Wednesday, May 26, 2010

Patents Generated is a poor measure of innovation success

I see that Forbes has ranked Raleigh one of the most innovative cities in the US.  This was based on a couple of factors:

  • Number of technology and science jobs
  • Number of "creative" jobs
  • Number of patents per capita
  • Amount of venture capital invested per capita
I'll not spend my time debating these criteria, although I could easily argue that several of them are not measures of innovation but are outputs and outcomes of successful historical innovation.  In other words, some of these measures are really "rear view mirror" metrics.  They tell us about what happened in the past, but not what's likely to happen in the future.  To a certain extent we are benefiting from the decisions in the 60s and 70s to create Research Triangle Park and the recruitment of IBM and other research oriented firms.

After all, if we were to travel back in time in the heyday of Bausch&Lomb, Corning and Eastman Kodak, I'd be willing to bet that Rochester, New York, to use one example, would have ranked highly using these criteria.  I don't notice Rochester on these lists anymore.

But I digress.  My real beef with these statistics has to do with patents.  Every firm and university sets as one of its innovation goals the number of patents filed each year.  While patents are an interesting measure of ideas that have been documented and protected, and new intellectual property created, they aren't a good measure of innovation for a company, or more importantly a city or geography, for several reasons.  First let's establish a definition for innovation:  people or companies putting new ideas into action that has value for customers.

First, many patents are filed and awarded and absolutely nothing is done with that intellectual property.  This means that a smart bunch of people worked for a significant period of time to define and protect some insight, process or method that their own organization won't commercialize and won't license to someone else, usually because there's not enough commercial appeal.  In this case, while we are "racking up" the patents, we are actually encouraging economic behavior that detracts from adding real value to the corporation.  We'd be better off in the Keynesian model of paying people to dig and fill holes.  At least more people would be employed.  These patents don't add much value to the company or to customers, so it's hard to argue they are driving innovation.

Second, many patents are filed to defend turf or to stake out turf before a competitor can claim it. Thus many patents are defensive in nature, meant to ward off competition and protect a company's investments and assets.  This stifles competition and isn't good for the consumer.  It delays new product or service introductions or inhibits the introduction all together.  These patents aren't adding new value to customers or consumers, and similar to the patents described above can't really be considered innovative.

Third, it's misleading in the extreme to state that Raleigh has a lot of patents generated per capita, since we have three research universities within a twenty mile radius and one of the largest patent generation machines in the world, alternatively known as IBM.  I wonder what our ranking would look like if we subtracted just IBM's contribution to patents from our numbers.  While IBM's presence in Raleigh is a real blessing for jobs and for spin-offs, most if not all of the patents created there don't further Raleigh's innovation capabilities, and don't have direct impact on the local economy.

Fourth, let's consider the research universities where publish or perish is becoming patent or perish.  There is a tremendous amount of great research underway in our local research universities, much of it achieving patents.  Once again, however, those patents aren't getting translated as quickly and as efficiently as possible into new products or services, due to a host of technology transfer issues, revenue recognition issues and so forth.  While NC State, Duke and UNC pile up new patents, these universities and their associated new IP aren't having nearly the effect on Raleigh that Stanford has on Silicon Valley or MIT has on Boston.

If patents received is a poor measure for innovation success, what is a better measure?  How about patents converted into new products or services within X years of the issue of the patent?  Then we eliminate all the patents that weren't going to be commercialized, and all of those defensive patents at the start.  Next we'd also measure how adept the organizations and communities are at translating new IP to new viable products and services, which is really what innovation is all about.  Ask yourself, would you rather have one really compelling idea that you launched in the market that may be difficult to defend, or several patents that you have no intention of commercializing?  As Theodore Roosevelt said, credit belongs to the man who is actually in the arena...
AddThis Social Bookmark Button
posted by Jeffrey Phillips at 1:32 PM

4 Comments:

Anonymous Kate Shore said...

One advantage of the patent system, and one of the reasons it is used so widely, is that the information is publicly available. I'm sure that Forbes opted for metrics that were easily available and broadly comparable (e.g. not limited by a company's status, such public vs. private). The conversion metric that you proposed (patents used on commercial products) would not be publicly available. Most companies are not that sophisticated about their own patent portfolios, and even if they were, they would not publicize those metrics. So I guess I'd cut Forbes some slack for using patents as one metric in their study, but that's just my inclination.

I do agree with you that patents are not synonymous with innovation. But another way to look at a patent is that the company or individual felt, at the time of filing, that the invention was important enough to formally document and invest in the patent filing fees. That does demonstrate some value of the invention to the patentee at that point in time. You can actually measure patent abandonment, which is a partial measurement of the utilization of patents.

Thanks for the provocative post!

2:16 PM  
Anonymous Miguel Fernández said...

Just passing by and saying hello.

You might find this thread interesting. It's a discussion on how can we measure innovation and it's social impact.

http://www.google.com/buzz/timoreilly/TUZ2kg8FTip/I-posted-this-on-my-blog-at-http-radar-oreilly-com

6:25 PM  
Blogger Adit Creation said...

This comment has been removed by a blog administrator.

11:09 PM  
Blogger Unknown said...

Thanks a lot for this time sharing of innovation about PATENTS GENERATED IS A POOR MEASURE OF INNOVATION SUCCESS. This is really the best website about innovation i have ever read.
stop pre ejaculation

6:08 AM  

Post a Comment

<< Home