During the drafting process, claims in a patent application may be written, edited, revised, reviewed in-house, revised again, reviewed by a client, and revised yet again, or have some subset of these applied to it.  But, is the claim good enough?  Here are some tests we like to perform on claims.

Basic tests include grammar, spelling, punctuation, format, correct establishment of antecedent basis and reference, correct claim dependencies and other formal aspects of claims.  Beyond that, one should ask: does the claim capture the intended point of novelty as conveyed by the inventor(s) and as determined by the practitioner(s)?  Does the claim distinguish from known art?  Whether or not the client has directed to perform a search or provided references, does the claim distinguish from general, widely known examples in related fields?

Does the claim have appropriate scope?  Is the claim broad enough to ensnare an infringer who makes a clever variation of one or more parts of the invention?  Is the claim narrow enough, with sufficient detail in the claim elements, to not read on the general examples in the art?  This is a variation on some of the above questions, and brings focus on a recommendation.  The claim should be broad with respect to description of elements, or even inclusion of elements, that are not close to the point of novelty, and should be provided with sufficient detail or narrowness of breadth with respect to the point of novelty.  If there is more than one point of novelty in the claims, careful consideration should be given to whether to move one point of novelty into the dependent claims (or a continuation or parallel application with another claim set) and focus on one of the points of novelty, and only one, in the independent claims.

Are the claims clear as to meaning and scope?  Is it clear, upon reading a claim, what the invention is?  Can one understand the claim just by reading the claim, without having read the specification or looking at the drawings?  Hint: Examiners may do just that during patent prosecution.  Is there any ambiguity, which can be addressed by punctuation, moving claim terms, adding or deleting a phrase or other editing?  Once ambiguity is removed, do the claims establish a clear boundary as to what is claimed?  This is just another way of asking if the claims are clear as to scope, but by asking it another way, one may catch what is otherwise missed.

As to scope, does the claim protect what the client actually sells?  A claim that requires induced infringement or multiple parties to join together in infringement is not as strong or desirable as a claim that would catch a single infringer directly.  If a specific claim (e.g., a dependent claim) is directed to an embodiment that isn't precisely what the client sells, this may be useful as a defensive posture to prevent an infringer from selling a variation, but an entire claim set that doesn't encompass what the client actually sells may provide little protection for a direct copy by an infringer.

Does the claim have a good claim strategy behind it?  There are many claim strategies (some would argue there are as many claim strategies as there are patent applications or inventions).  Here are a few.

– Aim broadly, to capture as much scope as possible, knowing it is likely the claims will be amended in light of cited art during patent prosecution.

– Aim narrowly, to carve around close known art, knowing it is likely patent prosecution will face an uphill battle.

– A "picture claim" describes, more or less exactly, a specific embodiment.  These are usually very narrow, and may be employed when the point of novelty is not known or not readily discerned.  Hint: it is not usually a good idea to draft a claim without having a strong understanding or belief of what the point of novelty is, in the invention.  A picture claim may be a fallback position in lieu of a more targeted strategy.

– Aim for a claim that can survive patent prosecution with little or no amendment, in order to preserve the doctrine of equivalents.

– Aim for a claim that can survive both patent prosecution and attempts in the courts to invalidate the claim.  This is the so-called "bulletproof" claim.

– Write what you've got.  The client may insist that the invention has novelty, even though there is close or even overlapping art.  Do your best to represent the client's invention in the claims and advise the client as to the likely difficulties to be faced during prosecution.

Once the claims are drafted, and the specification is written and accompanied by appropriate drawings, the claims should be tested to see if each and every aspect of the claims is supported in the specification and in the drawings.  Read each claim, and each claim element, and ask if this is found in the specification.  Exact literal support in the written description is not necessary, as the claims themselves are part of the specification, but each part of the claims should be found in both the drawings and the specification.  This is a vital test, as claims can be rejected if the drawings and/or the specification do not support the claims.  Yes, the drawings and the specification can be amended during prosecution, but new matter is prohibited.

Each field, and even each specialty within the fields, may have its own jargon, some of which are terms of art.  Also, each field and specialty may have preferred styles and types of claims.  For example, software-based claims often encompass an algorithm in the claim, and the claim can be tested to see if the algorithm in the claim behaves as a computer or processor with appropriate programming would.  Claim language, especially claims to software-based inventions, may need to be debugged, much as a computer program is debugged.  A claim including an algorithm with recursive or self-referenced elements may present difficulties for antecedent references.  If the claim doesn't quite work correctly, spend the time to analyze why this is so, and revise it.  A method claim should be tested to see if following the method results in actions that are novel or results in an embodiment having the novel features of the invention.  An apparatus, system or other structure claim should be tested to see if it embodies the essence of the novelty of the structure of the invention.  A tangible, computer-readable media claim should be tested to see if it embodies what a processor (or computer or other machine) would do if it read the computer-readable media.

One good test of dependent claims is to see whether some of them act as a fallback in case an independent claim is invalidated.  This could happen during patent prosecution, in which case moving some of the limitations of a dependent claim into an independent claim, by amendment, may be a good strategy.  Or, bringing in limitations from part of the specification into the independent claim, and keeping the dependent claim as a fallback may work well.  An independent claim could also be invalidated in a court case, and a fallback dependent claim might survive.  Writing the claims so as to support this possibility is a good strategy.

Do the dependent claims encompass all of the variations that the client discloses, and all of the variations that a person of skill in the art would devise upon reading the specification?  If there are any gaps, it may be worth revising the dependent claims for additional breadth, or adding dependent claims or recommending the client file a continuation application with another claim set.

Another test, based on the "less is more" concept of claims scope, is to ask whether the claim could be written in fewer words.  Fewer words in a claim generally (but not always) broadens the scope of the claim.  An overbroad claim is usually not desirable, unless the strategy is to set forth a very broad claim and amend it during prosecution.  But, a claim might be unintentionally narrow.  By looking at each claim term, asking if this claim term is necessary in the claim, and deleting unnecessary words or moving some words to dependent claims, one can pare down the claim to an appropriately broad scope.

One last test (for the present article, although this test could be applied at any time during the drafting process) is to ask whether the claims, particularly the independent claims, match both the preambles to the claims and the title for the patent application.  If so, good, if not, it may be worth revising one or more of these.  And, after revising for any or all of the above, read the claims one more time, to catch anything you might have missed.  This is all part of the art of patenting.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.