It is a very daunting task to review the claims in your own application which were written by a professional - presumably, you are confident that he knows his job well enough to avoid obvious mistakes. If you're not so confident, you might want to see our page "pitfalls in drafting your own claims" for a list of the top ten mistakes in claim drafting - and make sure your professional hasn't made any of them.
Assuming that your attorney or agent hasn't made one of the basic errors in claim drafting, then everything should be fine, right? Wrong. You still need to review the claims very carefully. He or she might know patent law, but no one knows your invention as well as you do. Start with the independent claims, and ask yourself these questions:
1. Does the claim really claim what I thought I invented?
This is pretty basic - is the claim really focused on what you think your invention is? If you think you invented a new hammer, does the claim start "An improved nail"?
It is common, in fact desirable, to have a number of independent claims describing the invention in varying breadth. For example, an improved fuel injection system for an automobile engine might be claimed as a fuel injection system, an internal combustion engine having the fuel injection system, a motor vehicle having an engine with the fuel injection system, a fuel injection system made according to a specific method, and a method of injecting fuel into an internal combustion engine.
The inclusion of claims based on the larger assembly (the engine or vehicle, in this example) is done in an attempt to increase the possible royalty base for an infringing device - you'd rather get 2% of the sales of an entire car than 2% of the sales of a fuel injector. However, this technique should not be used to excess, and make sure that the first claim is focused on exactly what the invention is - the fuel injector, not the car.
If there are "product" claims and "product by process" (that is, claims to the product made according to a specific method), make sure that the process really would produce the product.
If you invented an improved process of making an existing product, then you should not have any claims to the product itself. A claim to the method of making the product would be proper. A product-by-process claim would not be proper, unless there was some new feature in the product which is a result of the process (in other words, both the product and the process must be new for a product-by-process claim).
You cannot patent a known device, even if you are using it in a new way. A new use for an old device can only be claimed as a method. If you know that the basic device is old, and your invention is in the way it is used, be sure that your application only claims the use (as a method) and not the old device.
2. Is every element recited in this claim really necessary?
In many cases, the first draft of a claim may include elements which appeared in the description of the preferred embodiment, but it is not really necessary to recite these elements in at least the broadest independent claim. For example, if an element is simply part of the environment (say, a housing or frame), or is so much a part of the prior art that any person having skill in the art would realize that it is part of the device (for example, everyone knows that an internal combustion engine requires a fuel source), then it does not belong in the independent claim, unless it is needed for antecedent basis later on.
If an element is necessary for antecedent basis, but is not really part of the invention, as such, it can be moved to someplace where it is not positively claimed. Sometimes, if they are important enough to mention, but not absolutely essential, such elements belong in dependent claims. Or, if they are essential to give the invention structure, but are not part of the novelty, they can be put in the claim preamble (before the "comprising" connecting word). Maybe the claim would be better written in Jepson form?
In method claims, are there any steps which are merely there to prepare for the actual inventive method?
For example, if the method starts with a step of cleaning a workpiece, is that really needed? What if you started with a workpiece which was clean enough, or if the method would work (perhaps not quite so well) if the workpiece were not clean? Move the step to a dependent claim.
You should question the inclusion of any preliminary steps which just recite "providing (an element)", which is used or processed in a later step in the method. These steps can be omitted, since it will be assumed that the element has been "provided" when you start to work on it, and they are really meaningless. If you include such steps (and they are very common), you will give a potential infringer the opportunity to avoid infringement by saying that he did not "provide" the element, whatever that means.
3. Is there any element of my invention which is essential, but which is missing from the claim?
This does not mean that there is some part of the device as a whole which might be needed to make the device work - focus on the novel invention itself. Is there a part of the invention which is necessary to the operation which does not appear in the claim?
This is especially easy to miss when your invention is an improvement to part of an existing product. As an example, every car needs a motor, tires and a fuel source, but these elements do not need to be included in an independent claim if the invention is an improved turn signal. On the other hand, you would probably want to include an electrical power source, lights, a turn signal switch and a way of making the light flash.
4. Are any of the words in the claim too limiting? Is there a more generic word which could be used? Must the word be used at all?
Watch out for words which are unnecessarily limiting - if two parts need to be connected by a threaded connector, the claim should use that term and not "hex bolt". Claims shouldn't say things are "epoxied together" if they just need to be "adhesively fastened". In fact, if all that matters is that the parts are connected, the claim should just say so ("an A connected to a B") and not positively claim the connecting structure at all.
Your patent attorney may not be aware that a technical term is limiting - look at every word and see if you know a more generic term. Do you always use the term "polyethylene", when "plastic" or "polymer" will do? Or must the element really be polyethylene and not polyvinyl chloride?
Did your attorney use any word in a claim which is a trademark? For example, rather than claim "Velcro", which is a trademark, the claim should say "hook and loop fastener material" (at most - "separable fastener" would be better, unless it had to be hook and loop type).
5. Are there any alternatives to the elements in the claim which are not covered by the wording used?
Perhaps your preferred embodiment was written in a specific way because of manufacturing concerns - maybe you decided to drill a hole in a certain way because it was easier with your equipment, or you used a specific substance because it was less expensive. If there are alternate ways of accomplishing the same result, which are equally novel, make sure that the broadest claim, at least, covers those alternatives.
Review every adjective - must a body be "cylindrical", must a substance be "rigid", must an attachment be "pivoting" when "flexibly attached" would do?
6. Must the parts of the device be arranged in the specific way described?
Question "over", "under", "beside", "inside", "outside" if merely "adjacent" would do. Must a part really be "on the front end", or just "on an end"?
The same is true for method claims - must the steps in the method be performed in exactly the order given? Question any use of "first", "next", or "followed by..." or other wording implying an order, when no order is actually required.
7. Are the dependent claims correct?
Look at the dependent claims, and ask the same questions - but remember that dependent claims must be read in combination with all of the claims upon which they depend, and they are intended to be narrower than all of the claims upon which they depend. See the basic "How to read claims" page, if you're not clear on how to read dependent claims.
Is each dependent claim dependent on the correct "parent"? It is a common mistake to lose track of what claim depends on which, when claims have been repeatedly added and deleted in drafting. If a dependent claim does not further define or restrict a term in the parent, then perhaps it should be dependent upon a claim further up the chain.
Are the terms used in the dependent claims really narrower than the equivalent terms in the parent claims? For example, if you said an element was "polyethylene" in a parent claim, you could not have a dependent claim which says that the element is "a polymer" - that term is broader than "polyethylene".
Is the dependent claim really necessary? If all it adds is an element which is purely conventional, maybe it can be omitted entirely.
8. Are the claims properly supported by rest of the application?
Make sure that all of the elements of the claims are adequately described in the specification and shown in the drawing. More than that - make sure that every word used in the claim is defined, or at least described, somewhere in the specification.
Make sure terminology is consistent. If a screw is "a screw" in the detailed description, it should not be a "threaded fastener" in the claims (better yet, have the specification changed to "threaded fastener").
If you are using a word in a way which is not entirely standard, be absolutely sure that your specification has a clear definition of the term ("As used herein, 'cylindrical' means any structure which has a cross-section which is rounded, including circles, ovals and rectangles with rounded corners.").
9.For method claims - who is going to perform the method?
If you are reviewing a method claim (that is, a claim which starts with "A method for (x) comprising the steps of: ..."), take a look at each step in the method and ask yourself, "when this method is being performed, who will do this step?"
The reason for this review is that the whole purpose of writing claims is allow the patent owner to sue others to stop them from making, using or selling the invention defined by the claims. This act of making, using or selling without permission is called "infringement". If you can't define who is an infringer, or the claim is written so that it would be infringed by the wrong person, you might wind up with a valid claim you can't enforce.
As a first step in looking at a method claim, ask yourself, "who would I want to sue for infringing this patent?" Then, look at the claim and ask, "would the person I want to sue actually be the one peforming the steps in this claim?"
For example, suppose you invented a new way to allocate seats in an arena for basketball games. The claim you are reviewing reads:
1. A method of selecting seats for basketball games comprising the steps of:
(a) looking at a seating chart for the arena where seats are color-coded according to sight lines, price, comfort and availability;
(b) determining how many seats you need;
(c) finding all of the areas in seating chart where there are at least the needed number of seats which are contiguous and coded as available;
(d) choosing contiguous seats which are coded as being in your price range; and
(e) buying the tickets for those seats.
This might be a perfectly good and patentable claim, but who would infringe it? The sports fans buying tickets. Would you really want to (or be able to) sue each of the thousands of fans buying tickets for the next Celtics game? Probably not, if for no other reason than that the damages which you could recover from any one fan would be tiny, at best.
It would be better if the claim read,
1. A method of allowing a user to select seats for basketball games comprising the steps of:
(a) a computer server presenting a seating chart to the user, wherein seats are color-coded according to sight lines, price, comfort and availability;
(b) the computer server accepting from the user a count of how many seats the user needs;
(c) the computer server displaying on a seating chart where at least the needed number of seats are contiguous and coded as available;
(d) the computer server accepting from the user a selection of contiguous seats; and
(e) the computer server issuing the tickets to the user for the selected contiguous seats.
Now, the claim would be infringed by the arena (or Ticketmaster, or whoever runs the computer server).
There's another glitch, too - under the "all elements rule", it is only infringement if all elements of at least one claim are present in the accused product or method. For an apparatus claim, this is more-or-less straightforward - does the product have all of the parts which are listed in the claim, or not? For method claims, though, it is a bit more complicated. In order for a method claim to be directly infringed, not only do all of the steps have to be performed, but, according to the courts, but all of the steps have to be performed by one actor (preferably the party you are suing for infringement).
If even one step is done by someone else, then there is no direct infringement. Using the above example, suppose the claim read:
1. A method of allowing a user to select seats for basketball games comprising the steps of:
(a) a computer server presenting a seating chart to the user, wherein seats are color-coded according to sight lines, price, comfort and availability;
(b) the user entering a count of how many seats the user needs into the computer server;
(c) the computer server displaying on a seating chart where at least the needed number of seats are contiguous and coded as available;
(d) the user selecting the contiguous seats he wants to buy; and
(e) the computer server issuing the tickets to the user for the selected contiguous seats.
Now, steps (a), (c) and (e) are done by the server (i.e. the arena or its ticket-issuing company), but steps (b) and (d) are done by the user. This claim cannot be infringed because no one actor performs all of the steps. Note that there have been some cases which still found infringement for multi-actor method claims, but these have usually been in unusual cases where one of the actors was controlled by the other (perhaps a central control computer and a remote client computer executing instructions sent by the central computer) or was effectively the same party (say, two different computers but both owned and operated by the same company, who was the party being sued).
As you can see from these examples, it is possible to write essentially the same method in different ways to capture different actors. So, you might consider having a number of different method claims, each looking at the method from a different point of view - a server-side claim, a client-side claim, maybe even a claim from the point of view of the user of the client. Make sure your attorney understands these distinctions and ask him or her how this can be done - don't assume that the attorney knows the method as well as you do, so that he or she will necessarily see that more than one actor (or the wrong actor) is involved.
10. Is anything missing?
Do a final double-check for consistency and completeness.
Do it now! Before you file is the time to add things to the patent application - once it is filed, you cannot add "new matter".
<- Back to How to Read Claims
Back to "How to read a patent" main page