Category: R&D Management

Impact of Incentive Bonus Plan

Here is a cool article from Management Science on Empirical Examination of Goals and Performance-to-Goal Following the Introduction of an Incentive Bonus Plan with Participative Goal Setting:

Prior research documents performance improvements following the implementation of pay-for-performance (PFP) bonus plans. However, bonus plans typically pay for performance relative to a goal, and the manager whose performance is to be evaluated often participates in setting the goal. In these settings, PFP affects managers’ incentive to influence goal levels in addition to affecting performance effort. Prior field research is silent on the effect of PFP on goals, the focus of this paper.

The authors studied retails store performance (I believe retail stores have a much better handle on performance bonuses than most R&D organizations I know)

Using sales and sales goal data from 61 stores of a U.S. retail firm over 10 quarters, we find that the introduction of a performance-based bonus plan with participative goal setting is accompanied by lower goals that are more accurate predictors of subsequent sales performance. Statistical tests indicate that increased goal accuracy is attributable to managers ‘meeting but not beating’ goals and to new information being impounded in goals.

So, managers lower the goals and then meet them!

we find that prior period performance has incremental power to explain goal levels in the postplan period. Our results provide field-based evidence that PFP and participative goal setting affect the level and accuracy of goals, effects that are associated with both information exchange and with managers’ incentives to influence goals.

Take home message is to be very careful with setting up an incentive bonus plan.  In R&D organizations, it is even more difficult because the results are often not measurable and incentives tend to get disconnected from performance to start with.  Please let me know if you would like to discuss this further.


Specification and Design of Embedded Hardware-Software Systems

For last few months, I have been working on developing a new design flow that brings ASIC like reuse and semiconductor like cost curve to all R&D.  The idea is that semiconductors have increased in complexity and performance exponentially, while costs has come down continuously.  How can we replicate the same for all system R&D?

One of the earliest papers on the topic was  Specification and Design of Embedded Hardware-Software Systems.  In retrospect, the place where it should have come up first anyway – system where electronics/semiconductor and other technologies interact.

“System specification and design consists of describing a system’s desired functionality, and of mapping that functionality for implementation on a set of system components, such as processors, ASIC’s, memories, and buses. In this article, we describe the key problems of system specification and design, including specification capture, design exploration, hierarchical modeling, software and hardware synthesis, and cosimulation. We highlight existing tools and methods for solving those problems, and we discuss issues that remain to be solved.”

The paper suggests five tasks:

  1. Specification capture: Specify desired system functionality
  2. Exploration: Explore design alternatives
  3. Specification refinement: Refine specifications based on exploration
  4. Software & Hardware design:
  5. Physical design:
Much more on this in the future.  But a good paper to start thinking about things.

Top 10 R&D spending Firms

Some good benchmarking data from Booz & Co. and Christian Science Monitor at R&D spending: Here are the Top 10 firms:

Apple, Google, and 3M may top Bloomberg’s list of the world’s most innovative companies, but they’re not the biggest research and development spenders – not even part of the Top 20. Out of 1,000 publicly traded companies with the highest R&D spending in 2009, here are the Top 10, according to a survey by management-consulting firm Booz & Co.

Here is the list for 2009 R&D budgets(Clearly dominated by Drug companies):

  1. Roche Holding $9.b
  2. Microsoft $9B
  3. Nokia $8.6B
  4. Toyota $7.8B
  5. Pfizer $7.7B
  6. Novartis $7.5B
  7. Johnson & Johnson $7B
  8. Sanofi-Aventis $6.3B
  9. GlaxoSmithKline $6.2B
  10. Samsung $6B


China’s Drones Raise Eyebrows at Air Show – WSJ.com


Here is an interesting article in the WSJ with significant impact on long-term R&D strategy: China’s Drones Raise Eyebrows at Air Show. 

Western defense officials and experts were surprised to see more than 25 different Chinese models of the unmanned aircraft, known as UAVs, on display at this week’s Zhuhai air show in this southern Chinese city. It was a record number for a country that unveiled its first concept UAVs at the same air show only four years ago, and put a handful on display at the last one in 2008.”

Amazing progress on Chinese front. During the cold war, USA and Russia kept pumping money into R&D.  This long-term research provided sustainable lead to countries and was a source of significant innovations such as ASICs, Interenet, etc.
I think the difference between the cold war and now is the significant increase in the rate at which technology is changing. Slow progress over decades just won’t be sufficient against newcomers because they will be starting from a much more advanced computing platform.  They will be able to model new environments/materials and manufacture with increasingly more capable machines.  In fact, in many cases a long legacy is  a drag on new innovations.
The answer, however, is not the complete elimination of long-range research.  The answer is to develop more robust R&D plans, so that impact of changes in one technology can be propagated quickly across the entire system development.  The answer also is a frequent re-balance of R&D portfolios to account for changing technology/market/geopolitical landscapes.
I guess R&D managers need even more powerful tools and processes.


NSF Innovation Survey

The National Science Foundation has released preliminary results of their innovation survey: nsf.gov – SRS NSF Releases New Statistics on Business Innovation – US National Science Foundation (NSF). Below are some important take aways:
Defintion of what is innovation:

In the Oslo framework, innovation is the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organizational method in business practices, workplace organization or external relations.”[6] Further, “The minimum requirement for an innovation is that the product, process, marketing method or organizational method must be new (or significantly improved) to the firm. This includes products, processes, and methods that firms are the first to develop and those that have been adopted from other firms or organizations.

Lots of interesting data.  Definitely a source to get back to whenever you are trying to benchmark innovation.


Case study on too much of a vision?

The article on Ars Technica about OLPC’s Negroponte offers to help India realize $35 tablet could be an interesting case study on how too definite a vision could actually be counter productive to over all results.  The single minded focus on $100 per laptop could have been a good stretch goal, but then project execution problems and may be technology challenges led to pretty big problems.

OLPC designs low-cost education computing devices for developing countries. The project aimed to produce a ubiquitous $100 laptop that would bring constructivist learning theory to the developing world. The project has fallen far short of fulfilling its initial goals due to serious setbacks, ranging from technical and logistical failuresto divisive ideological conflicts. OLPC was forced to reorganize and downsize much of its development staff last year as its funds dwindled. Despite these cuts, the organization was able to continue moving forward by narrowing its focus and pursuing a less ambitious strategy.

Add to it the problems of multi-organizational R&D where goals of “partners” were often at odds with the overall goals of the project and it became a significant mess:

When the OLPC project first launched, Negroponte argued that OLPC’s agenda could only be achieved by harnessing economies of scale. The laptops would be sold to governments in massive quantities in order to reduce overall manufacturing costs. Negroponte contended that competing efforts and alternative low-cost laptop products were harmful to OLPC’s vision because they would fragment the market and undermine OLPC’s ability to achieve the level of scale that he believed was necessary for success. 

This became a contentious issue that isolated OLPC from potential partners—particularly Intel, which parted ways with OLPC and built its own competing Classmate PC. Intel frowned on OLPC’s one-size-fits-all approach and argued that diverse offerings were needed in order to encourage adoption of low-cost educational computing. 

None of the problems here are easy to solve and I applaud OLPC for achieving everything they have.  However, there are several lessons we could learn for more effective R&D management in large cross-organizational R&D:

  1. Flexibility in vision (not at the expense of drive and focus)
  2. Ability to identify, express clearly and discuss diverse goals
  3. Ensure that some team members are not going to work to weaken the overall effort
  4. More collaboration and inclusiveness

Looks like OLPC is at least becoming more flexible:

The world needs your device and leadership. Your tablet is not an ‘answer’ or ‘competitor’ to OLPC’s XO laptop. It is a member of a family dedicated to creating peace and prosperity through the transformation of education,” Negroponte said in his letter. “[I offer] full access to all of our technology, cost free. I urge you to send a team to MIT and OLPC at your earliest convenience so we can share our results with you.


Strategic Thinking and Dependence on Forecasts

An interesting article in McKinsey Quarterly with significant implications or R&D strategy: Applying global trends: A look at China’s auto industry.  The overall idea is that forecasts are important and can drive strategic plans and R&D plans. However, forecasts are often wrong and R&D organizations need to be careful not to be tied to strongly to one particular forecast.  In fact, strategy / R&D plans need to be flexible enough to support multiple outcomes.

Strategists can challenge conventional wisdom and better prepare for uncertainty by analyzing the complex and not-so-obvious ways global trends interact in their industries.

Here is a good example of how forecasts were wrong in adoption of mobile phones in Africa:

Only a dozen years ago, for example, authoritative predictions for the coming decade envisioned no more than a few million mobile-phone users throughout Africa. Local income, consumption, technology, infrastructure, and regulatory conditions seemed to hold little promise for significant growth. Less than ten years later, though, Nigeria alone had 42 million mobile subscribers—80 times more than initial forecasts predicted—as growth skyrocketed, largely as a result of the interaction between just two trends: improved income levels and cheaper handsets.

Scenario analysis is an effective approach to address such vast mismatches between expectations and reality:  The article proposes a four step approach that might facilitate better scenario analysis:

1. Establish the reference frame: Scenarios are difficult to construct.  Sometimes, it is possible to find published scenarios that might form the foundation for analysis (For example, CIA  global economic/political scenarios).  However, in absence of such background material, it helps to have a frame of reference.  For example, as the article points out, if one is worried about Chinese automotive entry in USA, it might be worth thinking through Korean market entry back in the 80s.  Or may be, Japanese entry even before.  It might also be worth looking at other industries and see what has happened in them.

The right frame of reference—a specific problem statement and a clear sense of the industry context for long-term shifts—is a critical starting point.

2. Expand the solution space: Now it is time to modify/update the reference frame based on current geopolitical scenarios, technology and industry specific environment:

Having carefully defined the problem and the industry context surrounding it, the challenge for strategists is to broaden the potential solution space by challenging conventional wisdom through the lens of global trends. Most companies have a broad range of experts who can help, yet these people are often tucked away in organizational silos that make it difficult for them to connect the dots.

3. Define scenarios
4: Quantify Industry Impact (under different scenarios).
The article also had a great listing of the Delphi tool that is helpful.


Finding Competitive Advantage in Adversity

HBR article Finding Competitive Advantage in Adversity has some important learnings for R&D managers:

Unlike many managers whose instincts are to hunker down and play it safe during difficult times, entrepreneurs like Bush hear a call to action in the oft-repeated advice of Machiavelli: “Never waste the opportunities offered by a good crisis.”

Opportunity 1: Match Unneeded Resources to Unmet Needs – That is to find resources that are not being used because of adversity and then repurpose or redirect them to meet unmet needs. This is extremely important in R&D management. As budgets are reduced, a lot of projects are cancelled midstream. Wise managers will build on projects being cancelled to deliver something new.

Adversity comes in many forms—acute, cyclical, long-term, and systemic. It sometimes affects individuals or single firms; other times it cuts across a wide swath of entities. However, its pathology is consistent: Adversity constrains a key resource, which then depresses demand, supply, or both. That gives rise to unmet need and releases other resources that become redundant. An opportunity emerges for inventive entrepreneurs who can reroute the redundant resources to fill the unmet need.

Opportunity 2: Round Up Unusual Suspects – somewhat less directly relevant to R&D. However, take away is to look for usual areas for cost cutting and unlikely areas for reuse.

Adversity is also characterized by missing or inadequate elements at critical points in the business system. These may include key inputs, capital, technologies, or partners in the supply, distribution, and marketing chains. Entrepreneurs who can creatively identify unlikely, alternative candidates are able to get a leg up. However, the art of aligning the incentives of an unorthodox coalition and maintaining equilibrium among the members is no small challenge.

Opportunity 3: Find Small Solutions to Big Problems – Take away is that do not make a big project out of repurposing R&D projects that are being cancelled (self explanatory).

The more severe the adversity, the harder it is to change the status quo. Comprehensive solutions that require many changes can appear to be dead on arrival, leaving only tiny cracks as points of entry to break the mold. The message for the intrepid entrepreneur: Small innovations can be huge. First, they are potentially more affordable and can be produced with less initial outlay. Second, they economize on features and complexity and may be just good enough to fulfill an unmet need. Third, their size can help minimize environmental effects or other negative externalities. Finally, they may be easier to integrate into the current model, with only minimal adjustments. In fact, four characteristics that, according to Trendwatching.com, define future consumer priorities may be the tiny cracks to look for: affordability, simplicity/convenience, sustainability, and design informed by local knowledge about product usage. Small solutions that fit within these tiny cracks represent major opportunities.

Opportunity 4: Think Platform, Not Just Product – This is probably the most important consideration for R&D managers.  By thinking long-term and focusing on platforms, the costs can be spread out over longer term.  The benefits can also be larger because platforms have larger revenue potentials.  Finally, stretching out R&D allows organizations to target solutions for when the adversity is mitigated.

In general, the underlying factors that constrain one situation of adversity also constrain others. This offers an opportunity to invest in a meta-solution that can address several unmet needs simultaneously, either in multiple market segments or various product markets. The multifaceted character of the opportunity also hedges the entrepreneur’s risk and helps the venture grow beyond the initial point of entry. Clearly, entrepreneurs can expect varying levels of success, but the broader the venture’s reach is, the greater the value to be unlocked. The profit potential comes from the capacity to enhance the business model at three possible leverage points: customer value, cost management, and growth-vector creation.


Exploratory and Exploitative Market Learning

A quick note from Using Exploratory and Exploitative Market Learning for New Product Development divides R&D organizational learning into two types: Exploratory and Exploitative:

More specifically, this study argues that exploratory market learning contributes to the differentiation of the new product because it involves the firm’s learning about uncertain and new opportunities through the acquisition of knowledge distant from existing organizational skills and experiences. By contrast, this study posits that exploitative market learning enhances cost efficiency in developing new products as it aims to best use the currently available market information that is closely related to existing organizational experience.

So we can think about Innovation as exploratory learning and invention as exploitative learning.  The paper explores this theory based on a survey:

“This study is based on survey data from 157 manufacturing firms in China that encompass various industries. The empirical findings support the two-dimensional market learning efforts that increase new product differentiation and cost efficiency, respectively. The study confirms that exploratory market learning becomes more effective under a turbulent market environment and that exploitative market learning is more contributive when competitive intensity is high. It also suggests that because of their differential direct and moderating effects on new product advantage either exploratory or exploitative market learning may not be used exclusively, but the two should be implemented in parallel. Such learning implementations will help to secure both the feature and cost-based new product advantage components and will consequently lead to the new product success.

To summarize: Innovation is useful (and probably caused by) a turbulent market with lots of changes and discontinuities, while invention or sustaining development is useful in stable / competitive markets.