Nokia R&D spending – a lesson in Portfolio Balancing

In a previous post, I discussed how bureaucracy stifled innovation at Nokia. I ventured a guess that a formalized R&D portfolio balancing process might have helped counteract the effects of bureaucracy. An engadget post finally led me to an article with some concrete data and useful learnings for R&D managers.

First observation is that Nokia’s R&D budget was almost five times that of Apple in 2007. Furthermore, Nokia is focused only on mobile communications, as opposed to Apple, which has significant markets outside of iPhone. Hence, clearly, Nokia’s problems with innovation did not come from lack of R&D budget.

Now lets look at the breakdown of R&D investments across the development portfolio:

It can be seen that Nokia invested as much in hardware development as on Symbian. Each of these were as large as Apple’s TOTAL R&D budget. So there are three fundamental problems here:

R&D budget is disproportionately large compared to competitors
Spending on software and OS is much larger than competitors
R&D investment is clearly NOT delivering the kind of performance competitors are able to get

The question Nokia should be asking when distributing R&D budgets are:

What is the right amount of R&D budget. A bottoms up analysis is one way to get to an answer. The other would be to compute affordability as a fraction of sales. A competitive positioning graph such as the one above could provide some much needed perspective.
What should Nokia expect in return for the R&D investment? It is absolutely critical to have CLEAR top level objectives and metrics. These objectives can help focus R&D community and drive innovation. A benchmarking study can help decide if the objectives are reasonable. For example, if Apple can develop a hardware / software ecosystem for $300M / year, should it not be possible for Nokia to achieve at least as much (may be much more because of economies of scale).
How should the R&D budget bet distributed across different requirements – Hardware vs. software vs long-term research? Again, a bottoms-up estimate can be a good starting point, but it has to be balanced by a external perspective. If none of the Nokia competitors are spending $1.2B in software development, what is the business case for doing so at Nokia? As we get into details, competitive information will be hard to find. The answer is to set up some objectives / metrics and then track performance. Many R&D managers do not track results because it might take a year or two to understand the consequences. However, not tracking means there is no way to change the direction or even to gage if the investments are performing. More on this later….

In summary, R&D investments in this changing and highly competitive environment need to be guided by 1) clear objectives, 2) precise metrics for objectives (both qualitative and quantitative), 3) competitive benchmarking, and 4) longer-term metrics tracking.

At Apple, the Platform Is the Engine of Growth – NYTimes.com

The article At Apple, the Platform Is the Engine of Growth: has some good lessons about platform-based R&D management:

Apple has hit that magical combination of gradually shifting from a product to a platform strategy,” says Michael A. Cusumano, a professor at the Sloan School of Management at M.I.T. and author of “Staying Power: Six Enduring Principles for Managing Strategy and Innovation in an Uncertain World” (Oxford University Press, 2010).

We have discussed Steve Job’s uncanny ability to find duds in Apple’s R&D portfolio. I think platform-based portfolio management is even more important than eliminating poor projects. It is not just about selling a single product – like a great phone. But about developing an ecosystem of related products and services that provide a much better value to the customers.

Apple provides the underlying technology and marketplace: iTunes software and the iTunes Store for managing, downloading and buying music and media; iPhone and iPad software for creating applications; and the App Store for sampling and buying them.

R&D managers need to look at this carefully. If we can make portfolios of related projects, each connected based on underlying platforms, we can do a much better job of funding R&D and ensuring all the different pieces are available simultaneously. This is clearly not easy. Even Apple has had trouble:
Things look different for Apple in the market for mobile devices. There is some debate whether

Apple has become a platform strategist by design or by default: When it introduced the iPhone in 2007, Apple did not have software tools for outside developers to make applications. That came in 2008.
“The iPhone was such a great product that lots of people wanted to write applications for it,” says Marco Iansiti, a professor at the Harvard Business School. “This was a case of the hit leading to the platform, and not necessarily voluntarily for Apple.”

But then they have Jobs… He will take fix the problems in the portfolio, right?

Intel Spreads Its University Research Bets

NY Times had a brief article on open innovation in Intel Spreads Its University Research Bets.

So Intel, the chip giant, has decided to take a more distributed approach to financing university research. The company said this week that it would pour $100 million over the next five years into projects at universities. Each of the projects will involve a few Intel researchers, typically four, with far-flung teams of researchers from several universities.

One key lesson here is teaming between Intel and academic teams. I think this is key to avoiding foreign body rejection in open innovation. Some Intel team members will feel ownership is the accessed innovation and will probably be proponents to get the innovation integrated into internal R&D. Another advantage here is that internal researchers remain “fresh” and involved in innovation. I think this is a great idea.

Intel has been accessing innovations from universities just like other companies. As the Intel press release points out, the old approach was to have collaboration centers near universities. The new approach is to set up innovation centers that are jointly run by the university and Intel.

Until now, Intel Labs ran open collaboration centers near research universities and a substantial portion of the company’s funding focused on operating, maintaining and staffing these facilities. The new centers will be Intel-funded and jointly led by Intel and university researchers. They are designed to providemore dollars in the hands of researchers, and to encourage tighter collaboration between academic thought leaders in essential technology areas such as visual computing, security and mobile computing.

This is an intriguing idea. Clearly, having universities integrally involved in management will add to the innovative environment. However, it will also reduce control and visibility because the environment will be less structured. Another interesting idea is the focus of innovation. As you can see above, Intel has already defined the areas in which it will access innovation. This will hopefully help provide some structure to these centers. They have even a tighter focus on the first center:

As an initial step, Intel Labs also announced that Stanford University will be the hub for the first center, which will focus on improving visual computing experiences for consumers and professionals. Researchers at Stanford will collaborate with a community of researchers from seven other universities. The recently introduced 2nd Generation Intel® Core™ processor with combined visual and 3-D graphics will be a key R&D platform for researchers to develop innovations which improve the quality and the way images are captured or created, manipulated or interpreted and ultimately displayed to the viewer.

I look forward to seeing how this experiment works out. Clearly, we can learn a lot from this.

Toyota’s ‘Devil’s Advocate’ Policy Looks to Restore Quality

We have talked about Toyota recalls and the announcement by Mr. Todyoda to fix the problems. You might recall that Mr. Toyoda’s solution to reducing errors was to increase quality control. Well, Toyota has followed through. The article Toyota’s ‘Devil’s Advocate’ Policy Looks to Restore Quality provides some details.

The ‘devil’s advocate’ approach to vehicle design is a key element of the new Toyota. Under the plan, the company gives engineers four extra weeks to tear down and evaluate new vehicles.

The goal is to use the car in ways the owner’s manual doesn’t even consider. That’s because Toyota found out the hard way last year that customers use cars in unpredictable ways. It traced some unintended acceleration cases to gas pedals being jammed by stacked floor mats — an ill-advised practice for which Toyota engineers didn’t plan.

Interesting! So, if the customers use the car in an unpredictable way, why think of it AFTER the design is complete? Also, if the use is unpredictable, how can you account for it? If it can be predicted, why not change design objectives to address those use cases? You might have guessed it: I do not believe this way of implementing a devil’s advocate process will do much to mitigate risks of recall.

But not everyone agrees Toyota has solved the problems that led to last year’s crisis. One skeptic is Mr. Ditlow of the Center for Auto Safety. He said Toyota’s new approach to dealing with complaints of quality glitches such as unintended acceleration still is “pretty much the same-old, same-old.”

What is needed is a new process to manage automotive R&D. You might also recall that the root cause of the recalls was NOT poor quality control, but a large increase in complexity or vehicles being developed. I believe that adding extra quality control steps at the END of development does little to address the complexity. Here is what I suggest:

Make sure the system objectives are defined correctly. Use the devil’s advocates in helping balance and guide objectives.
Link objectives to detailed requirements across the entire system hierarchy
Link requirements to risks of failure. That is, perform concurrent risk assessment and integrate quality control in the design process.
Integrate devil’s advocates in the risk assessment process and fix it while there is time!

If you don’t do this, you run into many problems. The so-called Toyota Devil’s Advocate quality control process did this :

Toyota and the supplier switched to crisis mode. They designed a new pinchless wiper and the Yaris still made the scheduled start of production in November.
“We really had to push hard,” recalled Katsutoshi Sakata, Toyota Motor Corp.’s lead executive for quality research and development. “But there’s a new mindset here that we will address even the smallest of issues.”

We can surely do better.

GE – Interesting perspectives on R&D strategy

The NY Times article For G.E. and Jeffrey Immelt, a Return to Basics has some interesting sound bites from the CEO Jeffery Immelt:

Strategies are useful, he says, but only if they can quickly adjust to nasty real-world surprises. “In the words of the great philosopher Mike Tyson,” Mr. Immelt says, smiling, “everybody has a plan till they get punched in the mouth.”

The focus seems to be adaptability in face of a rapidly changing environment

And what wisdom is on tap at the United States Military Academy? “Adaptability” and “resiliency” amid uncertainty, says Mr. Immelt — skills as vital to surviving in business as they are on the battlefield.

I think effective R&D planning and portfolio management is key to adaptability. If market forces require a change in products under development, managers need to have detailed visibility into every single R&D project. This access needs to be so comprehensive that executives can quickly perform what-if scenarios on what new products can be developed by minimal tweaking of existing projects. More on this below.

The other GE strategic focus is Technology-differentiated Manufacturing:

He’s most animated talking about heavyweight products that take patience and piles of cash to develop, weigh tons and last for years — next-generation jet engines, power turbines, locomotives, nuclear plants, water-treatment systems, medical-imaging equipment, solar panels and windmills. Mr. Immelt notes, for example, that the cost of a good-sized solar-panel plant, about $70 million, is more than twice the total investment in Google in the six years before it went public in 2004.
The costs and complexities of such businesses, he adds, make it hard for just any company to compete. These are markets, he says, that have “big moats. They’re tough to get in.”

So, if there is significant investment (both capital and R&D) in developing these moats, then being adaptable is going to be rather difficult. So, if the jet engine market shifts from higher efficiency / lower operating costs to higher speeds, how will GE be able to respond? Or if the solar business sees new entrants with low cost organic cells, how will GE be able to change its $70M plant? This will be rather difficult:

Despite the financial crisis and recession, Mr. Immelt has kept investing for the long haul. Research and development spending increased last year to $3.3 billion, and will be still higher this year. “It would have been easy to say times are tough and we’ll pull back on research spending and long-range projects, but he didn’t do any of that,” says Vijay Govindarajan, a professor at the Tuck School of Business at Dartmouth and a former consultant to G.E. “Jeff Immelt really held onto his technology-led-innovation agenda.”

It is not just that the Mr. Immelt held onto his strategy, it is also that it is very difficult to change these strategies mid-stream. Most differentiating technologies take years to develop and a start-stop-restart approach just will not work. Furthermore, most R&D is still managed using gut feelings. It is extremely difficult to adapt to new markets when one does not know what drove old decisions to start with… We need a new paradigm to managing R&D!

Apple’s R&D portfolio strategy – “Get Rid of the Crappy Stuff”

Some interesting R&D portfolio management advice in the Fast Company article Steve Jobs’s Strategy? “Get Rid of the Crappy Stuff”

“Do you have any advice?” Parker asked Jobs.
“Well, just one thing,” said Jobs. “Nike makes some of the best products in the world. Products that you lust after. Absolutely beautiful, stunning products. But you also make a lot of crap. Just get rid of the crappy stuff and focus on the good stuff.”

There are two fundamental problems in executing this advice:

How do you decide what is crap? Sales to date, forecasts, executive opinion?
Why did the crap get into the portfolio to start with? Should it not have been eliminated before development was concluded?

Clearly an executive like Steve Jobs can answer these questions. As the article points out:

It takes courage to reduce the number of products a company offers from 350 to 10, as Jobs did in 1998. It takes courage to remove a keyboard from the face of a smartphone and replace those buttons with a giant screen, as Jobs did with the iPhone. It takes courage to eliminate code from an operating system to make it more stable and reliable, as Apple did with Snow Leopard.

In most companies, answers to two questions are much more difficult. There are pet projects for executives that can not be touched. A R&D portfolio manager I know went through an elaborate exercise to evaluate the portfolio and find duds. He found many projects that had gone on for years and were no closer to delivering results. He brought the list up to the senior executive’s attention. The answer from the CEO was immediate and clear – you can not touch these. No explanation was given nor any improvements suggested to the portfolio management process to ensure it did not bring up the same projects again.

Another problem is interdependencies between different projects in the R&D portfolio. Killing one project means others get impacted. Companies need to really invest in maturing and enhancing their portfolio balancing processes.

A Bright Future Requires Bright Ideas from R&D

Another survey of executives and another validation that innovation (or any current hot buzz word) is key. This time from Corporate Executive Board in A Bright Future Requires Bright Ideas from R&D. We now have a significant increase in number of executives wanting innovation…

From CEB: Number of executives wanting innovation drive growth

Really? Is innovation driven growth better than any other growth? Here is the key message

The best way for R&D to help their companies is to focus on the beginning of the innovation process – the ‘front end’. This is the part of the process where concepts for new products and services are sourced or created. RTEC researchers found that firms that were best at sourcing big new commercially viable ideas were good at two things: Securing organizational buy-in for funding big growth bets. Ensuring that growth ideas are differentiated and connected to pressing market needs.

I guess those are good qualities to have. As we have discussed before, the key problem with open innovation is inability to overcome foreign body rejection tendencies of R&D organizations…

McKinsey Survey on Innovation and commercialization 2010

Another survey on innovation with some interesting data (you will find many more here)

As companies begin to refocus on growth, innovation has once again become a priority: in a recent McKinsey Global Survey,1 84 percent of executives say innovation is extremely or very important to their companies’ growth strategy. The results also show that the approach companies use to generate good ideas and turn them into products and services has changed little since before the crisis, and not because executives thought what they were doing worked perfectly.

Further, many of the challenges—finding the right talent, encouraging collaboration and risk taking, organizing the innovation process from beginning to end—are remarkably consistent. Indeed, surveys over the past few years suggest that the core barriers to successful innovation haven’t changed, and companies have made little progress in surmounting them.

So, we all agree that innovation is important and we recognize the challenges in becoming innovative. BUT, we continue to use same old management processes and hoping that we will get new results! Even more importantly, we continue to believe that we are good at innovation, but not really know why we are better than others…

Just over half of all respondents, 55 percent, say their companies are better than their peers at innovation, a figure that hasn’t budged since 2008. Another consistent pattern is that far fewer respondents say their companies are good at the specific processes and tactics frequently tied to successful innovation—such as generating breakthrough ideas, selecting the right ideas, prototyping, and developing business cases. Respondents say their companies are best at adapting once they’re in the market, with 58 percent claiming to be successful. As in the past, executives have the most difficulty stopping ideas at the right time, with only 26 percent of respondents to this survey saying they do this well.

The graph below says a lot:

Only 39 percent of respondents say their companies are good at commercializing new products or services. This overall assessment seems to have a few different sources (Exhibit 4). Commercialization was a serious concern in 2007 as well; in that year’s survey, nearly a third of senior leaders selected making handoffs from ideas to commercialization as one of their biggest challenges, and 43 percent said the top challenges included choosing which ideas to move forward.

Much more good info in the article…

Down with fun

Another great article by Joseph Schumpeter in the Economist: Down with fun.

These days many companies are obsessed with fun. Software firms in Silicon Valley have installed rock-climbing walls in their reception areas and put inflatable animals in their offices. Wal-Mart orders its cashiers to smile at all and sundry. The cult of fun has spread like some disgusting haemorrhagic disease. Acclaris, an American IT company, has a “chief fun officer”. TD Bank, the American arm of Canada’s Toronto Dominion, has a “Wow!” department that dispatches costume-clad teams to “surprise and delight” successful workers. Red Bull, a drinks firm, has installed a slide in its London office.

I am not sure how important FUN is to R&D teams.

This cult of fun is driven by three of the most popular management fads of the moment: empowerment, engagement and creativity. Many companies pride themselves on devolving power to front-line workers. But surveys show that only 20% of workers are “fully engaged with their job”. Even fewer are creative. Managers hope that “fun” will magically make workers more engaged and creative. But the problem is that as soon as fun becomes part of a corporate strategy it ceases to be fun and becomes its opposite—at best an empty shell and at worst a tiresome imposition.

This does feel like an excuse for poor R&D management in my opinion. I do not know what makes my employees more innovative or more effective. I am going to assume that creativity is related to fun. So, I am going to encourage my workers to have more fun at work! As the author rightly points out:

Behind the “fun” façade there often lurks some crude management thinking: a desire to brand the company as better than its rivals, or a plan to boost productivity through team-building. Twitter even boasts that it has “worked hard to create an environment that spawns productivity and happiness”.

And here is the clincher…

The most unpleasant thing about the fashion for fun is that it is mixed with a large dose of coercion. Companies such as Zappos don’t merely celebrate wackiness. They more or less require it. Compulsory fun is nearly always cringe-making.
…
Wal-Mart tried to impose alien rules on its German staff—such as compulsory smiling and a ban on affairs with co-workers…

The innovation machine

The Economist has an interesting article by Joseph Schumpeter The innovation machine:

Hardly a week passes without someone publishing a book on the subject. Most are rubbish.

I tend to agree. There is so much hype around innovation, but no one really defines what they believe is innovation. Even the Oslo Framework definition talks about new products and new methods. What is the difference between invention and innovation? All new product lines automatically become innovation? I am thinking we should focus on R&D effectiveness and not really worry if it is innovation or not. Anyway, back to the article:

Last year Mr Govindarajan and Mr Trimble (hereafter: G&T) published a seminal article, with Jeff Immelt, the head of General Electric, on frugal innovation. In their new book they address two subjects that are usually given short shrift: established companies rather than start-ups and the implementation of new ideas rather than their generation.

I like frugal innovation as a concept because it focuses the R&D team and drives innovation. The two new thrusts are:

The fashion these days is to focus on the supply side of innovation: for example, by encouraging everyone to think big thoughts. 3M, the maker of Post-it notes, expects its workers to spend 15% of their time on their own projects. Google expects them to spend 20%. This approach is attractively democratic: by giving everyone a chance to innovate, it makes everyone feel special. Or so the theory goes. G&T are ready with the cold water. The let-them-loose approach spreads resources thinly and indiscriminately. Companies dissolve into a thousand small initiatives rather than focusing on a few big problems. It also produces far too many ideas: managers have to spend weeks sorting through the chaff to find a few grains of wheat.

That is give people time to think of new things and innovation will happen. I agree that is a good idea. However, it is important to remember that R&D management becomes even more important when you are letting technologists roam free. One needs to guide and nurture right ideas and to do so requires a good evaluation scheme to identify those ideas that are worthwhile.

A second approach focuses on closing the loop between ideas and results. Nucor Corporation, a steelmaker, gives its workers bonuses if they can produce steel more efficiently. Deere & Company, a maker of farm machinery, has produced a detailed playbook on how to design new tractors. G&T concede that this approach is an excellent way of making incremental improvements to existing products and processes, but suggest that it has little chance of producing a big breakthrough.

Interesting debate! More on it below. However, the authors insist that the only way (or the primary way) to drive innovation is to break out a group that can innovate.

G&T argue that companies need to build dedicated innovation machines. These machines need to be free to recruit people from outside (since big companies tend to attract company men rather than rule-breakers). They also need to be free from some of the measures that prevail in the rest of the company. But they must avoid becoming skunk works. They need to be integrated with the rest of the company—they must share some staff, for example, and they must tap into the wider company’s resources as they turn ideas into products. And they must be tightly managed according to customised rather than generic rules. For example, they should be held accountable for their ability to learn from mistakes rather than for their ability to hit their budgets.

I have seen many innovative ideas rejected because they were not invented by the team. Setting up a separate organization just brings foreign body rejection a little bit less difficult. More often than not, the innovation organization becomes an outcast and root cause of tremendous organizational conflict. This has happened many times with the original Skunk Works (at Lockheed) it self. Many people do not realize that Skunk Works was successful because they were paid by the government to do work. They were a profit center. Trying to replicate Skunk Works in non government area is going to be quite difficult. More importantly, innovation to invention transition from Skunk Works has been far from effective.
So, what are my recommendations? Effective R&D management, of course! Only managers can provide challenges that help teams innovate, identify bright ideas and fund them so that they can be transitioned into inventions and products and avoid organizational conflict / foreign body rejection. Much more stuff is here.