Impact of IT on R&D management

MIT Sloan Review has an article on “The Four Ways IT Is Revolutionizing Innovation.”  The factors according to the article are:

Measurement, experimentation, sharing and replication.

Of these factors, sharing and replication are clearly universal drivers enabled by IT.  Most industries are benefiting from improved sharing and replication of information across R&D organizations.  The impact of other two is not universal.

The examples given in the article are mainly Internet and networking companies.  Clearly, they have the luxury of using IT for measurement and experimentation.  A lot of their data is easily available and IT can really support experimentation. 

In many other industries (high-tech, electronics, bio-tech, aerospace and automotive, etc.), IT does not actually measure R&D results or aid experimentation.  In fact, a lot of the challenges in managing R&D stem from the difficulty in measuring R&D results and finding ways to experiment with R&D processes/tools to improve efficiency.  Any thoughts?

Making cost cuts stick

According to one recent study, 90% of the organizations fail to sustain cost savings beyond two years.  McKinsey quarterly has an interesting check-list to ensure cost cuts stick:

  • Assign accountability at the right level
  • Focus on how to cut, not just how much
  • Don’t let P&L accounting data get in the way of cost reduction
  • Clearly articulate the link between cost management and strategy
  • Treat cost management as an ongoing exercise
  • All very interesting points.  I think the first two are quite relevant in an R&D environment.  However, several are very difficult to implement in R&D.  I remember an old boss talking to me about his problems predicting R&D costs.  He remarked that when he went to get his car serviced, they had a pretty good handle on what different services would cost and how long they would take.  Why could we not do the same for R&D?

    Predictability normally comes with repeatability.  Unfortunately, finding repeatability in R&D is pretty tough – as R&D always involves developing something new.  In fact, it is more developing knowledge that can be used to produce that new item.  Without repeatability and associated cost data, it becomes very difficult to understand what things should cost to develop in the first place.  How do we define how much cost to save?  As we get to more and more complex devices, this will likely become even more difficult…