Advertisement

http://insurancethoughtleadership.com/wp-content/uploads/2014/04/bg-h1.png

Facebooktwitterredditpinterestlinkedinmail

June 3, 2015

The Dangers of Standing Still

Summary:

By being cautious on driverless cars, Nissan may underscore the lesson of Kodak about the perils of standing still in innovative times.

Photo Courtesy of Emran Kassim

One of the most telling episodes of Kodak’s slide into bankruptcy was how it incorporated digital capabilities into its Advantix camera system.

Kodak spent more than $500 million to develop and launch the Advantix in 1996. The system capitalized on emerging digital capabilities— especially the digital sensors that Kodak engineers had invented two decades earlier—to capture date, time, shutter speed and lighting conditions to produce better pictures. The strategy culminated in the Advantix Preview camera, which allowed photographers to preview shots and mark how many prints they wanted. Kodak gave users no ability to save the digital images, however. The Advantix required traditional silver halide film and prints.

Advantix flopped. Why buy a digital camera and still pay for film and prints? Kodak wrote off almost the entire cost of development.

Kodak’s strategic blunder was not because of a lack of technological prowess; it was because of an inability to embrace business model innovation. Kodak was the market-leading photo film, chemical and paper business. It bet its future on “the hope that demand for digital images would sell more film.” As a result, Kodak protected its traditional business to the bitter end—until others leveraged digital to make film irrelevant.

Judging from recent comments by Carlos Ghosn, Nissan’s chief executive, we might one day read about how Nissan repeated the pattern of Kodak’s decades-long blunder and demonstrated the dangers of standing still during a period of industry innovation (like what’s happening in insurance).

Ghosn has championed his company’s efforts to develop autonomous driving technologies to allow cars to operate without human intervention. And, unlike some other large automakers (such as Toyota), Ghosn does not dispute the technical feasibility of driverless cars.

But Ghosn views the choice of semi-autonomous vs. driverless as a strategic decision—and he is clear that his choice is to use autonomous technologies as incremental enhancements to cars with drivers. As reported by the Associated Press via the New York Times: Ghosn said Nissan sees autonomous vehicles as adding to driving pleasure, and a totally driverless car is not at the center of the automaker’s plans. The autonomous driving Nissan foresees will assist or enhance driving. Nissan may end up with a driverless car, but that was not the automaker’s goal, he said. “That is the car of the future. But the consumer is more conservative. That makes us cautious.”

In other words, Ghosn’s strategy is to hope that the demand for autonomous technologies will sell more cars. Like Kodak, he is aiming to reinforce Nissan’s current business model rather than embrace business model innovation.

By being cautious, however, Ghosn risks emulating Kodak’s failure by waiting for others to leverage driverless technologies to make traditional cars irrelevant. He also risks ceding emerging business innovations to Google, Uber and others willing to make driverless cars their explicit primary goal.

The unanswered question is whether Ghosn, behind the scenes, is parlaying his technological forward-mindedness into strategic preparedness.

Carlos Ghosn need not shed his caution. But, as I previously argued, trillions hang in the balance. Given those stakes, has Ghosn hedged Nissan’s strategic bets in case the driverless “car of the future” comes more quickly than he expects?

Some argue that, of course, Nissan won’t be caught flat-footed even if driverless cars come sooner than expected. Look, for example, at its research partnership with NASA. But research is not enough.

A trap that market-leading companies fall into is believing that they can catch up if their initially cautious strategies turn out to be wrong. One lesson that Paul Carroll and I found in our study of thousands of large company failures is that it is very hard to excise denial from multiple layers of the organization—even after objective evidence argues for doing so. Another lesson is that, while it is possible to catch up on raw technical expertise, it is hard to catch up after yielding multiple product-oriented learning cycles to competitors.

Take electric hybrid cars. A former senior technologist of one of the big automakers told me his company considered but rejected hybrid electric cars before Toyota launched the Prius. The automaker was at first dismissive of the Prius and then surprised by its market success. It did jump into the market with its own offering. But, the technologist bemoaned, it has not been able to catch up. With each model, Toyota gets further ahead. The company ceded too many learning cycles to Toyota.

The same could be happening with driverless cars.

Nissan espouses caution about driverless cars. Whatever research is going on in its labs is mostly hidden from the public (perhaps to not confuse the market or provide succor to competing strategies).

Google, on the other hand, will soon release 25 prototype driverless cars onto the streets of Mountain View, with plans to launch 75 more. Google’s self-driving cars have logged a collective 1.7 million miles and are adding about 10,000 miles per week, mostly on city streets. Google has not cracked all the issues involved with driverless cars. It has, however, created the ability to learn faster.

Kodak, as evidenced by its own tongue-in-cheek marketing video, ended up play “grab ass” for years with digital photography. Late attempts to “get serious” were too late. Even now, 40 years after Kodak engineer Steven Sasson invented the digital still camera, Kodak still struggles to realize the potential of its IP portfolio.

Likewise, every market-leading department retailer of the 1950s and ’60s, such as Macy’s, Woolworth’s and Ames, thought it could contend with discount retailers like Wal-Mart if the need arose.

Only Dayton Hudson took the discounting business model seriously. Rather than watch and wait, Dayton Hudson formed a discounting business unit and unleashed that subsidiary to compete as hard as possible against the traditional business. That discount subsidiary was named Target. Of the more than 300 department-store chains in the U.S. in the late 1950s, only Dayton Hudson/Target successfully moved into discount retailing. Most of the others preceded Kodak on the path to bankruptcy.

Rather than following in the footsteps of Kodak and all those defunct department stores, Nissan should be more like Dayton Hudson.

Instead of just betting on caution, Nissan should also unleash innovators to create its own driverless offering and charge them with competing as hard as possible against its traditional business.

description_here

About the Author

Chunka Mui is the co-author of the best-selling Unleashing the Killer App: Digital Strategies for Market Dominance, which in 2005 the Wall Street Journal named one of the five best books on business and the Internet. He also cowrote Billion Dollar Lessons: What You Can Learn from the Most Inexcusable Business Failures of the Last 25 Years.

+ READ MORE about this author ...

Like this Post? Share it!

Add a Comment or Ask a Question

blog comments powered by Disqus
Do NOT follow this link or you will be banned from the site!