price (and more sales), or to higher margins. Since there are lots of ways to improve efficiency, it follows that there are an equal number of ways to screw up efficiency, thus increasing the price (and reducing sales) or reducing margins.

With that in mind, here is a quote from the ars technica article:

"They ought to stop by on a frequent basis. We're going to flow hardware. It's not that it's going to show up only on one occasion. It's going to be constantly flowing in."
Nintendo of America President Reggie Fil-Aime
The phrase "constantly flowing in" stands out to me. Toy manufacturers have a classic problem related to the Christmas buying season. First, their revenues all come at one time during the year. If Nintendo sells 50% of their annual demand for their game console in one month, this means they will sell very little during all the other months. Second, the factory has to deal with this same spike.

Although this is a supply chain problem, lets talk about it in terms of our own lives. Imagine that you have a large extended family that includes 12 other households. That is, there are 12 other families that might want to come visit you in your house. Now you love your family, and you want them to stay with you when they visit, so you buy a house with a guest room. If they each visit one month per year, you are all set. But what if they all want to visit during the holidays? Should you build a house with 12 guest rooms, and leave them empty for 11 months out of the year? That is certainly not an efficient way to entertain relatives, and it is not an efficient way to run a business.

Nintendo is just saying that they cannot run their business efficiently by designing it to meet seasonal spikes. "Well, since you can never find a Wii on the shelf now, surely they can increase production another 10% or so." Another good point. But to do this, Nintendo will have to invest money -maybe a lot of money. For example, what if they are currently running at full capacity on one manufacturing plant, and to increase just a little, they will have to bring another plant on-line?

In any case, any change will probably require Nintendo to invest some money, and the first question to ask is "will we get that investment back?" There is lots of uncertainty in consumer goods - especially toys, but I am pretty sure that Nintendo has thought about this, and they are thinking that retooling their business for last-minute shoppers is not a good investment.

Doing things the Hard Way

I like to do things the hard way. Well, it's not so much that I like it, but after I have done something, it usually turns out to be the hard way. Part of the problem is that I have a pathological need to understand how things work. For example, in high school, I could never memorize the the rule-based procedures for solving a problem. Instead, I tried to understand the relationships at work, and then re-create the method as needed. This was cool for explaining the concept to the girls in the class, but it was murder on exam day. Eventually I realized that you just don't have time to reinvent the wheel on every problem. Still, I was never satisfied to have memorized a procedure that would produce the correct answer, without understanding all the moving parts.

A few years ago, I realized that this problem shows up in other areas of my life. I know that most men don't like to read instructions, but apparently I take this to an extreme. I didn't go to college right out of high school for a number of reasons, some good, some not so much. (As I get older, I move more of my reasons from the former category into the latter.) What I eventually realized was that deep in the dark crevasses of my subconscious, I believed that taking a class was an admission of failure. If I was really as smart as I thought I was, I should be able to figure things out for myself. You know, stuff like chemistry, and physics, and solid-state semiconductor theory. I'm not kidding. I worked on military and space-grade electronic systems for nearly 10 years with nothing but an associates degree. And not one of those trade-school degrees where they actually taught me some technical stuff - no, I studied Anthropology, Ecology, Psychology, History, and English. Sure, I took about 3 classes from the Electronics department, and a bunch of math, but half the math was to make up for what I didn't do properly in high school, and the other half was only sort-of useful.

From a fundamental perspective, Calculus was cool, but I am pretty sure that I have never had to calculate an integral outside of the classroom.

The funny thing is, most of my colleagues consider me to be a pretty hard-core engineer. Anyway, after 20 years of reasonable success in my career, I decided that maybe there was something to this whole idea of letting a teacher show me how to do something, instead of reinventing it for myself. So I found a reasonably good university that didn't laugh too hard at my previous attempt at an education, and managed to earn a BA degree. Of course, after significant work experience, a BA degree seems like pretty entry-level stuff, so my pathology of having to understand how things really work kicked in, and forced me into graduate school.

All of this is just a long-winded way of explaining how an uneducated, stubborn guy like me finds himself in top-tier MBA program. Some of the comments in this blog will come from the uneducated, stubborn guy, and some will come from a seasoned professional with a top-notch education. Which comments are which will be left a an exercise for the reader.