Intel Disrupted: Why large companies find it difficult to innovate, and what they can do about it

In the 21st century it’s harder for large corporations to create disruptive breakthroughs. Disruptive innovations are coming from startups – Tesla for automobiles, Uber for taxis, Airbnb for hotel rentals, Netflix for video rentals and Facebook for media.

What’s holding large companies back? Here are four reasons:

First, companies bought into the false premise that they exist to maximize shareholder value – which said “keep the stock price high.” As a consequence, corporations used metrics like return on net assets (RONA), return on capital deployed, and internal rate of return (IRR) to measure efficiency. These metrics make it difficult for a company that wants to invest in long-term innovation. It’s a lot easier to get these numbers to look great by outsourcing everything, getting assets off the balance sheet and only investing in things that pay off fast. To do that, companies jettisoned internal R&D labs, outsourced manufacturing and cut long-term investment. These resulting business models made them look incredibly profitable.

Second, the leaders of these companies tended to be those who excelled at finance, supply chain or production. They knew how to execute the current business model.

Intel under their last two CEOs delivered more revenue and profit than any ever before. They could point to record investment in R&D for more expensive chip fabs yet today the writing is on the wall that Intel’s leading days are over.  Why?

Over the last decade, Intel missed two important disruptive trends. First, the shift away from desktop computers to mobile devices meant that Intel’s power-hungry x86 processors weren’t suitable. ARM, a competitor, not only had a better, much lower power processor, but a better business model – they licensed their architecture to other companies that designed their own products. Intel attempted to compete, (and actually owned an ARM license) but fell victim to a classic failure of ignoring a low-end disruptor and hobbling their own chances by deciding not cannibalize their own very profitable x86 business. All of Intel’s resources – fabs, manufacturing strategies, and most importantly executive mindset — were geared towards large, expensive x86 processors, not low-cost mobile cores of someone else’s design.

The result, Intel just laid off 12,000 people, 11% of their company.roadkill

But it’s not over for Intel. Their most profitable segment is very high-end processors used in data centers in servers and the cloud. Today that’s built on the premise that an x86 architecture is the one best suited for big data. It’s becoming clear that extracting intelligence from that big data requires machine learning architectures which are better implemented with non x86 chips from companies like NVidia. It’s possible that by the end of this decade history might repeat itself in Intel’s most profitable segment.

The third reason why companies find it hard to innovate is the explosive shifts in technology, platforms and markets that have occurred in the last 15 years–personal computers moving to mobile devices; life science breakthroughs in therapeutics, diagnostics, devices and digital health; and new markets like China emerging as consumers and suppliers.

Which brings us to the fourth reason it’s harder for large corporations to offer disruptive breakthroughs: startups.

For the first 75 years of the 20th century, when capital for new ventures was scarce, the smartest engineering talent went to corporate R&D labs.

But starting in the last quarter of that century and accelerating in this one, a new form of financing – risk capital (angel and venture capital) — emerged. Risk capital has provided financing for new ideas in the form of startups. Capital is returned to these investors through liquidity events (originally public offerings, but today mostly acquisitions).

Startups have realized that large companies are vulnerable because of the very things that have made them large and profitable: by focusing on maximizing shareholder return, they’ve jettisoned their ability to do disruptive innovation at speed and scale. In contrast, startups operate with speed and urgency, making decisions with incomplete information. They’re better than large companies at identifying customer needs/problems and finding product/market fit by pivoting rapidly. Their size lets them adopt flatter and more agile organizational structures while providing incentives that reward risk-taking and collaboration.

Startups are unencumbered by the status quo.  They re-envision how an industry can operate and grow, and they focus on better value propositions. On the low-end, they undercut cost structures, resulting in customer migration. At the high-end they create products and services that never existed before.

As we’ve seen, corporations are very good at maintaining, defending and refining existing business models, and they’re pretty good at extending existing models by identifying adjacencies. But corporations are weak, and have become weaker, in identifying new disruption opportunities.

Innovation can come from inside the corporation, by adopting Lean Startup language and methods, developing intrapreneurship, and fostering innovation-driving behaviors such as GE’s FastWorks program. And corporations can foster innovation from the outside by promoting open innovation and buying startup-driven innovation. Google has bought close to 160 companies in the last decade. Its acquisition of Android may have been the biggest bargain in corporate history.

So to succeed, corporations must re-think and then re-invent their corporate innovation model, replacing a static execution model with three horizons of continuous innovation: This requires a corporate culture, organizational structure, and employee incentives that reward innovation. It requires establishing acceptable risk level and innovation KPIs for each horizon.

And it also requires understanding the differences between executing the existing business model, extending the business model and searching for and disrupting the business model.

Lessons Learned

  • Even the most innovative companies eventually become yesterdays news
  • To survive companies need to run three-horizons of innovation
    • Horizon 1 – execute their existing business model(s)
    • Horizon 2 – extend their existing business model(s)
    • And for long-term survival – Horizon 3 – search for and create new/disruptive business model(s)

(this article first appeared in the Peoples Daily.)Peoples Daily

14 Responses

  1. Peoples Daily link is broken. Help


  2. Chunka Mui has made a career telling big companies “You, too, can be innovative; in fact, you have some significant advantages over startups if you can figure out how to use them.”

    I think his Rule #1 is “Don’t let the finance guys anywhere near your innovation teams until late in the process.”


  3. Intel should out-compete by creating a new HW platform and lifecycle that, while sacrificing features and performance, delivers unprecedented and ultra-high-assurance levels of trustworthiness. It would apply extreme trasnaprency and public verification relative to complexity, and meanwhile prevent malevolent use of the publicly verifiable designs.

    It could do that by leading globally the establishment of a Trustless Computing Group (and paradigms), in opposition to the Trusted Computing Group (and paradigms): Here’s how:


  4. I think it is true that it is easier for small companies to get to know their market, but when it is the right moment to start looking for new marketing?


  5. A small quibble: the word “disrupt” gets bandied about a lot, without observing the original meaning Clay Christensen gave it in the landmark Innovator’s Dilemma. As he described it, a disruptor is a supplier who enters the market at the low end, with a cheaper, inferior product that is nonetheless “good enough” for a new class of price-sensitive, previously underserved customers. Over time, the company moves up-market, refining its products until it threatens the incumbent leaders (who tend to be invested in higher-cost business models).
    By that standard, Tesla is not a disruptor. Probably the best example of one in autos is Toyota–>Lexus.


    • @ Robert Parker – thanks for refreshing our memories on the “real” meaning of disruptive technology. I also thought that there was another condition in the marketplace – that being that the product really generated a new market.


      • Well, yes, that’s basically what I meant by “a new class of price-sensitive, previously underserved customers”.


  6. Let’s not forget the “not on my watch CEO” where disruption can effect the clean run to a nice parachute into bliss. Do I come first or the company? Sports teams always fail with that attitude.


  7. Well written! It seemed like the core message was that big enterprises need to innovate rather than be financial focused, got that. Article didn’t really go into how they can do that except for the last three bullets.
    Can we followup with your thoughts in how to go about this? Not sure how big the GE initiative is and also Googles growth by acquisition isn’t really internal innovation story


  8. More than just startups changed the landscape. It used to be that only huge companies had the money, expensive capital equipment and knowledgeable workers to innovate in the key new tech areas and then manufacture at scale. Giant fabs to make chips, scanning electron microscopes to see tiny things, supercomputers to calculate stuff. That now has all changed. In many fields startups can innovate toe-to-toe in some of the most arcane areas, renting supercomputer power, buying time on worldclass semiconductor fabs, runnning complex experiments on automated biology equipment and then get giant ODM’s to manufacture it for you at best-in-class economics.

    Like you pointed out, big-company-itis is a crippling disease for these changing times, but so is the new landscape for doing world-class tech.


  9. Agree with all if your points, Steve. Small typo, Tesla is misspelled first paragraph …


  10. Intel should ragaein a leading edge by creating a new HW platform and lifecycle that, while sacrificing features and performance, delivers unprecedented and ultra-high-assurance levels of trustworthiness.
    Intel would apply these paradigms:
    (1) Complete verifiability, extreme compartmentation and minimization, and sufficiently extreme verification relative to complexity of all critical HW&SW; made possible by an initial extreme minimization of features and performance;
    (2) Extreme oversight centered on offline citizen-witness and citizen-jury processes of ALL critical technical and socio-technical lifecycle components, including ICs fabrication, and server-room access, including for constitutional lawful access requests;
    (3) Highly technically-proficient and citizen-accountable ICT assurance certification.

    It could do that by leading globally the establishment of a TRUSTLESS Computing Group (and paradigms), in opposition to the Trusted Computing Group (and paradigms): Here’s how we are setting out to do it and get it rolling with “ultra-minimum” viable product and certification body for 4-5M€:

    We were inspired to dig deep by Steve Balnk blogpost on Intel in 2013, and found Schneier agreed we should assume all main CPUs to be compromised. Then found that rot goes beyond CPU design and firmware upgrade mgmt, but down to fabrication processes of critical HW parts.

    In TRUSTLESS managed to find a way to economically get even that part to be extremely trustworthy, while at the same time prevent malevolent use, with the CivicSite processes, which we worked on with 4 of our Trustless partners: (1) American Minifoundry (whose tech lead used to be the Technical Director of NSA Trusted Foundry Program); (2) our Brazilian CPU partner, maker of the only general-purpose CPU publicly-verifiable in HW and SW; (3) Lfoundry, the only a independent 200mm sub 110nm left in Europe; (4) The German DFKI, the largest AI R&D center in the world by number of staff and external funds.

    This is what DoD should do by the way to regain democratic top down contral on their techs, through transparency or effective traslucency of all critical internal processes.

    Come join us to an event where we’ll be talking about this stuff and more in NYC on July 21st:


  11. Same plot as in ´90s The Innovator´s Dilemma now revisited to this century. There´s though a spicy ammend: speed. Liquid environments of ever changing technology are the cause of this sort of innovation outsourcing concept. I do not think big companies are weak to innovate. Truth is that startups are strong because hundreds of them are right now working and testing different ideas at the same time. This is the real power of startups.


  12. Same as it ever was, only faster.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: