Can MBA Programs Learn from Yoga?

Here’s an article in the NYT Business section on Harvard Business School’s efforts to create an online offering without eroding the value of their traditional MBA program. The central hitch (aside from the irony of a Business school struggling with a business plan, I guess) is that the MOOC approach, where thousands of people watch the same lecture online, doesn’t work for HBS’s discussion-heavy case study education format. So they’re attempting an elaborate, custom-designed video setup where the professor can create virtual case study discussions. Unsurprisingly, HBS professor Clay Christensen is skeptical:

Professor Christensen, for one, worried that Harvard was falling into the very trap he had laid out in “The Innovator’s Dilemma.” “I think that we’ve way overshot the needs of customers,” he said. “I worry that we’re a little too technologically ambitious.”

What if technology weren’t the answer?  If the case study methodology relies on small discussions, then it seems that the most logical way to grow would be to increase the number of people who are qualified to lead those instructions.  And maybe make it so all those people didn’t have to live in Cambridge.

What I have in mind is a train-the-trainer model akin to what you find in many professional certification programs, or martial arts or yoga study.  There’s a methodology, and the experts in that methodology train others to teach it.  Then the instructors continue their training over the course of their lives, getting better and better and teaching more people.  The “gurus” might all be in Cambridge, but you’d have experts in every major city offering HBS-branded education.

The internet turns out to be a great complement to these sorts of programs, allowing students to locate the expert in their city, or find out about trainings going on in other cities they might like to attend.  Maybe there are week-long intensives on HBS campus.

Eventually, HBS might even phase out the two-year MBA in favor of continuous lifetime learning.  Aren’t we deluding ourselves in thinking that one can become a “master” in two years anyway? True mastery should take much longer.

Disrupting Amazon

I was catching up on the excellent Critical Path podcast, and was struck by Horace Dediu’s encapsulation of the last 100 years of retail. The gist of his argument is that revolutions in retail shopping always ride in on the back of infrastructure improvements:

  • Railroads enabled long-distance distribution, and with that came catalog retailers such as Sears, Roebuck
  • Streetcars, steel, and elevators enabled skyscrapers and intra-city transit, and with that came urban department stores like Macy’s
  • Cars and freeways enabled suburbanization, and with that came the shopping mall
  • Bigger freeways and longer distances enabled exurbanization, and with that came the “power center” anchored by stores like Best Buy
  • And finally, the internet and modern logistics enabled internet shopping, and with that came

Dediu wonders what will come next, and whether it will disrupt Amazon the way the internet disrupted Best Buy, or the freeways disrupted department stores.

What I wonder is whether mobile computing potentially represents that next disruption. Is the ability to shop in a store, with a smartphone, revolutionary enough? Mobile phones are going to change in-store radically in the next few years. iBeacons are just the beginning. Google Shopping Express is also interesting in this regard. Does it constitute disruption? Time will tell.

Amazon became the biggest retailer on the web by providing three main services:

  • Discovery – if you search for anything, you’ll find it on Amazon
  • Purchase – 1-click shopping and stored credit cards
  • Fulfillment – a massive network of efficient warehouses

Each of these is a potential “attack vector” for a competitor.

In the discovery space, Google is the only serious contender, which is why Amazon is keenly interested in Google Shopping and at one time launched its own search engine, A9. Thus far, Amazon has done a good enough job of being your “front door” to shopping to keep Google at bay.  Could a physical retail store with sufficient technology provide a similar discovery experience? it’s a tall order, but not impossible. Mobile presents some interesting opportunities here.

With purchase, there are several well-funded competitors in market, including Apple, Square, Google, and PayPal. Each of these is intensely focused on mobile. Expect Amazon to introduce its own mobile payment service to compete here as well.

That leaves fulfillment. This would appear to be Amazon’s ace-in-the-hole. No one has yet been able to match the scale and speed of their distribution network, their investments in automation, etc. The risk is that this fulfillment service rides on the back of infrastructure that Amazon does not control – the trucking industry, the road system, and logistics suppliers like UPS. And as Dediu argued, changes in infrastructure beget disruptions in retail.

If Amazon wants to survive the next disruption, then, their best bet is to control more of the infrastructure over which their products travel.  Amazon Web Services is one way of doing that for  internet infrastructure.  As for product distribution, well… send in the drones.


Thoughts on SIC 2013

Photo via Admosis Media

Photo via Admosis Media

The 2013 Seattle Interactive Conference wrapped up today. This is my second year attending. While I think the organizers have the makings of a decent conference on their hands, with a few tweaks, I think it could be great. Here’s my unsolicited advice:

1. Crack down on marketing in the talks
XOXO down south has a pretty strict no-marketing rule in this regard, as do some other conferences. I don’t know that SIC has to be quite as strict, but I shouldn’t be subjected to talks that consist almost entirely of the speaker’s case studies or demo reel. Or worse, a talk that’s quite literally a sales pitch. Tell me something I can’t learn by reading your website.

2. Vary the time limits
An hour is too long for most speakers. Most talks either ended after 30 minutes or should have, even with Q&A. IXDA does a nice job of including 15-minute slots for younger speakers along with 45-60 minute slots for more established folks.

3. Ease off on the theme
The theme of the conference was “transparency.” How do I know? Because nearly every speaker bent over backwards to weave the word into their title or presentation. It felt forced (for the most part). The curator’s job is to select the theme and then pick talks that play off it in interesting ways. Instead I felt like I was being clubbed over the head.

4. Figure out what you want to be when you grow up
The website claims that SIC is about “the convergence of online technology, creativity, and emerging trends in one of the world’s most innovative cities.” To me it felt like it was mostly about online marketing. Which is fine! I work in online marketing and I’ve got no qualms with it. But maybe we should rename it the Seattle Interactive Marketing Conference and just own up to what it is.

5. Try fewer speakers
Instead of having 6-7 speakers every hour for three straight days, start with 3-4 speakers every hour for 2 days.  Improve the signal-to-noise ratio with better content.  A 2-day conference is totally respectable!

See you in 2014!

Business Strategy Starts With Good UX Design

Recently, I’ve become interested in business strategy, so I’ve been devouring books like The Innovator’s Dilemma and The Lean Startup. And what I’ve come away with more than anything is how much of business strategy is just good old-fashioned user-centered design.

To take one example, startup consultant and founder Steve Blank encourages startups to “get out of the building” and listen to their customers.  He advocates for a systematic process of making hypotheses and testing them with real customers.  Perhaps I’m oversimplifying it, but this sounds like good old fashioned user research to me.  Eric Reis, a student of Blank’s, takes the idea even further by marrying it with agile development in the book, The Lean Startup. I came away from Ries’ book more convinced than ever of the value of prototyping and user testing in helping to create a minimum viable product.

Clay Christensen, the author of Innovator’s Dilemma, popularized the notion of “the job to be done” where you design a product around a customer’s need, not that customer’s superficial demographic profile.  In other words, understanding the job to be done requires direct observation of the customer. In essence, it’s contextual inquiry!

Now, I’m sure I’m oversimplifying these principles somewhat to make them relate to me and my work. But it shouldn’t really be that surprising that basic user-centered design goes a long way towards making a great company and a great product.  If UX designers started talking less about user research methodologies that put executives to sleep and more about “minimum viable product” and “the job to be done” we might win more allies in the C-suite.

Five Practical Job Skills I Learned Studying Theater

Being a curious and somewhat confused undergrad, I graduated college with two degrees: in theater and computer science. After working in the tech industry for over a decade, I can honestly say the former taught me more about how to succeed in the workplace.

Now, I should note that I’m not a programmer. Your mileage may vary. But as someone who works with a team of people to ship software products, I’ve found that my day-to-day work has much more in common with the work of producing a play than it does with the computer science exercises I did as an undergrad.

Inspired by Scott Berkun, I thought I’d write down a few things I learned:

1. Collaborate Respectfully

Working on group projects in most classes was a bit daunting to me. You’d be put in a group of 2 or 3 people and expected to work together on a presentation or a paper or something. It was never clear who was in charge. Oftentimes, what ends up happening is that the person who’s most concerned about getting an A does all the work and everyone else sort of coasts. Not always, but that happened to me on more than one occasion.

By contrast, in theater everyone’s roles are obvious and named up front. When we work on this project, I’m the director and you’re the actor, and that means we each have distinct responsibilities. Next time, the situation may be reversed. This mutual dependency is at the heart of all teamwork. As the Clarence Otis, Jr., the CEO of Darden Restaurants, put it,

The thing that prepared me the most — where the team was front and center — was theater, which I did a lot of growing up, in high school, during college, law school and even for a couple of years after law school. I would say that probably is the starkest lesson in how reliant you are on others, because you’re there in front of an audience. It’s all live, and everybody’s got to know their lines and know their cues and know their movement, and so you’re totally dependent on people doing that.

2. Solve Problems

With a $200 budget, you might have to convince the audience you’re in a castle, then a forest, then finally a mountaintop. No CGI. Time to get creative. Steven Soderbergh put it well:

Art is also about problem solving, and it’s obvious from the news, we have a little bit of a problem with problem solving. In my experience, the main obstacle to problem solving is an entrenched ideology. The great thing about making a movie or a piece of art is that that never comes into play. All the ideas are on the table. All the ideas and everything is open for discussion, and it turns out everybody succeeds by submitting to what the thing needs to be. Art, in my view, is a very elegant problem-solving model.

3. Prototype

One key aspect of the play development process is the staged reading. You take the script and a group of actors, and you sit on stage and read it aloud to an audience. Before the first costumes are sewn, before the first set is built, you will have a pretty good idea of what aspects of the play are working and what aren’t. The staged reading is the minimum viable product

4. Tell Great Stories

Storytelling is at the heart of great theater, great marketing and great businesses. From the macro mission statement down to the most inconsequential PowerPoint deck, a great story is powerful. One thing I’ve learned from my years in the workforce is that the data is less important than the story you can tell around the data. We usually aren’t running double-blind experiments in the corporate world to see what works and what doesn’t. What tends to happens is: a project gets done, and the story of the project’s success or failure propagates throughout the company. If the story is a good story, it will propagate faster, and it’s more likely that the project will be held up as a model (or a warning sign). The story you can tell about the project is often as important as the project itself.

5. Ship

The first thing you learn in theater is that you have a ship date. It’s called opening night, and barring an extreme emergency, it’s not slipping. So immediately you’re forced to focus and prioritize. This is a skill that’s surprisingly hard to learn in college. You can usually ask for an extension. Or even if you can’t, it won’t hurt anyone but yourself. In the real world, no one’s going to to your job for you, and everyone else is relying on you to do your part.

To quote Clarence Otis again, “You could have your piece down, but if one person on the team doesn’t, you’re in trouble, and it’s embarrassing because people aren’t used to seeing errors in theater. Theater is seamless every night.” Or, as the Navy Seals say, under pressure, you don’t rise to the occasion, you sink to your level of training.

So why does any of this matter?

I’m not trying to talk anyone out of a CS degree. CS is fantastic. It’s a total shame that my local public university, the University of Washington, can’t make enough room for all the undergrads who want to enter the CS program. It’s doubly outrageous that we, the citizens, underinvest in our youth in such a way.

However, in an era where undergrads are running to so-called “STEM” degrees out of fear of unemployment, I’m worried we’ve put too much emphasis on what we deem a “practical” education to be. The arts matter, and not just because Steve Jobs studied calligraphy at Reed College (though that’s a helpful anecdote). The arts give us new ways of looking at problems, as Soderbergh elegantly put it. Which is why I applaud the efforts of folks like John Maeda at RISD to move from “STEM to STEAM.” Because the things you learn in school are profoundly important, just not always in the ways you assume.

Social Profiles

Edison Research, Learning to Love LinkedIn:

That means, of course, that 69% of LinkedIn users do not use Twitter, so they didn’t read your awesome tweet. That’s millions of Americans who are willing to engage with others online, willing to put their profile information on display, and who spend at least some time each week “social networking” but who are not engaged (and perhaps not even interested) in “tweeting.”

What’s most interesting to me is the way the survey question is framed: Facebook and LinkedIn users are asked if they “have a profile,” while Twitter users are asked if they “ever use” the service. As I tried to allude to in the comments, LinkedIn needs you to do more than just “have a profile” on the service. It needs the profile to be active and up to date, even if you’re not actively hunting for a job. This is likely why they’ve introduced features like endorsements and are ramping up their news strategy with the purchase of Pulse. These features give you a reason to keep coming back to LinkedIn every day and, hopefully, keep your profile up to date.

Why does LinkedIn care if your profile is up to date? Because LinkedIn’s real customers are recruiters, who pay for premium accounts. If the profile data goes stale, then the recruiters will start looking elsewhere. I’m sure the rise of Twitter resumes has LinkedIn a bit spooked in this regard.

In some ways, LinkedIn and Facebook are similar in this respect. Remember when Facebook announced OpenGraph and everyone thought it would make a killer dating service? The problem is the quality of the profile data:

“If I asked you what your favorite movies are and went to your movies section on Facebook, I don’t think they would match up. On OkCupid, people curate their profiles with a scalpel.”

And that’s true as far as the traditional Facbook profile stuff goes. But one of the great things about Facebook Timeline is that it turns profile maintenance into a passive act. Check in somewhere, upload a photo, like something: all of these things will have the effect of keeping your interests and activities up to date without you ever having to click “Edit Profile.” I’d expect LinkedIn to do something similar, and I think endorsements are a first step in the direction of passive or croudsourced profile curation.

You Manage What You Measure

Two points from this week’s episode of The Talk Show worth highlighting. Gruber and Om Malik are talking about how Apple struggles with the Internet, and how that might be related to the fact that the company pre-dates the web.

First, John mentions at one point that Apple products, since the company’s inception the experience design has been around the unboxing: you bring an Apple product home and use it by yourself. This can’t be overstated. I would go even further and say this blind spot isn’t just about the Internet, it’s about collaboration in general. Apple products are designed for a solo experience. It’s about the brilliant artist, tucked away in an attic, creating a masterpiece and unleashing it upon the world. If you work collaboratively with others on a day to day basis, as most people do, Apple’s products will continually frustrate you.

To take one example, surely no one at Apple has ever had to co-author a Keynote deck with several people, or the experience of doing so presumably wouldn’t be nearly as dismal. As it stands, you end up emailing 85MB files around to one another because you can’t store it on the server (doesn’t play nice with versioning) and iCloud doesn’t support granting other users access to your files. Oh, and there’s no easy way to bundle the fonts in with the presentation, so the app is constantly throwing errors about missing fonts. Again, none of this matters if you never interact with other people, which is presumably how Apple thinks you’re supposed to use its products.

The second point John makes is that iCloud backups work really well. He says iCloud backups have really high adoption rates, and that being able to bring a dead iPhone to an Apple store and have your backup magically appear on the replacement device is a wonderful thing. I have no doubt about that. But it’s worth pausing a second to think about why iCloud backups work so well. My guess is because someone’s measuring it. Apple has a huge financial incentive for a customer to walk into a store with a dead iPhone and walk away happy. It’s someone’s job at Apple to figure out how to optimize that experience. So that person is going to push and push for iCloud backups to be rock solid.

For Google and Amazon, improving server performance by several milliseconds can mean tens of millions of dollars in additional revenue. So there’s a huge incentive to make improvements. What’s the financial incentive for Apple to improve iCloud Core Data? It will make some developers happier, which in turn might make them make better apps, which in turn might make more people buy iPhones. It’s a triple bank shot, in other words. And it’s really hard to put into a spreadsheet.

IxDA 2013 Recap

Below are a few quick thoughts from my trip to the IxDA conference in Toronto last week.

Who owns our social graph?

Jer Thorp of NYU (TED talk here) gave a compelling presentation full of beautiful data visualizations using his personal data collected from Twitter, Foursquare and a variety of other sources. Thorp’s presentation was followed by a panel discussion later in the day on big data: who owns it, what are companies doing with it, what happens to it after we’re gone? As is his wont, Jared Spool took to the mic to ask if all this data lead to increased knowledge, or wisdom, or is it just noise?

The next morning Paul Adams of Facebook (and author of Grouped) gave some tips on designing effective socially-enabled experiences. Social relationships, he argued, are about many, many lightweight interactions over time. This is counter to how brands often seek to design “immersive” social experiences. People don’t want to be immersed; they want something that’s part of their everyday lives.

And that leads to a tension that ran throughout the conference: we know we can design better experiences using the social graph, but what responsibility do we have with the data we’re using?

Lean on Me

The influence of “lean UX,” inspired by Eric Reis’ The Lean Startup, could be seen everywhere. Designers are discussing ways to get the “minimum viable product” out the door more quickly. While you might be concerned that this rush to market inhibits the ability for user research, Josh Seiden argued it’s actually quite the opposite. “Lean” practitioners believe in getting something in the market very quickly, and talking to customers to validate their assumptions and test their hypothesis. In other words, it’s prototyping but with a real product and real customers. Paul Adams went so far as to say it’s impossible to design social software without real users using it. He noted that Facebook deploys code twice a day to its billion users.

So where does interaction design fit in the corporate hierarchy? Johanna Kollmann and Martina Schell discussed how they help startups realize their vision. Cindy Chastain of R/GA made the case that UX is in large part change management, and UX’ers need to think of themselves as business consultants.


Speaking of lean, there were several presentation that touched on the limitations of wireframes as a communication tool. Derek Vaz made the case for interactive prototypes, while Peter Stahl gave an interesting talk (older version here) on the need to consider rhythm and flow in our designs. He said we should think like musicians, with a beat in our heads, as we design the flow from screen to screen. If we aren’t defining these rhythms, someone else will define them for us.

Beyond the screen / Internet of Things

Kate Chapman, IxD board member and OCAD professor (TED talk here), showed some whimsical student projects from her “wearable computing lab.” One was an internet-connected skirt made of earthquake debris that vibrates when an earthquake occurs anywhere in the world. Carla Diana’s presentation on IoT contained more great examples than I can list here; I’ll be sure to send out the deck when it’s available (in the meantime, there’s an older version of the deck here).

Learning from Marketing

We may dismiss traditional marketing, but marketers can teach us something. Cliff Kuang from Fast Company suggested that interaction designers lack the storytelling and emotional tricks of brand gurus, and we could become more influential if we improved our storytelling chops when trying to sell our ideas.

On a similar note, Matt Walsh of CP+B presented on the need for tension in design. Tension can create drama or humor. Interaction design is often obsessed with removing tension from UI, but sometimes tension can be used creatively to deepen emotional engagement and create interest. For example, the self checkout at the grocery creates tension by transferring control to the shopper. Create a fear of missing out (FOMO!) by showing other people having fun.

The Social Impact of Design

The theme of the conference was “social impact.” There were sessions on Design for America, designing for “distressed interactions” at the Mayo Clinic, and a project to redesign a school lunchroom. Robert Fabricant of Frog suggested a couple of times that it was time for interaction designers to take on bigger problems, like health care and transportation.

IxDA Awards

On that note, the awards this year recognized some obvious products – Nike Fuel, the Obama for America Apps – along with some non-traditional choices like 21 Baloncoires and Studio H: Bertie County.

Design Process

It wouldn’t be a conference without a little navel-gazing about process! Jason Ulaszek and Brian Winters from Manifest Digital laid out their discovery process in a solid talk. Stakeholder interviews? check. User research? check. Personas and mental models? check and check. Yeah, we got that. Dan Saffer pimped his new book, Microinteractions.

One note I wrote down, which I can’t recall where I heard it: grest digital companies are built around one or two core user interactions, which they polish obsessively.


Gamification came up several times, but the most interesting was from HUGE’s Kunal Patel, who called badges the “backup quarterbacks of game design”: useful, but only when paired with the right team. Think about how gamification can solve your problems. For example, Stack Exchange rewards pro users with advanced features, not gimmicks. This incentivizes better behavior on the discussion threads, which makes the site more useful for everyone.

Did you attend IxDA? What were your favorite sessions?

The iPhone Classic II


Released in 1989, the Macintosh SE/30 is widely considered the apotheosis of the black-and-white Mac era and perhaps the finest Mac of the 1980s. With a Motorola 68030 processor, a dedicated FPU, a high-density floppy drive and an expansion slot, the SE/30 was built for performance, and sold for nearly 3 years before finally being discontinued.

It also cost $6,500.

When Apple decided to revive the black-and-white form factor in the early 90s with the Mac Classic and later the Classic II, it did so with an eye towards value.  Instead of turning the SE/30 production lines back on and selling it for less money, Apple opted to redesign it completely.   While the Classic II was superficially similar, it wasn’t nearly as powerful.  Here’s Wikipedia describing the difference:

[The SE/30] was followed in 1991 by the Macintosh Classic II, which, despite the same processor and clock speed, was only 60% as fast as the SE/30 due to its 16-bit data path, supported no more than 10 MB of memory, lacked an internal expansion slot, and made the Motorola 68882 FPU an optional upgrade.

The Classic II retailed for $1,900, less than a third of the SE/30′s original cost.  It’s unlikely Apple could have sold the SE/30 for so little money, simply because it was never designed to hit that kind of price point.

Which brings me to the rumors of a “low cost iPhone” that have picked up steam in recent weeks, and which John Gruber said might hypothetically constitute the “next incremental step” in Apple’s product plans.  The average sale price of the iPhone is $650.  If Apple wants to sell iPhones to billions of customers in the developing world While the iPhone 4 is still on the market and can be had for “free” with a 2-year contract, it’s not designed for low cost.  Getting it down to an unsubsidized price point of, say, $200, to sell in India might be difficult.

As Horace Deidu notes on a recent Critical Path, it’s highly likely Apple will expand the iPhone product line at some point. A lower-cost iPhone will be built from the ground up to hit a lower price point.  It will look like an SE/30 but perform like a Classic II, perhaps using an older (A5?) processor and a non-retina screen.  Every Apple product starts with a mission statement and then every design tradeoff is made with that mission statement in mind.  The low-cost iPhone’s mission statement will be value.

An App for That

Anniversaries are a cheap way of finding an excuse to talk about something.  Nonetheless, the 5-year anniversary of the iPhone is as good a time as any to talk about how mobile app design has evolved over the past five years.

The slogan, “there’s an app for that” is cliche today, almost quaint: for every task, an appropriate app.  Before the iPhone, one might have been tempted to say “there’s a google search for that” or “there’s an AOL keyword for that.”  But now: apps.  They are how we do things.

But not everything, of course…

Implicit in the “there’s an app for that” slogan is the idea that “that” is a single thing, an atomic task: reading the news, doing a crossword, booking a flight.  Whether designing an app for Android, iOS, or Windows Phone, the mantra has been the same: do one thing and do it well.

This laser-like focus has generally served app designers well. When the App Store launched, the conventional wisdom was that one should design for quick, easy-to-consume experiences that relied on a mobile user who might only be using one thumb, one eye, and a jacked-up AT&T Edge connection.  And so we designed fairly shallow, low-information-density apps.

Over the past five years, though, the constraints have slowly but inexorably changed.  We now have:

  • GPS standard in every phone
  • Fast mobile processors
  • Better, smarter back-end data services that can provide only the most relevant content
  • high-resolution screens
  • faster, more stable LTE connections (in some places, anyway)
  • More mobile use at home and in the office, where data connections and attention spans are pretty decent
  • a wider array of “mobile” devices, including large-screen tablets
  • more sophisticated users who better understand mobile conventions and gestures
  • authentication services like Facebook that allow for richer customization

All of this means re-thinking how we design apps.  Today’s apps can – and should – aim to do more than they did 5 years ago.  Functionality that we might have spread across two or three apps can now be consolidated in one. Apps can be more ambitious.

In the past, my advice to clients asking about app strategy has generally been to be super-specific and focused. But I might be starting to re-think that.  My 2013 resolution is to be more ambitious about these sorts of things.