Should companies outsource innovation?

Outsourcing has had a negative connotation in the US.  However, the trend seems to be here to stay.  As product complexity increases, and it is not easy for companies to do everything in-house.  Furthermore, competitive pressures are compressing product development cycles and companies have to use approaches such as co-development to grow.  If companies are using suppliers for components and subsystems, should they not use them for innovation as well?  Collaboration with strategic suppliers has been shown to enhance innovation.

More and more organizations are sourcing innovation from the outside through open innovation or borderless innovation.  Leading companies such as Intel work with universities to source innovation. Companies from J&J to Ortho have setup innovation hotbeds to access innovation from the outside. There are interesting approaches to crowd-source innovation and reach-out to the masses for innovation.
Even so, there is a significant amount of confusion and controversy around the benefits of innovation outsources.

A great article in R&D Management journal provides a synthesis or current research in innovation outsourcing (Controversy in innovation outsourcing research: review, synthesis and future directions):

There is a growing stream of research into the outsourcing of innovation activities within the innovation, management, marketing and economics disciplines. Understandably, this coincides with the practice becoming more commonplace in industry. Here, we attempt to synthesize research surrounding the question of whether to outsource or internalize innovation activities and the performance implications of this decision.

As expected, there are many open issues and conclusive consensus has not emerged about aspects of innovation outsourcing.  Some research shows that innovation outsourcing reduces costs while other shows that it increases them!

As innovation outsourcing research has progressed, several controversies have emerged in the literature and remain unresolved. For instance, case-based research provides evidence that outsourcing innovation activities can lead to faster product development and cost savings; yet, empirical research shows that outsourcing may lead to higher costs and slower new product development

However, there are some clear trends that R&D managers should pay attention to:

Outsourcing is most likely when specific assets are required, behavioral uncertainty is low, intellectual property is well protected, the activity is not seen as a path to developing competitive advantage and when low cost is not the primary goal of the development effort. Also, large firms have a greater tendency to outsource.

So, it makes sense to outsource when:

  • Supplier has specific equipment, laboratory or assets that are expensive to acquire.
  • It is easy to understand and evaluate innovation results from the supplier (low uncertainty about supplier behavior)
  • Sourced innovation is not critical to the company nor is it a significant competitive advantage (critical innovations are normally controlled by the company)
  • Costs are not a key driver (companies still tend to feel that sourcing innovation is more expensive than doing it in-house)
  • Large firms are more likely to outsource innovation than small firms
There is a lot more info in the article, but it is a bit difficult to read…

Should Performance Reviews Be Fired?

Many observers have suggested that performance reviews cause too much stress (for the benefit generated). Even when reviews work as advertised, R&D managers do not have the right tools to use performance evaluations.  There are no easy ways to reward good performance, as neither financial incentives nor stock options may be very effective.  Furthermore, &D managers do not often have the right approach to address poor performance.  A new article from the Wharton School of Business provides some useful information about performance reviews (Should Performance Reviews Be Fired? – Knowledge@Wharton):

Performance reviews typically are not done often enough and all too often are done poorly. A good performance review gives employees constructive, unbiased feedback on their work. A bad one demonstrates supervisor bias and undermines employee confidence and motivation.

Clearly, a vast majority of companies use performance reviews

Cappelli cites studies showing that 97.2% of U.S. companies have performance appraisals, as do 91% of companies worldwide.

So, what are the main problems with performance reviews and what can we do about it?

  1. Frequency: Annual feedback on performance is just not enough in today’s fast paced world.

    In addition, reviews encourage employees not to speak out about problems they observe because it could adversely affect their career paths and compensation, Culbert states. As examples, he points to “employees at Toyota, BP and the nuclear reactor site in Japan who knew about defects” in their companies’ products, but failed to report them because of a lack of trust between employees and management.

    It might be better to have a project / task based continuous evaluation, a quarterly or semiannual development reviews followed by annual incentive / raise review.

  2. Competing Agendas: Most organizations have two completely different goals for reviews. One is to help employees improve performance and the other is to rank them to decide who gets what rewards. Combining the two in one review ensures neither objectives are satisfied.

    Performance reviews “are rarely authentic conversations,” writes Daniel Pink in “Think Tank.” More often, “they are the West’s form of Kabuki theatre — highly stylized rituals in which people recite predictable lines in a formulaic way and hope the experience ends quickly.”

    As discussed above, it is probably better to separate performance, development and reward reviews from each other.

  3. Performance Metrics: In many cases, especially in R&D, it is not clear how to measure performance and compare it.

    “Companies are concerned that if it isn’t a quantifiable, very objective measure, then it’s not a good measure.”

    However, if we do 1 and 2 above, it is possible to make performance reviews more quantitative:

    But in recent years, with the explosion of knowledge-based companies, “the ability to assess performance in a subjective and qualitative way” requires a process that looks at “first, what are the key performance criteria that are important, and second, how do you measure them when they are qualitative.” He suggests asking employees during the assessment process “how they do their job, what [competencies] they have developed and whether they are continuously improving their knowledge skills.”

  4. Checking the box:  Many managers do reviews just to check the box. This actually is the worst of both worlds – neither do employees know how to improve performance, nor do they feel rewarded for their work.

    Finally, some performance reviews under the auspices of human resources departments focus only on getting reviews completed — “100 percent compliance” — not on their quality. A Sibson Consulting/WorldatWork survey found that 58% of HR executives give their performance management systems a “C” or below, in part because managers don’t receive the training they need to deliver effective appraisals.

    Points 1, 2 and 3 above might help managers move away from checking the box.

The article shares a case study about SAS which saw improvement when they moved from annual performance reviews to more frequent appraisals.  The article also talks about a new performance management tool (for software companies):

Indeed, the importance of frequent feedback crops up in almost every discussion of how to improve performance reviews. Daniel Debow is co-CEO of Rypple, a Toronto-based social software company that creates products designed to help people share continuous real-time feedback and provide coaching. Rypple’s target market is the 50- to 1,000-person knowledge worker firm focused on creative collaboration “where the model of a social network describes what is going on.”

Finally, here are some more ways to improve performance reviews.

Article first published as Should Performance Reviews Be Fired? on Technorati.

Pfizer says 24% cut in R&D is good for the company

Most high tech companies have ferociously guarded their R&D spending even through the great recession.  In fact, cuts in R&D spending have been much smaller than the reduction in revenues.  Hence, the R&D as a fraction of overall expenditure has actually increased through the recession.  It is known that R&D spending does not guarantee increased profits.  Many observers have pointed out that companies might actually be protecting the wrong investments.
Reuters had an interesting article recently about Pfizer cutting their R&D budgets by 24% in 2011 (See Pfizer R&D chief upbeat despite smaller budget): 

Mikael Dolsten, Pfizer’s president of worldwide research and development, said making choices about research priorities was a ‘sign of a healthy company culture.’ ‘Our action was more a thoughtful deliberation after looking at how the industry has performed as a whole,’ Dolsten said in an interview on the sidelines of the Bio-Windhover Pharmaceutical Strategic Outlook conference in New York. ‘We feel that the amount of investment in R&D that we are committing to is really the right number to drive the priorities we have put in place.’

May be the cut in R&D spending will actually force managers to think through the R&D pipeline and remove pet projects and dead wood.  This is especially important because Pharma R&D seems to have declined in efficiency and return on R&D investments have been falling (See Big pharma’s stalled R&D machine).

However, it is also quite common for CTOs to claim that cutting R&D budgets is a good thing AFTER the R&D funding has been cut.  TI’s CTO suggested that their 25% cut in R&D spending actually sharpened their focus.  What can R&D managers do to actually get the positive results?  Freescale’s CTO made some great points about cutting R&D budgets:
  1. Have a clearly defined strategy that drives investment decisions
  2. Decide on what R&D you are NOT going to do and what you are.
  3. Decide what R&D will be done internally and what will be outsourced to strategic partners
  4. Tie marketing into the R&D planning and align roadmaps with customers
Great advice because pet projects have a way to stick around no matter what.  Also, 90% of all cost cuts are reversed in three years unless there is a purpose and drive behind them.  Here are some portfolio management best practices that you could follow.  There is a lot more about portfolio balancing here.

Article first published as Pfizer Says 24% Cut in R&D is Good for the Company on Technorati.

Does modularity reduce innovation?

The Journal for Product Innovation Management had an interesting article on The Impact of Product Modularity on New Product Performance.  We recently discussed the benefits of modularity to combat component shortages (Impact of component shortages on R&D).  The article points out that modular design may have an impact on innovation.  Availability of a large number of alternate modules allows designers to try multiple alternate solutions and select the best alternate:

In light of problem solving, system complexity, and dominant design theories, some researchers suggest that modular product design promotes product innovation through experimenting with many alternative approaches simultaneously. This leads to rapid trial-and-error learning and accelerates new product introduction. 

The problem with modularity is that it requires compatibility and limits the solution space because modules need to fit together.  Also, if a module is fulfilling 80% of the requirements is available, designers may not push for the rest and hence not be as innovative.

However, others argue that modular product design inhibits innovation because common modules can be overly reused, the degree of freedom for innovation is limited due to module compatibility, and knowledge sharing among module teams is weakened.

The paper has results based on a survey of 115 electronics companies that suggest that the relationship between modularity and innovation is indirect.  The main recommendation is for R&D managers to be vigilant and monitor the negative impacts of modularity.  One red flag may be too many alternate configurations.  Another key concern is communication across different module R&D teams.

If there are any signs of diminishing product innovativeness, problems with poor communication across module teams, or excessive design alternatives, the manufacturers should stop further modularizing their products. Alternatively, manufacturers can take steps to reduce the negative effects of modularity. For instance, manufacturers can develop ways to strengthen communication among module teams. They can also use a set of design rules to reduce the number of design alternatives systematically or a design method to balance product commonality and differentiation during the development processes.

It is clear that as R&D becomes more modular, R&D teams for each module will become less engaged with other teams and more virtual.  We have discussed several approaches to boost productivity or drive satisfaction in virtual teams.  We could also try project networks to enhance communication.

Article first published as Does modularity reduce innovation? on Technorati.

Steve Jobs Methodology for Apple R&D

Apple innovations such as iPod and iPhone have had a wide ranging impact on the technology industry.  New York Times recently discussed How the iPhone Led to the Sale of T-Mobile.

Until Apple introduced its highly popular touchscreen device in 2007, which went on to become the world’s leading smartphone, Deutsche Telekom had been generating decent sales from its American operation, with growth in some years surpassing that achieved in Germany. But after the iPhone went on sale, sold exclusively at first by AT&T in the United States, T-Mobile USA began to lose its most lucrative customers, those on fixed monthly plans, who defected to its larger American rivals — AT&T and Verizon Wireless, which began selling the iPhone in February. The percentage of T-Mobile USA’s contract customers fell to 78.3 percent in 2010 from 85 percent in 2006, according to the company’s annual reports. During 2010 alone, T-Mobile USA said it lost 390,000 contract customers to rivals.

Nokia’s inability to compete with the iPhone led to its move to Windows Phone 7 and resulted in significant layoffs. (See DailyTech – Nokia Contemplates Deep Job Cuts Due to Windows Phone 7)

While Nokia’s growth has stalled, competitors like Apple and Android phone makers (Motorola, ZTE, HTC, etc.) have soared.  Now the company is forced to make a big transition as it prepares to move away from the Symbian operating system to Microsoft’s Windows Phone 7.

Apple was not the only reason for these upheavals.  Nokia was known to have become an inefficient bureaucratic organization which stifled innovation.  But the fact is that Apple’s innovative products and competitive positioning are material contributors to the market landscape.  Many observers have speculated that Steve Jobs has been the driver of Apple innovation.  But conventional wisdom is that that management involvement actually drives down innovation. In fact, it has been shown that risk averse management can prevent employees from innovating.  Research also shows that unless analysts categorize a company as a growth company, pressure from Wall Street also drives down innovation.

So what does Steve Jobs do differently and what can we learn from him about R&D management?  Steve Jobs clearly personifies some of the characteristics of innovators.  However, is there anything specific R&D managers can do to make their organizations more innovative?  I was thrilled to find a treasure trove of information on the Steve Jobs Methodology at the website Cult of Mac (In the transcript of an interview with ex-Apple CEO John Sculley On Steve Jobs). Here is what I think are key lessons:

1.       User experience centric design: Steve Jobs always started from the perspective of what the user experience was going to be.  However, he did not do that by doing market research or by asking consumers what they wanted to see in the product.  He rightly suggested that users did not know what the new products or technology could do.  It is the role of the R&D team to decide the user experience. (More discussion here)
2.       Long-term vision: Technology may not be mature enough to implement the entire user experience.  R&D managers need to be able to map out a development plan that achieves this vision in manageable steps.  Jobs started working towards a convergence device like iPhone in 2002.  The technology was just not ready at that stage.  He had an intermediate product in the ROKR in 2005 and finally reached iPhone in 2007. (More discussion here)
3.       Deep leadership engagement: It is not enough for a manager to have a focus on user experience and a long-term vision to get to it.  Just like other leaders of successful companies (like Bill Gates at Microsoft, Zukerberg at Facebook) Jobs is actively engaged in R&D.  From user experience to industrial design to retail store layout, he ensured a consistent theme through the entire operation.  (More discussion here)

4.       Small focused teams: Building strong focused teams is a challenge for any R&D managers.  Jobs had some interesting approaches to building strong teams.  A clear vision and deep involvement from management will motivate most engineers. Even more importantly, designers reported directly to Jobs and were respected more than all other skills at Apple.  He insisted on knowing all engineers by name and limited the team size to a number that he can know personally. (More discussion here)
5.       Razor-sharp focus on the niche: Apple had a sharp focus on its niche – ultimate user experience.  This is extremely important because the broad strategy always leads to conflicts between different goals.  The original MacBook Air had a wasteful industrial design to get it to market in a reasonable time, but it definitely delivered on the user experience.  (More discussion here)
The results continue to be stunning.  Here is what the Japanese consumer electronics engineers had to say about the iPad 2:

Through the teardown of the iPad 2, we noticed that Apple’s design philosophy is clearly different from that of Japanese makers. It seemed that the priority orders of various features and functions were determined based on the company’s aesthetic sense, and it designed the iPad 2 while giving first priority to realizing them. The iPad 2 made us wonder if products developed based on the philosophy of prioritizing costs and specifications can compete with it.

Article first published as Steve Jobs Methodology to Manage Apple R&D on Technorati.

Impact of component shortages on R&D

In Apple R&D and Steve Jobs Methodology: User Centric Design, we discussed how digital technologies let Apple focus on user experience.  The ability to focus on user experience, in turn, made Apple succeed where Japanese manufacturers failed – because Japanese companies focused primarily on components.

Apple’s success has had a big impact on the industry landscape.  There has been significant consolidation in component manufacturers.  More importantly, other companies have increased their focus on user centric design.  The result is that everyone is demanding the same set of components from a decreasing pool of suppliers.  The balance of power is now shifting again – from system designers to component manufacturers.

The article Getting Through the Shortages: No More Being Choosy in Nikkei Electronics has some very interesting data for R&D managers and strategy developers: “

The shortage in key components that began in summer 2009 is shaking the electronic equipment industry, and bringing about major change in the balance of power between equipment and component manufacturers. In response, equipment manufacturers are beginning to take action to ensure continued access to essential components at low cost.

Here is a great graphic from the article showing an strong increase in profits at the component manufacturers:

 So, what are the lesson for R&D managers:
1. Plan and design modular products: If one component becomes hard to obtain, you should be able to swap it out with another.  Modular products are always a great idea, but in case of supplier concerns, they become even more important.  The article had a great example of HTC Desire that shipped with an OLED screen, but had to be converted to LCD because of supply problems at Samsung.

2. Find commonalities between products:  If you can use same components across all your products, your volumes will increase and it will give you a greater clout with the suppliers. This is a challenge for R&D, because common parts will inhibit complete performance.optimization for each products.  Here is the graphic from Nikkei:

3. Secure supply by prepaying for parts: Self explanatory.  But still important for R&D managers because you will be locking in a particular component for a long term.  Designs around them will need to be robust enough to accommodate the parts from the long term supplier.
There is a lot more about R&D strategy and planning at the R&D Management Blog.

Article first published as Impact of Component Shortages on R&D on Technorati.

The psychology of change management

I have discussed the problem with process improvement and change management in the past.  We have seen that 90% of cost cuts obtained through organizational change are reversed within 3 years. It has been shown that senior executives are much more likely to imagine that change management projects are successful than middle managers.  I liked this 2003 McKinsey article that has a different take on the problem: The psychology of change management:

Companies can transform the attitudes and behavior of their employees by applying psychological breakthroughs that explain why people think and act as they do.

The article suggests that there are four conditions to make changes stick:

  1. A purpose to believe in: The leaders have to develop and describe a story of why the change is needed and why it is important. The story needs to be communicated to build a sense of purpose within the organization around the change. People are more likely to change their individual behaviors if they believe in the purpose (especially when change might cause some people to loose power / benefits).
  2. Reinforcement systems: We have talked about this previously.  Metrics and rewards needs to be aligned to make change stick.  However, psychological research shows that people get bored of rewards.  Rewards alone are not enough to change behavior over the long-term.  Hence, the other four conditions have to be satisfied.
  3. Skills required for change: Changes do not happen because many people do not know how to change.  We have seen research suggesting extended involvement of mentors (such as six sigma black belts) to make sure change happens and sticks.  This will only work if the organization launches a few, highly visible change programs, with a clear purpose. 
  4. Consistent role models: I have seen many change management projects fail because the senior managers did not actually change their behavior.  Their words were different from their actions.  People can easily see the difference between a real change and one in name only.  Again, we have seen research that shows that managers need to stay involved in the change for a long time to make results stick.
Overall a great article and worth reading….

Social networks helpful in improving R&D efficiency?

MIT Sloan Review article The “Unstructured Information” Most Businesses Miss Out On has some interesting benchmark information on the role of social networks in driving knowledge collaboration (and hence efficiency) in R&D environments.  The article details an interview with, K. Ananth Krishnan, the CTO of Tata Consultancy Services.  As you might remember, I have not been very impressed with the role of social networks in complex R&D.  TCS seems to have great success using social networks to share knowledge between geographically disconnected employees:

Well, let’s talk about the use of social webs inside the enterprise. Here at TCS, we are having a lot of success in saying that if you’re dealing with a particular problem and you need help, you go into our social platform and you just ask. You type in a question saying, “This is a problem I’m having. Has anybody solved this before?” And you might get five responses in 30 seconds from people who have done exactly what you tried to do, and they have their solutions. 

That is great.  Can we replicate this trend in any R&D environment?  What are some of the challenges involved in using social networks effectively in an R&D environment? Mr. Krishnan points out one:

Of course, three responses might say one thing and two might say something totally different. So you still have to use the intuition and the judgment.

So, one key problem with social networks is not knowing the veracity or accuracy of solution provided.  The other key problem is ability to describe the problem in sufficient detail that someone could suggest a meaningful solution.  Let us dig in:
Mr. Krishnan is in the IT business where a more or less common language / jargon is shared between all employees.  Most physical system require multiple disciplines (mechanical, electrical, etc) that all speak different jargons.  Furthermore, most problems occur at the intersection of different disciplines.  It is extremely hard to describe such multidisciplinary problems in sufficient detail in a social network.  Even more importantly, there are only a handful of people who could understand the problem and provide a realistic solution.  Social media would probably not be the most effective means to reach those people.  We should probably consider project networks?
That said, clearly, homogeneous environments such as software development or ASIC design can benefit from social networks.  TCS seems to be very enthusiastic about it:

We are today probably one of the largest users of the social web inside the enterprise, and we have improved our ability to look at the structured and the unstructured opportunity. In the last three years we have really launched into the exploitation of the social web as a means for ideation, as a means of finding the expert, as a means of learning. We use the web to form groups to look at specific problems and tapping into a collective intelligence.

TCS seems to have a key innovation in the use of social networks here: They use unstructured information from the social network to supplement structured information and to drive discussion.  I think that is a key requirement for the success of social networks in the R&D environment.

All those things supplement the way we look at our structured information, and they get some of these subjective insights into what we should be doing as a business.
For example, I have a blog inside the company, and I just finished writing a blog post which will go live tomorrow morning on the ideation process. There are a lot of things that I as the CTO of India’s largest software company should be looking at. Obviously, I don’t have the bandwidth to look at all of them. So I’m asking my readers to help me find out what am I missing. What are the three things they feel I should be paying attention to? Hopefully I will get a few hundred responses, and then I and my staff will go through and make sure that we pick the top three from there.
I do this quite often to supplement what I’m reading from all the other sources of information. The kind of insights that our business leaders might need for creating a new service offering or going after a new market or whatever, many of those get validated by this softer data.

So, I seem to be coming to the obvious conclusion that all tools can be helpful or harmful.  It depends on how one uses them.  Hence, if R&D managers want to use social networks, they have to get involved, show the team how to use the social network by example, structure the interactions on the social network AND give it adequate priority (probably by using it themselves).

Impact of Employee Stock Options on Performance

Knowledge@Wharton has a great article Incentive or Gift? How Perception of Employee Stock Options Affects Performance.  It appears that the idea that stock options drive people to work harder to increase the stock price is not quite true.

The story is not that people work harder to make the share price go up,’ Cappelli noted. ‘It is that if the share price goes up and people make money, they feel an obligation to work harder. That’s a bit of a surprise.

Employees seem to feel that the company has given them a gift ONLY if they can exercise the options at a profit.  Even then, they only work hard in that they feel some gratitude for the gift!  This is important for R&D managers to understand because options are increasingly a part of benefit packages.

The issue is significant because over the last two decades, American firms have both greatly increased use of a stock option plan as a form of employee compensation, and broadened the class of eligible employees to include more than just the most elite executives. According to the National Center for Employee Ownership, only one million U.S. employees held stock options in 1990. That figure has since skyrocketed to nine million workers now participating in roughly 30,000 different plans.

We have discussed incentives to drive R&D performance many times.  Most recently, we discussed that financial incentives alone might not be the best way to drive performance.  That post has a list of related articles that you might want to check out.  This article follows the same theme that we have discussed in the past. Lets dig in…
Incentives have to be aligned with required behavior to improve behavior. Most R&D team members can not define what they can do to improve share prices, so they do not actually see options as an incentive to work harder.

Boosting the research effort was a large amount of data provided by a major American public corporation. The firm granted stock options to the 4,500 employees — primarily store managers — based solely on their level within the company hierarchy, as opposed to job performance. Because these lower-level managers were largely responsible only for the sales performance in individual stores, there was little chance that their day-to-day work would actually directly influence share price, or that the manager would perceive such an impact.

So, what are some of the positives of employee stock options?  Clearly, employee retention is better – employees are not likely to leave if they have a lot of un-vested options.  The impact on performance seems to come from improved morale and is related to the actual profit employees make from the options.

… employees who profited handsomely from exercising their stock options appear to give a lot of the credit to the positive attributes of the company.

The article claims that there is significant improvement in performance.  It looks quite small to me and I wonder if a much better improvement can be gained from bonuses tied directly to required behavior…

The researchers found that a doubling of the profits from a sale of options resulted in a 1.3% increase in performance appraisal scores and a 1.1% decrease in performance-related dismissals — numbers that were statistically significant and meaningful in the broader context of the company data. According to Cappelli and Conyon, the data suggests that a company would have to increase the number of options awarded to employees by sevenfold to achieve the same impact on worker performance as a doubling of the profits from options.

The key problem is that profit from options are seen as gifts and any improvement in performance comes only once the profits are realized.

This shows that we’re not strictly economic in our relationships,” Cappelli says. “The proof of that is these folks who feel that they have been given a gift that they didn’t expect for their performance. There is nothing that requires them to reciprocate by performing better, but they do anyway.” Indeed, the research suggests that the improved performance after a profitable exercise of stock options typically lasts for a year or longer.

So instead of driving performance now, when we need it, we will drive performance when the share prices are actually higher…  Not quite a direct approach to improving performance and quite unpredictable in efficacy.

But the findings also raise questions for the many firms that offer stock options as a benefit, since the research shows that the impact on employee performance really depends on the price at which employees sell their shares, which changes in ways that are essentially unpredictable — and mostly out of the control of company leadership.

Finally, the most important problem: Employees need to know when to exercise the options to receive maximum profit.  The data shows that employees are not good at that (Personally, I have been very ineffective at timing).

One of the most critical variables was showing that the size of the stock profits realized by the managers was essentially the result of good luck in when they decided to sell, and not an indicator of either inside company knowledge or a special knack for timing the stock market. “Even top executives don’t do better than the average investor in making these selling decisions,” Cappelli states, noting that the majority of the employees in the study were store managers, not experienced stock traders. “Markets are pretty unpredictable, and there’s strong evidence that they don’t do well in timing the market.”

In summary, R&D managers should probably look elsewhere to motivate their teams – stock options are not it.

What Will Your Customers Buy Next?

A quick post about an article in Technology Review about What Will Your Customers Buy Next?

Using sophisticated math and vast amounts of data, predictive analytics software can help forecast and influence purchasing behavior. So why aren’t more companies using it?

The thrust of the article seems to be about predictive analysis software that can help companies better forecast evolving customer preferences:

With predictive analytics software, companies can see which customers are most likely to buy a given product. The process begins by ranking customers according to how recently they purchased, how frequently they buy, and how much money they spend.

Clearly, having data is not enough.  We need a way to generate knowledge and information out of the data.  The idea is to evolve parameters that can be used as a foundation for the predictive analysis.  The key question in my mind is how can we link this data about customer preferences into R&D plans?  I am not sure if anyone is working on that…