<img height="1" width="1" src="https://www.facebook.com/tr?id=1483239291704574&amp;ev=PageView &amp;noscript=1">

the hype innovation blog

The Innovator’s Dilemma: Narrow Theory, Widely Applied

Posted by Tim Woods on Jan 31, 2016 10:05:19 AM


It’s been nearly twenty years since the publication of The Innovator’s Dilemma, the go-to book for the disruptive generation. The Economist named it as one of the six most important business books of all time. However, the term ‘disruptive innovation’ has been hijacked and used in all sorts of ways which certainly do not fit the original theory put out by Christensen. It’s worth looking again at what the theory actually entails, particularly for those that haven't read the original work; then look at some criticisms and whether the term 'disruptive innovation' should remain inherently linked to Christensen’s original idea.

Sustaining vs. Disruptive Innovation

Almost all innovation that happens, in every industry, is of the sustaining type. Where companies improve what they already have, making it more appealing and useful to their existing customers. Getting this distinction right is crucial to deciding whether something is a disruptive innovation according to Christensen’s criteria. Apple’s iPhone, for example, is a sustaining innovation, because it merely improved an existing technology, and served the same market as other mobile phone manufacturers.

Disruptive innovations begin at the foothold of the market, meaning that they serve the low-margin, fringe customers, who have lower expectations on product performance. By definition a disruptive innovation is not as good as the existing technology, but it gains a foothold by focusing on one key dimension, such as speed, price, or convenience.

In his case study, smaller disk drives entered at the foot of the market, because they didn’t have the capacity needed for mainstream customers (who only wanted better mainframe computers). But fringe customers, such as makers of medical devices and cars, found a use for the smaller size and lower price point.

This distinction is why Uber is not considered a disruptive innovation, despite being disruptive to the taxi industry. It entered the market at the top end, competing with what many consider a better product performance offering.

Christensen has revised his original theory to now account for another entry point for disruptive innovations, which is that of new-market entry. The iPhone is an example, where although it was not disruptive to the mobile phone market, it was disruptive to the desktop and laptop market, as it became the ubiquitous way for consumers to access the internet, overtaking the PC market.

Using the Theory Appropriately - Process and Trajectory

The taxi industry and Nokia might find this distinction hard to accept. But Christensen makes an interesting point that we must not consider every successful company disruptive, because we won’t be able to learn any common principles or lessons. If we constrain our theory to specific criteria, we are more likely to find patterns which managers can actually use to make better decisions, under a specific situation.

Crucially, it is not whether a particular technology is disruptive at a given point in time, it’s whether it is following a disruptive trajectory. Netflix started out at the low-end of the market, serving only those fringe customers who were willing to wait several days for a DVD to arrive, and not necessarily concerned with the new releases. Netflix did not go for the mainstream market initially. Over time they increased their selection of DVDs far beyond what Blockbuster made available, and then moved into online streaming, eventually attacking the mainstream.


(to be clear, that's my bad sketch, not an official Harvard chart)

The process of moving from the fringe to the mainstream can take time, and all the while the incumbents are busy fighting with their mainstream competitors, which is why disruptive innovators appear to sneak up on incumbents and eat their lunch.

The Five Principles of Disruption

Christensen is primarily interested in why great companies fail. In the book he looks at the disk drive, mechanical excavator, and steel industries to find out why leading companies are toppled by under-resourced new entrants. What emerges are five principles (or laws). If managers can fully understand and be conscious of these, they can be better prepared to handle the threat of disruption.

Principle 1: Companies depend on customers and investors for resources

Companies will always be led by their customers and investors, and often by the biggest and most demanding customers. What these customers want is more of the same, but better - they direct a company’s resource allocation towards sustaining innovations. The largest customers don’t want the low-end, cheaper, lower quality product; they want the best possible quality, and they will pay for it. Because incumbents depend upon these customers for the higher-margin profits, they will organize themselves internally to best fit those needs.

“...money will be spent because companies with investment patterns that don’t satisfy their customers and investors don’t survive. The highest-performing companies, in fact, are those that are the best at this, that is, they have well-developed systems for killing ideas that their customers don’t want. As a result, these companies find it very difficult to invest adequate resources in disruptive technologies”

Principle 2: Small markets don’t solve the growth needs of larger companies

Because disruptive innovations begin at the low-end, they are able to grow on small turnover and margins, but established companies cannot do this. A $40m company which seeks to grow 20%, has to find only $8m of new revenue; but a $4b company seeking to grow 20% must find $800m of new revenue, and emerging markets initially aren’t large enough to provide that growth. By the time they are large enough, it’s already too late.

Principle 3: Markets that don’t exist can’t be analyzed

With emerging markets the one thing we know for sure, is that expert forecasts about how large the market will become are always wrong. The planning tools we use to understand sustaining technologies don’t work when used on disruptive technologies, Christensen says this is an exercise in ‘flapping wings’.

Principle 4: An organization’s capabilities define its disabilities

People are surprisingly flexible. You can take an individual out of a large company and put them in a small startup, and they can adapt and survive. But a company has processes and values - the way it produces output, and the way it makes decisions on resources and pursuing ideas. These processes and values are not flexible, they cannot easily change to support a different low-margin product, when they are organized effectively around producing a high-margin product. The new scenario causes a company’s strengths to now become its weakness.

Principle 5: Technology supply may not equal market demand

The incumbent companies compete aggressively to obtain higher-margin profits, by creating better product performance for the existing customers. Without realizing it, they overshoot the needs of the general market, and create a vacuum for lower-end, lower-cost technologies to enter at the foot of the market, servicing the basic needs of customers. Incumbents often don’t realize the speed at which they are moving up-market, seeking higher profits, because they are engaged in a competitive battle with other incumbents. Because new entrants can enter with lower price points, it creates a compelling reason for some customers to begin changing.

‘Good Management' is the Innovator’s Dilemma

Middle management play a critical role in weeding out ideas. They are attuned to putting resources behind ideas which have a high chance of success - sustaining technologies are easier to identify in this case. Well-run companies will naturally gravitate towards those ideas which keep the company moving upwards in terms of higher profit margins, and greater product quality for the customer. So good managers are doing exactly what they’re supposed to do when they shift resources towards sustaining technologies, and away from highly uncertain disruptive ones.

A manager can suffer tremendous career setbacks if they support a failed project. If it failed because of the technology performance, then it’s less harmful, but if it failed because there was no market for it, it’s highly damaging.

Even if the company decides to pursue a new technology, its own cost structure and value network can cause it to fail. For example, sales and marketing departments may not favor the new technology, because they know it doesn’t fit the immediate needs of their primary customers, which in turn mean thousands of little decisions going against the pursuit of the disruptive technology, and towards the sustaining. In total it means the organization is rejecting the new technology.

The problem is compounded by the fact that established firms try to take the new technology and push it onto their existing customers, which typically reject it, while the new entrants find a new market, or new type of buyer for the technology.

New hydraulics technology didn’t provide large bucket-size capacity, which was essential for the mainstream contractors. But it did provide narrow digging, for trenches, and a high degree of speed and maneuverability. This was the start the entrants needed to get a foothold on the market, and start moving their way up the value chain.

The Solution

What to do about it? The advice is quite simple, but very hard to execute. When you see a disruptive technology enter the bottom of the market, you should not tear down the existing business and make radical changes. Instead, you should create a separate entity in which the new technology can be pursued, outside of the existing value network and cost structure. Resources must be allocated, and not put at risk of being pulled back onto the core business. Expectations on short-term returns should be low.

In the early days of Apple (1977), they sold 43,000 Apple II computers. It was a huge success and the basis for the IPO. A decade later, as Apple grew to a $5 billion company, it released the Newton personal computer, and sold 140,000 of them. In such an emerging market, this should have been seen as a success, but because Apple was now a giant, and the sales amounted to about 1% of revenue, it was viewed as a total disaster.

“Small markets cannot satisfy the near-term growth requirements of big organizations.”

In a recent article in HBR, Christensen admits there is no panacea for this dilemma, and the best advice is the one previously mentioned:

“...universally effective responses to disruptive threats remain elusive. Our current belief is that companies should create a separate division that operates under the protection of senior leadership to explore and exploit a new disruptive model.”

It's not clear to me how Christensen views The Lean Startup, but I for one feel this is the most valid and appropriate solution available to handling disruption - from both the incumbent and new entrant's point of view. I won't labor this point, since I've done that elsewhere.

Narrow Theory, Widely Applied

In 2014 Jill Lepore, a Harvard Professor of History, took a swipe at Christensen and his theory of disruptive innovation, claiming that it thrives on panic; either disrupt or be disrupted! Lepore calls into question the method of choosing his case studies - handpicking them to fit the theory, and claims that Christensen ignores factors that don’t support his theory, for example the case of Bucyrus was supposed to show how an incumbent was ravaged by new 'hydraulic technology' upstarts, but in reality Bucyrus was not impacted that much by disruptive innovation itself:

“...actually, between 1962 and 1979 Bucyrus’s sales grew sevenfold and its profits grew twenty-five-fold. Was that so bad? In the nineteen-eighties, Bucyrus suffered. The whole construction-equipment industry did: it was devastated by recession, inflation, the oil crisis, a drop in home building, and the slowing of highway construction … Caterpillar didn’t think [Bucyrus was a failure] when, in 2011, it bought the firm for nearly nine billion dollars.”

There is good and bad in the article. On the bad side, Lepore is criticizing the entire ‘disruption’ movement, and seeking to put Christensen squarely at the center of it. He's made a career out of it, and sold a lot of books, but I'm not sure it is entirely fair to pin a whole movement and its implication of panic on him.

Lepore claims his theories are untouchable, above criticism, and creating an unwarranted sense of panic and obsession over constant change:

“Most big ideas have loud critics. Not disruption. Disruptive innovation as the explanation for how change happens has been subject to little serious criticism … in its modern usage, innovation is the idea of progress jammed into a criticism-proof jack-in-the-box.”

On the good side, she is right that there is a lack of criticism around the idea, and there is a feeling in the air that any criticism is met with a cultish response, “well, if you don’t agree with it, then you’re not in favor of change.” Haydn Shaughnessy, on observing the Lepore article, makes an excellent point that the real problem here is that it has taken an elite figure (Harvard Professor of History) to take on another elite figure (Harvard Professor of Business), highlighting how difficult it can be to have a proper debate around such important ideas which stem from the world of the elites.

“It matters not whether you are pro-Lepore or pro-Christensen, but whether you are pro this debate.”

The issue turns on the question of who owns the term ‘disruptive innovation’. And the fact is that Christensen has been quite clear what his theory is, and what it isn’t (particularly in the recent HBR article). When it comes down to it, we’re talking about a very narrow theory:

  • It only applies to new technology which begins life as a poor alternative to existing offerings. Critically, the product performance must be worse.
    • Uber is therefore not a disruptive innovation to the taxi industry.
  • It only takes hold either at the very foot of the market, or enters a new market altogether. It must not enter at the mainstream or high-end of the market.
    • Apple’s iPhone is therefore not a disruptive innovation for the mobile phone industry.
  • The technology is not disruptive itself (it’s not something a market leader couldn’t feasibly copy).
    • Smaller disk drives were not technically more difficult to manufacture than larger ones, the main players could easily have developed them.
  • It’s fundamentally about the journey upon which it moves over time.
    • Netflix mail order ‘technology’ was not a disruptive innovation, even its online streaming wasn’t. But the journey by which it moved up the market segments marks it out as a disruptive innovation.

The problem for Christensen is that his narrow theory has been applied widely. So widely that it’s been lost in the mix. As Christensen himself notes, most people claiming to understand disruptive innovation simply don’t get it:

“In our experience, too many people who speak of ‘disruption’ have not read a serious book or article on the subject … Many researchers, writers, and consultants use ‘disruptive innovation’ to describe any situation in which an industry is shaken up and previously successful incumbents stumble.”

And on that point, he’s not wrong.

Recommended Reading:

Want more? Listen to our bookclub discussing The Innovator's Dilemma:


Topics: Methods & Frameworks

Subscribe to the Blog

Subscribe to the Blog
Business Innovation Brief