Sun Tzu and the Art of Disrupting Higher Education

In 512 BC or thereabouts, the venerable Chinese general, Sun Tzu wrote The Art of War in which he proffered the following advice on military strategy:

The highest form of generalship is to balk the enemy’s plans…the next best is to attack the enemy’s army in the field; and the worst policy of all is to besiege walled cities.

Sun Tzu’s advice applies equally well in business, as disruptive new ventures are well advised not to initiate an all out war aimed at taking over incumbent market leaders’ core markets– their “walled cities.” Sebastian Thrun, the self-declared father of the modern movement to disrupt higher education with Massive Open Online Courses — aka MOOCs — has learned this lesson well.  And therein lies an instructive tale.

MOOCs and university walled cities

For those unfamiliar with this space, prior to 2011, Thrun was a tenured professor at Stanford University who taught one of the most popular computer science courses on campus, routinely packing his lecture hall with 200 students.

But after hearing Salman Khan’s inspiring TED talk in March, 2011, Thrun reckoned that he could reach a far larger global audience by transforming his course to video format.  He set out to convert his popular Stanford course to online delivery which initially featured videos produced with nothing more than a camera, a pen, and a napkin.  Despite the low initial production quality, many of his 200 Stanford students chose to switch to his video version because they could absorb the material at their own pace and on their own schedule. Eventually, the 200 student classroom dwindled to a group of 30. Meanwhile, the course’s popularity exploded online. Within two weeks of its announcement, 56,000 students from around the world had enrolled, swelling to 160,000 by the start of the virtual semester — more students than Thrun could reach in his physical classroom in 800 years.

The initial experience convinced Thrun that he could craft an even better course with interactive Web tools that adequately recreated the intimacy of one-on-one tutoring. The student value proposition seemed compelling: one of the best professors from one of the best universities teaching one of the most popular courses on his very best day (Thrun re-taped his lecture modules until perfected) — all for free!  The audacity (root for the venture’s ultimate name) and potentially vast scale of the endeavor — along with  a dash of founder hubris — prompted Thrun to leave the comfort and prestige of his tenured position at Stanford.  Thrun launched Udacity with a $5 million Series A investment from Charles River Ventures and $300,000 of his own money.

In response to its extraordinarily successful launch — at least as measured by student enrollments –  Udacity rapidly expanded its online course offerings, and was soon joined in this space by a similar and even larger Stanford spinoff venture — Coursera — as well as by edX, a MOOC joint venture between MIT and Harvard.  Collectively, the Big 3 MOOCs soon attracted >3 million course enrollments and over $100 million in venture investment.

As these new ventures entered their second year of operation, despite the lack of a clear sense of how they would convert eyeballs to dollars, it was hard to not get swept up in the noble aspiration to democratize the availability of first rate education to every citizen of the world within reach of a computer and a broadband network.  Thrun was one of the most vocal MOOC proselytizers, explaining his decision to leave traditional academia in January 2012 by saying:

I can’t teach at Stanford again, I feel like there’s a red pill and a blue pill. And you can take the blue pill and go back to your classroom and lecture your students. But I’ve taken the red pill. I’ve seen Wonderland.”  

At the same time, Thrun also boldly predicted that Udacity  would profoundly disrupt the traditional university system, noting:

In 50 years, there will be only ten institutions in the world delivering higher education and Udacity has a shot at being one of them.”

In essence, Thrun was declaring disruptive war on the walled cities of established universities, threatening a technological revolution that would blow up a number of cherished institutions and traditions.

Warren Hall Demolition - Sundt Construction_resized

Sacré bleu, could it really be that the grandchildren of today’s college students would no longer see ivy covered buildings adorned with honorific statues and gargoyles, large lecture halls filled with a mix of fascinated and napping students, tenured academic faculty pontificating on arcane research topics, guy- and gal-watching on manicured quad lawns and game day mania at college football and basketball games?!

Not so fast!
But less than two years after his bold pronouncement on the impending death of higher education as we know it, a discouraged Sebastion Thrun downgraded his messianic view and hyperbolic enthusiasm by noting:

I’d aspired to give people a profound education–to teach them something substantial, but the data was at odds with this idea. We were on the front pages of newspapers and magazines, and at the same time, I was realizing, we don’t educate people as others wished, or as I wished. We have a lousy product.

What precipitated such a dramatic fall from grace?

For one thing, Thrun and his fellow MOOC pioneers were undoubtedly troubled by stubbornly low student completion rates, which continued to hover in single digits for most courses, despite multiple attempts to improve pedagogy and student engagement.  Then too, the path to profitability  — at least for the VC-backed, for-profit ventures, Udacity and Coursera — remained cloudy.  Simply throwing more free courses from more universities on a MOOC platform appeared unlikely to yield a viable investment return, notwithstanding attempts to add “freemium” services such as test validation, completion certificates and recruiting data services.

But perhaps the most bitter disappointment for Udacity was its widely noted failure in a pilot program to replace in-class education at San Jose State University in California.  In late 2012, Thrun proposed a partnership to California Governor Jerry Brown, who had been struggling to cope with rising tuition costs, poor student performance, and overcrowding in state universities. At a press conference the following January, Brown and Thrun announced that Udacity would open online enrollment in three subjects–remedial math, college algebra, and elementary statistics–that would count toward credit at San Jose State University, a 30,000-student public institution.

Courses were offered for $150 each, and students were drawn from lower-income neighborhood high schools and the members of SJSU’s student body for whom there was no space in conventional classrooms. The pilot’s target students were struggling to keep up with requisite levels of educational achievement, prompting Thrun to declare at a press conference:

A lot of these failures are avoidable.  I would love to set these students up for success, not for failure.

But when test results from the first pilot courses came out, it was clear that Udacity’s online courses failed to deliver acceptable results. In particular, it was found that 74 percent or more of the students in comparable classroom courses passed, while no more than 51 percent of Udacity students passed any of the three online courses offered.

Notwithstanding the fact that this initial pilot targeted a particularly challenging student population — which in retrospect was ill-advised for a venture at such a primitive state of development –  adverse publicity led to a temporary suspension of the Udacity online course experiment. Critics — and there had been many all along the way — cheered Thrun’s comeuppance.  Some declared that the poor pilot performance was “predictable” while others lamented the “immorality” of public funds being used to pay a for-profit company to experiment on students with an unproven new approach to higher education.

Academics’ schadenfreude was understandable payback for Thrun’s hyperbolic zeal in promoting the pilot program, and many were quick to claim victory against Thrun’s unsuccessful attempt to invade their university walled city.

Immoral experimentation?
But if Thrun’s victory laps were premature, so were critics’ condemnations.

There are so many things wrong with such quick and sweeping condemnations of Udacity’s approach, intent and even morality that’s is difficult to know where to start.  But as this debate gets to the heart of the future of higher education, it is important to set down some markers.

Experimentation and fast learning from failure lies at the very heart of technological progress, so drawing sweeping conclusions from an initial pilot test demonstrates a self-serving lack of understanding of how innovation works.  As an analogy, I would hope we wouldn’t suspend research on all cures for cancer because an initial clinical trial regrettably failed to save the lives of some terminally ill patients.  Lest this sound like a deliberately draconian analogy, it is important to note that what compelled Governor Brown to seek out experimentation in this case was his recognition that California had neither the budget nor the resources to adequately educate the state’s young adults, particularly the most vulnerable students requiring remedial attention. Despite the fact that state education expenditures per capita had increased by nearly 25% over the past decade, too many young adults were failing to receive the education required to live a fulfilling, productive life.  Under the circumstances, those who so swiftly and harshly condemned the search for new solutions to a chronic and worsening education problem in California should reexamine the logic of simply throwing more money at a system that was not able to deliver acceptable results.

As it turns out, after a two month hiatus, Udacity’s online courses returned for a second semester, with dramatically better results. As noted in the table below, for some courses, online learning achieved better outcomes than traditional classroom formats (at a fraction of the cost).  Whether this is the result of a significantly different student sample or due to improvements made in course design between the two trials is unclear.  But that’s exactly the point.  We need continuous disciplined experimentation to determine the most cost effective approaches to higher education.  And we should learn from, not condemn inevitable setbacks along the way.

Screen Shot 2013-08-27 at 7.09.05 PM

Looking more broadly beyond the San Jose experiment, it is important to recognize that the higher education model in the US is broken and will be disrupted by Udacity or others committed to find more cost effective mechanisms to deliver higher quality, more relevant higher education.  To substantiate this assertion, let’s start by observing that higher education meets all of the conditions that measure the vulnerability of any industry to disruptive transformation.

Ed_Disrupt Vulnerability

The fourth point on this list is worth noting.  In my business strategy course at a Tier 1 university, I often ask my MBA students to cite some examples of the greatest technological breakthroughs in human history.  Common responses include the wheel (big improvement!), the printing press, electricity and, for those with shorter memories, the iPhone.  Then, as I stand before a blackboard poised to record the answers, I ask my students to cite the biggest technological leaps in the history of education.  But this question usually stumps the audience, drawing only tepid responses including “chalk” and “Powerpoint”.  Hardly the stuff of technological revolution!  Indeed the structural characteristics of my class (if not the caliber of teaching) differs little from the great Socrates — a predecessor of Sun Tzu.  We both gather a limited number of pre-selected students — he under a tree, and me in a classroom — to engage in an inherently labor intensive dialogue aimed at imparting knowledge, judgment and critical thinking.

When an instructor is talented and students are engaged, this method works quite well.  However there are inevitable drawbacks to this time-honored approach: inherently high cost, inflexibility (requiring students to metaphorically “gather under a tree”), and a chronic dearth of talented teachers.  This last point is worth stressing.  Despite the widely reported increase in the costs of delivering higher education, many scholars — for example Clay Christensen and Richard Arum — have also documented declines in student learning outcomes, notably college graduates’ critical thinking and writing skills.  In short, we’re paying considerably more for and getting less from higher education.

Is there something about education that makes it uniquely immune to technological progress?
Defenders of the status quo would say yes, pointing to the numerous past failures of communication technology breakthroughs — movies, radio, television and the Internet — to live up to lofty claims of changing education forever.  So why should we believe that MOOCs — or whatever else you choose to call emerging online education formats — will be different this time?

New York Times columnist Thomas Friedman perhaps answered this question best when he said:

Big breakthroughs happen when what is suddenly possible meets what is desperately necessary

What makes disruption of higher education “suddenly possible” is the confluence of four emerging technologies:

  • Web 2.0 interactivity tools that allow much higher forms of interactive student engagement — with instructors and with each other
  • Widespread high speed broadband availability
  • Plummeting IT and video production costs
  • Rapid advances in online pedagogical techniques, including adaptive, personalized courseware, automated grading (including free form text entry) and team-based project coordination

There have been some truly inspirational early success stories for online higher education as  captured by Coursera co-founder Daphne Koller’s mid-2012 TED talk.    But while we still have a lot of room for improvement — as Udacity’s San Jose State University experiment reveals — it’s important to remember that we’re still early in the process of redefining a set of institutions that were more than 2,500 years in the making.  And from that historical perspective, the progress achieved in exploiting disruptive technologies for higher education the past few years has been encouraging.

As for the second half of Friedman’s prediction, there should be little doubt in the “desperate necessity” to reform higher education in the US.  Recent trends  are more than discouraging, as noted below.

Hiegher Ed Characteristics

These points are beginning to take a toll.  From 2010 through 2012, freshman enrollment at more than a quarter of U.S. private four-year colleges declined 10 percent or more, according to a recent analysis by The Wall Street Journal. Perhaps the most compelling evidence of the inevitability of disruptive change to higher education is captured in the chart below displaying the growing gap between tuition costs and college graduate starting salaries.

As NYU professor Clay Shirkey has bitingly noted:

The value of a college degree remains high in relative terms, but only because people with bachelor’s degrees have seen their incomes shrink less over the last few years than people who don’t have them. ‘Give us hundreds of thousands of dollars and years of your life so you can suffer less than your peers’ isn’t much of a value proposition. More like a ransom note, really.

So the question is only when, not whether new approaches to higher education deliver more effective and more relevant learning outcomes at lower cost.

Udacity’s pivot
In August, 2013, Udacity announced it was changing its business model, from offering an eclectic array of massive open online courses for free, to contractual arrangements with major corporations to produce targeted skill-building courses primarily aimed at prospective and current high tech employees on a fee-per-course basis.  The new strategy had two important  initiatives:

  1. A partnership between Udacity, AT&T and Georgia Tech University to offer a fully accredited online MS in Computer Science for $6,600 — less than one-third of what an in-state student would pay at Georgia Tech, and one-seventh of the tuition charged to an out-of-state applicant.  AT&T kicked in $2 million of funding to subsidize the first year of operation, presumably motivated by the opportunity to recruit talented graduates.
  2. A partnership with Google, Intuit, Cloudera, Autodesk, Khan Academy et al in an “Open Education Alliance” wherein the memebers will fund and help design the creation of a defined curriculum designed for students pursuing jobs in technology.  According to Udacity’s press announcement, participating members will “assist in the curation and development of a new 21st century curriculum and connect learners with opportunities in industry,” .

Once again, critics were quick to pounce, including this reaction from a widely published academic specializing in technology and education:

Well, there it is folks. After two years of hype, breathless proclamations about how Udacity will transform higher education, Silicon Valley blindness to existing learning research, and numerous articles/interviews featuring Sebastian Thrun, Udacity has failed.

I strongly disagree with this conclusion, not borne out of a strong conviction that Thrun will succeed (he well may not), but because this condemnation is intellectually flawed in two important respects:

  1. It fundamentally misunderstands the importance and frequency of business pivots in startup ventures
  2. It fails to recognize that Thrun’s pivoted business model poses a far more serious threat to traditional higher education institutions than Udacity’s original approach

Let’s start with business pivots.  When an early stage venture pivots its business model, it does not imply that the business has failed or even that it is in dire straits.  It does mean that the venture has gained valuable insights about customer preferences and market dynamics in an initial trial of its business plan and is attempting to respond accordingly.

Business plan pivots reflect a logical and necessary evolution in developing successful businesses as the following examples attest:

  • Post-It Notes languished in 3M’s research labs for five years without a convincing business plan, then failed miserably in pilot tests in four cities, where the company tried to sell “Press & Peel Pads” (as it was then called) as a glorified sticky bookmark. Two years  later, in a last ditch effort to make the product stick (sorry!), 3M re-piloted what was now called Post-It-Notes in a massive free give-away.  Thousands of sample products were sent to office managers, purchasing agents, lawyers and hospital personnel, coordinated with 3M personnel stationed at customer facilities to explain possible uses of the product.  The pivoted marketing strategy was highly successful, word-of-mouth endorsements went viral, and Post-It-Notes emerged as a billion dollar brand.
  • Google received its first tranche of venture capital from Kleiner, Perkins (et al) in June, 1999 on the strength of its promising technology, but with no viable business model in place to monetize their search engine. In fact, the founders were initially opposed to “contaminating” their site with advertisements.  It took more than two years and several unsuccessful steps along the way for Google to pivot to a pay-per-click, ad-supported revenue model, and the rest is history.
  • Nespresso languished for over a decade as a money-losing, slow-growing provider of single serve coffee machines and coffee pods to food service and office establishments.  A new CEO, Jean-Paul Gaillard was hired in 1988 to reenergize the franchise, and refocused Nespresso’s business on the consumer market as a high-end gourmet product.  Gaillard’s strategy pivot involved a number of a critical decisions, including selling Nespresso’s single serve coffee pods only through its own phone-order  channel and shunning mass advertising in favor of word-of-mouth referrals — both radical departures from normal business practices at Nestlé.  Although Gaillard’s strategy pivot evolved over time — Nespresso now also sells pods through its own retail boutiques and over the web — the fundamentals of the pivoted business model remained intact, propelling Nespresso to highly profitable, rapid growth (CAGR=35%!)  over the ensuing two decades.

So obviously, pivots don’t necessarily signal failure. Entrepreneurs often reverse early setbacks by incorporating market feedback to relaunch successful businesses.

After two years of operation, Udacity undoubtedly came to the conclusion that its initial MOOC business model offered very little promise of financial viability.  The company always knew that free is not a business model, but it also discovered that secondary sources of “freemium” revenues were unlikely to offset the high development costs of online education.

Thrun’s pivot puts Udacity in a better position to disrupt higher eduction than its initial positioning.  Why?  Because instead of trying to outdo universities at their own game — that is, delivering a wide array of college courses  — Udacity has shifted to focus on a job that higher education institutions have traditionally largely shunned: corporate training.  And in so doing, Udacity now threatens to break down a major pillar propping up the  current business model of higher education –  thereby “balking their plans,” as Sun-Tzu said millennia ago.

Recall that the (weakening) justification for traditional college education has historically the attractive returns from higher lifetime earnings.  But Udacity’s new focus threatens to profoundly undermine this rationale. If employers experience positive results in hiring high tech employees who have acquired superior job skills from Udacity’s new low cost training initiatives, student interest in attending, and employer interest in recruiting from many conventional institutions of higher education will decline.

Udacity is trying to bridge a serious gap in our higher education system: the disheartening disconnect between college graduates’ growing difficulty in finding emotionally and financially satisfying jobs at the same time that employers chronically complain they can’t find enough qualified employees amongst the ranks of recent college graduates.  According to a recent study from McKinsey,  while 72 percent of educational institutions believe recent graduates are ready for work, only 42 percent of employers agree. There appears to be an equally large disconnect between college administrators and the  general population on how well colleges prepare students for their careers.  This is the gap that Udacity is now trying to close.

It is far too early to tell how Udacity will fare, as its new partnership programs have yet to be fully implemented.  But a few early indicators should be taken very seriously by academic leaders of our current higher education institutions:

  • It appears that a growing number of companies are prepared to become more directly involved in addressing the challenge of preparing their employees for 21st century business problems — and are putting their money where their needs are
  • Students are reacting positively to the opportunity not only to acquire accredited new job skills at sharply lower costs, but to do so with improved prospects for immediate employment from participating corporate sponsors.  Over 2,300 students applied to the first entering class of Georgia Tech’s new online MS degree program in computer science.  To put this number in perspective, since Georgia Tech created its on-campus master’s degree program in in computer science in 1991, fewer than 2,000 degrees have been awarded. Under the new effort, that many online degrees could be awarded in a single year.

What are the implications for today’s higher education institutions?
It is dismaying to see so many academics rushing to dismiss innovative initiatives which seek to improve our higher education system. Resistance from incumbent stakeholders will eventually be overcome by three large and powerful constituencies poorly served by today’s status quo: the 70% of US adults who do not have a college degree, the majority of college graduates who are dissatisfied with the value of their degrees and the large number of employers challenged by a skills gap in the recruiting marketplace.  The economic potential that can be unlocked by better serving these large constituencies will continue to attract investment in alternative education delivery models from both the private and public sector.

Higher education institutions can no longer ignore their imperative to do a far better job preparing students for more demanding and fast-changing careers.   While the hubris and hyperbole of some high tech zealots have been admittedly grating, hopefully, leaders of higher education institutions now recognize that the question is no longer whether, but only how and when disruptive technologies will reshape higher educational delivery models.

As in any disrupted industry, the speed with which disruption occurs will vary widely across the higher education landscape.  Many smaller colleges that rely almost entirely on tuition revenue are already facing severe financial distress.  But no institution should feel immune from the disruptive forces at play, including the most highly endowed Tier 1 universities.

My advice to academic leaders would be to commit to two inter-related strategic imperatives.

  1. Be prepared to broadly rethink the mission and priorities of higher education
    The factual evidence of a significant skills gap in college graduates’ preparedness for 21st century careers is clear. Therefore, higher education institutions need to ask themselves: what is our responsibility and action plan to address this societal problem?  Towards this end, colleges and universities will need to broadly rethink a wide array of business policies that currently inhibit the delivery of cost-effective, first-rate education.  Many of the shibboleths that define higher education today are deeply entrenched in institutional norms and academic psyches and as such, will not be easy to change.  The challenge of course is to assess future priorities and strategic imperatives from the standpoint of market and societal needs, recognizing that the way forward may be at odds with the perceived welfare of some academic stakeholders. But with long term survival at stake, the requisite strategy pivots are likely to call for very different behaviors and skill sets than found on many current campuses. As a faculty member of a Tier 1 university, I am as convinced of the ongoing value of an on-campus higher education experience as I am of the need for profound changes in how that experience is delivered. Academic leaders must be prepared to disrupt their own institutions before external forces foreclose current options.  These are not easy dialogues to initiate, but it will only get harder over time.
  2. Initiate and learn from multiple experiments on new pedagogies and delivery mechanisms
    No one can claim to know the precise pace and form that disruptive learning technologies will take over the next decade.  But I would argue that it is precisely because of this inherent uncertainty that the only appropriate response is to undertake extensive low-cost iterative experiments on college campuses. Universities need to discover for themselves how to best incorporate new technologies into their on-campus and extended learning environments.  I would like to see more higher education institutions aggressively undertaking and sharing experiences from multiple digital learning experiments, including video lectures to “flip”  classrooms, MOOC courses to extend  learning reach and to gain familiarity with online pedagogical techniques, greater use of video technologies to beam global thought leaders into our classrooms, and experiments with different forms of automated grading for larger online and classroom audiences.

None of this will be easy, and there will undoubtedly be a number of painful pivots along the way.  But leaders of higher education would be well advised to heed the advice of two recent pronouncements germane to the topic at hand.  The first is from the 2006 U.S. Department of Education Task Force Report: “A Test of Leadership: Charting the Future of U.S. Higher Education” who noted that:

History is littered with examples of industries that, at their peril, failed to respond – or even to notice – changes in the world around them from railroads to steel manufacturers.  Without serious self-examination and reform, institutions of higher education risk falling into the same trap, seeing their market share substantially reduced and their services increasingly characterized by obsolescence.

And the second piece of advice comes from Reid Hastings, CEO of Netflix who knows a thing or two about strategy pivots:

If you are not genuinely pained by the risk involved in your strategic choices, it’s not much of a strategy.