In response to outcry from many companies defending infringement suits that revolve around patents many agree should never have been issued, the US Patent and Trademark Office has endeavored to implement programs to ensure the granting of higher quality patents.
One such program is "The Second Set Of Eyes", a process that requires a second examination professional to review the work of the primary patent examiner.
There is also pending legislation intended to raise patent quality standards, including a provision Post Grant Opposition. This process would allow third parties to review newly granted patents, and provide relevant prior art which, had the patent examiner been privy to, may have caused the patent examiner to disallow the patent. The notion is that if patent applicant knows that their patent will be subject to public scrutiny, they will be more diligent in crafting patents that claim truly novel inventions.
And IBM recently announced a new corporate patent policy that included as two primary tenants:
Ultimately, "patent quality" correlates to "patent value" – either as determined by the licensing revenue that patent may earn during its useful life, or perhaps the ability of a patent to survive opposition or invalidity challenges in order to preserve or enhance a company's monopolistic market position.
With all this effort to ensure patent quality, the definition of "patent quality" still remains elusive. But one such measure is a qualitative analysis of patent quality using statistical processes.
PatentCafe, a provider of an advanced technology international patent search engine, and patent analysis software tools, took a look at the patents central to the IBM / Amazon litigation to see whether there are obvious quality issues.
The Patent Factor Index Report is a statistical analysis tool that assesses 20 different quality indices, looking separately at legal, commercial and technology factors. Using its Patent Factor Index Reports, PatentCafe analyzed the five IBM patents being asserted against Amazon.com (click on any patent to download the full Patent Factor Index Report):
"The "quality" of a patent has many dimensions, and the perspective of the quality of a patent changes depending on who's looking at it," says Andy Gibbs, PatentCafe Chief Executive Officer. "If we compare patent analysis to car-shopping, we’d see that a small, gas-miser car would earn a "high quality" score by a cost-conscience commuter, but a Ferrari would be considered "high quality" to someone who wanted speed at any cost. PatentCafe's PFI Reports take subjectivity out of patent analysis, and provide an objective score of the various qualitative elements."
PatentCafe's PFI Reports statistically analyze 20 different quality factors, and provide a weighting of each factor compared to the 100 most closely related patents. The results are not necessarily absolute, but rather highlight potential strengths and weaknesses in the legal, commercial and technology areas.
The following scoring charts show how the scores of the factors vary, contributing to the overall quality of each of IBM's patents.
Will statistical patent quality analyses replace humans?
Gibbs says "In a world where more information is required faster, statistical patent analysis helps cut down weeks of traditional front-end patent review work to just a few minutes – something that we'd expect a tool to do."
Humans will never be eliminated from the patent analysis formula. PFI Reports are a fast method to identify the elements of a patent that may be extraordinarily high – or low. Then it's time for licensing professionals or litigators to start dissecting the claims, the patent specification, and file wrappers.
But, from that point, it comes down to how skillful the litigators are in clearly presenting the patent quality information to a jury.