Sunday, March 28, 2010

Patent Basics III - Prior Art

(This is the third post in my series on Patent Basics; the first two posts can be found here and here. My previous series on Trademark Basics has been consolidated on my website at Future "basics" series will cover copyrights and trade secrets.)

What is “Prior Art,” and Why Should I Look for It?

Patents are only awarded for inventions that are “novel” -- in other words, new. You won't be able to get a patent for the light bulb unless your version of the light bulb is an improvement over all of the other light bulbs that have ever been invented and disclosed in some way to the public. Those earlier versions of the light bulb are part of the “prior art.”

Inventors have an obligation to tell the Patent Office about any prior art that they're aware of that may relate to their inventions. But that of course is not the main reason for conducting a search to find possible prior art before you file your patent application. There are a number of more practical reasons for doing so.

Prior Art Search Benefits
  • A prior art search will help you figure out whether your invention is truly 'novel.' Why spend thousands of dollars on a patent application for an invention that someone else has already thought of?
  • A prior art search may help you refine and improve your invention. “Stand on the shoulders of giants,” so to speak, and see what others have done. Their work may inspire you to improve on your original idea.
  • A prior art search will show you what you may need to 'design around.' Since you can't patent what's already been invented, you may have to develop a different way of doing what your invention does in order to avoid infringing an existing patent. A workable design-around can be valuable if the owner of the prior art patent refuses to license their invention on reasonable terms. A design-around can then become a cost-effective way to compete with a product or process that is protected by such a prior art patent.
  • A prior art search will help you understand whether your invention is a significant improvement over existing inventions, or an incremental advance. In either case, that information will help you decide whether it is worth investing in the patent application for your idea. An incremental advance that overcomes a significant marketing or operational problem that exists with the prior art, for example, may be very valuable. On the other hand, if your idea does not add much value to the existing invention, you may decide to invest your money elsewhere.

How to Conduct the Prior Art Search

There are services that will conduct a prior art search for you, of course for a fee. The quality of these searches can vary widely, and often depend on how much you spend. One thing that I explain to every client is that nobody will conduct as thorough a prior art search as someone who is accused of patent infringement. In other words, a prior art search that is paid for by an inventor, and may cost hundreds of dollars, will not be as thorough as one conducted by an accused infringer that stands to lose millions of dollars in a patent infringement suit.

That's a fact that patent owners simply need to accept. A reasonably-priced, reasonably-thorough prior art search is likely to provide a good indication of existing, similar inventions that are out there. But it cannot be considered to be definitive.

The patent search services tend to be very adept at searching U.S. Patent Office records, which in many cases are an excellent source of prior art. Existing patents and published patent applications are likely to form the bulk of the results of most prior art searches from such a firm. If your invention is one that is in a 'traditional' line of inventions – in other words, not a business method – then a Patent Office-focused search may be sufficient.

If, however, the invention is in a new field of science or technology, or is a business method invention, you should make sure that the prior art search goes beyond the Patent Office, and includes areas where publicly-available information about the particular field of endeavor is likely to be found. This includes the Internet, but can also include key university libraries as well as foreign-based information sources.

In some cases – mainly in areas where the inventor is at the cutting edge of the technology – the inventor may be an excellent source of prior art. In any case, it's always a good idea to check with the inventor to find out what information he or she has about the existing state of the technology and other inventions that may be out there.

Where to Search

If you want to do some or all of your prior art searching yourself, here are some good places to start. Of course Google or some other comprehensive search engine is first, provided you are adept at crafting searches that will yield enough relevant 'hits.' Google also has its own patent search engine, available at, that is a great help insofar as it goes. Clicking on the “Advanced Search” option opens up a detailed search menu that you can use to help narrow down your search results. I have found, however, that Google Patents is not kept up to date, and many recently-issued patents and recently-published applications are not included in the scope of its search, making this otherwise excellent search tool something that is ultimately unreliable for prior art searching.

There is, of course, the United States Patent Office website at The main page will give you a link to the Search section, where you will quickly learn that the PTO is not exactly on the forefront of intelligent search technology. Still the information is there, and if you know something about a particular inventor or company whose work may be part of the prior art, a PTO search may be an efficient way to see what they've patented. If you really want to dive in to the nitty-gritty of how to search using the PTO's patent classification system and other similar tools, you can look at what the PTO tells its examiners about searching prior art here: .

Finally, don't forget that a foreign patent or invention disclosure can also be prior art. There are a number of databases where you can conduct a search of foreign patents. The World Intellectual Property Organization provides a short guide to some of them here: .

Some sort of a prior art search is an important part of the application process for any patent application. Whether your resources permit you to hire a search firm, or you conduct the search yourself, if you move forward with a patent application a prior art search will be a good investment.

The next post in this series will cover the difference between provisional and non-provisional applications, and the basic structure of the patent application.


  1. Excellent post!!...

    Most of the people use prior art search just to check the patentability of their invention. However, I believe prior art is much more than that.

    Prior art can be used to find out additional things that can be integrated with your invention. Prior art can be used to expand your invention into other application areas which you might not have thought about. And a lot more...

    Nice article... Will surely be subscribing to your blog from now on.



  2. Great article and good introduction for those looking to dip their toes in on their own prior art searching. Newbies to the prior art search game might also be willing to try their hand at a little classification searching after they have exhausted the Google Patents avenue. A colleague of mine has just written a couple of blog posts on classification searching if anyone is looking for a primer:

    My company also hosts a free web resource, called Intellogist, for anyone wishing to learn more about prior art searching, and there are Best Practices articles available by technology area - these include initial classification suggestions by technology area, and other tips, and they are available at

    By the way, the point you make about professional search services being experts at patent searching, but less focused on literature, is definitely something that professional prior art searchers need to work to improve in general. I work for the search service Landon IP, and I know we have placed emphasis on literature search offerings to attempt to mitigate this patent-heavy focus in the field. For example, we offer a Scour the Earth search service that is designed to be extremely comprehensive, encompassing the Library of Congress and other university library sources that offer public access. An explanation of the sources searched is available at the following URL:

    Of course, as you have put it, this kind of in-depth search is going to be more costly than a simple patentability search, just due to the amount of effort involved. However, patentability searches will still encompass a basic literature search.

    The point you make that a reasonably thorough prior art search is still never going to find *everything* is a valid one - it's impossible to "prove a negative" given the vast amount of patent and other literature that has already been published in the world. Patent owners need to be aware of this reality, and prior art searchers need to be extensively trained to search the best sources and to mitigate this risk as much as possible.

    Thanks for the great review, I think this is one of the best blog posts introducing prior art searching that I have read.

  3. Thanks, Deepak - good point there. And Kristin, your well-written and informative comment is appreciated, and you inserted enough flattery to get me to overlook the advertising.

    Seriously, your point about the depth to which a prior art search can be taken is a very good one. It's always going to be a judgment call, and you'll be balancing the money you have to spend, the possible value of your patent, and the time it will take to conduct and analyze the prior art search when you decide what kind of search to conduct.

  4. Haha, thanks Kelley, thanks for overlooking the link list there. As someone who has done prior art searching, you want to work with the client's budget and expectations, and you just have to hope that the two line up! For those very short, cut-it-off-after-3-hours type projects where the customer can't spend for a full patentability search, it's very difficult to feel good about stopping when you just know there is so much more out there. But I think that if you have the option to hire someone who has been trained and has a lot of search experience, you are going to get a lot more bang out of those first hours.

    As a current library school student and someone who is witnessing the decline of the reference librarian and the rise of the end-user search on Google, I am all for empowering users to do their own searching with the caveat that I think they should know what they are getting into before they jump into the deep end!

  5. Seriously, your point about the depth to which a prior art search can be taken is a very good one. It's always going to be a judgment call, and you'll be balancing the money you have to spend, the possible value of your patent, and the time it will take to conduct and analyze the prior art search when you decide what kind of search to conduct.