What Separates an Enforceable Patent From a Decorative One

There is a category of patent that exists mostly to be framed. It has a ribbon, a number, and claims so narrow that no competitor would ever accidentally fall inside them — and no competitor would deliberately, either, because avoiding them costs a design change of about twenty minutes. The owner paid full price for it, lists it on the website, and mentions it in every investor deck. It has never stopped anyone from doing anything.

The difference between that patent and one worth enforcing is not luck and it is not the examiner. It is a handful of decisions made during drafting and prosecution, most of which are invisible to the inventor at the time. Understanding them will not turn you into a patent attorney, but it will let you ask the right questions of one, and it will stop you from treating a grant notice as proof that you own your market.

1. Claim breadth against validity — the central tradeoff

The claims are the patent. Everything else in the document exists to support them, a point worth internalizing from how patent claims actually work. A claim is a set of elements, and a competitor infringes only by practicing every single element. Each additional word you add narrows the claim; each word you remove broadens it.

So why not draft the broadest claim imaginable? Because the same words that catch competitors also catch prior art. A claim broad enough to read on every competitor is usually broad enough to read on something published before you filed, and that makes it invalid. A patent is strong when its independent claims sit at the widest point that the prior art still allows — and the only way to know where that point is, is to have searched properly first. This is why a serious prior art search before drafting is not an optional expense; it is what tells the drafter how much room there is.

Watch for the tell-tale sign of an over-narrowed claim: strings of specific dimensions, materials, or numbers that were added purely to get past a rejection. A claim reciting "a housing of anodized aluminum having a wall thickness between 1.2 and 1.8 mm" has effectively told the world that 2 mm of glass-filled nylon is free.

2. Enablement and written description

Section 112 of the patent statute requires the specification to describe the invention well enough that a person skilled in the field could build it without undue experimentation, and to show that you actually possessed the full scope of what you claim. Both requirements are validity landmines for thin applications.

The classic failure is the specification written from a single prototype. It describes one motor, one sensor, one geometry, then the claims reach for a whole category. If the disclosure never explains how the invention works with anything other than that one embodiment, a court can find the broad claims unsupported. The fix is unglamorous and cheap at drafting time: describe the alternates. Every substitute actuator, every alternative sensing principle, every material family, every range you have reason to believe works. Adding three paragraphs to a specification costs almost nothing. Adding them after filing is impossible — that is new matter, and the rules on amending an application after filing will not let you.

3. A prosecution history that did not concede too much

Everything you say to the examiner becomes public and permanent. When you narrow a claim to overcome a rejection, or argue that your invention differs from a reference because of some specific feature, you generally cannot later assert that the patent covers what you gave up. That is prosecution history estoppel, and it is the reason a granted patent's real scope is often much smaller than its claim language suggests.

Weak prosecution looks like this: a first rejection arrives, the applicant amends every independent claim to add the first distinguishing feature they can find, writes a paragraph explaining why that feature is essential, and gets an allowance in one round. Fast, cheap, and the resulting patent is decorative. Strong prosecution argues the prior art first, amends only when argument fails, amends by the smallest increment that works, and says as little on the record as the rejection requires. It costs more rounds and more money, and it is the single largest determinant of whether the issued claims are worth anything. The mechanics are covered in our guide to responding to an office action.

4. A family kept alive

The best patents are rarely the ones filed first; they are the ones filed later in the same family, once the market has shown you what competitors actually build. That is only possible if something in the family is still pending. File a continuation before the parent issues and you retain the ability to write new claims, supported by the original specification and carrying the original priority date, aimed at a product that did not exist when you drafted.

Let the last application in a family issue with no continuation on file and that door closes for good. It is a modest annual cost for a large strategic option, and it is the reason experienced patent owners almost never let a family go completely to grant while the product line is still evolving.

5. Infringement you can actually detect

An unenforceable-in-practice patent is one where you cannot tell from the outside whether anyone is infringing. If your claim recites an internal control algorithm, a firmware state machine, or a step performed on a server you will never see, you may be right that a competitor is using it and still have no way to prove it without discovery in a lawsuit you cannot afford to start on a hunch.

Strong claims are written toward what is visible in a purchased unit, a published datasheet, a teardown, a user manual, or a marketing claim. Where the valuable idea genuinely is invisible, the honest answer is often that it should not be a patent at all — that is exactly the situation trade secret protection exists for. Publishing an undetectable method in exchange for an unenforceable right is the worst trade in intellectual property.

Quick self-assessment

Apply the same test in reverse when a competitor's patent is pointed at you. Most patents that look terrifying in a demand letter fail two or three of these rows, which is why designing around a patent works so often, and why challenging validity is a live option rather than a desperate one.

Projects House is a product development engineering firm, not a law firm. What we contribute to a strong patent is the engineering underneath it: identifying which alternate embodiments genuinely work so they can be disclosed, producing the drawings and test data that support breadth, and reviewing whether a claim would be detectable in a shipped product. Drafting and prosecution belong with a registered patent attorney. To bring engineering rigor to a filing you are planning, reach us through the contact form.