Can My Idea Be Patented? The Two Questions I Ask Every Client

What do clients really ask when they bring me an idea?

In my 20 years of practice, very few people walk in asking the abstract question "is this idea patentable?" The most common situation is more practical: the client already has an R&D result or a product, and does not know whether it can be patented at all, or how to file so that the product is protected as fully as possible.

Those are two different questions. The first is about whether there is anything patentable. The second is about how to claim it so that a competitor cannot simply step around the patent. This article deals with the first one, because if the answer there is no, the second one does not arise.

Why can't you patent an idea on its own?

Every major patent system draws the same line, just in different words. In the United States, 35 U.S.C. §101 covers processes, machines, manufactures and compositions of matter, and the courts exclude abstract ideas. The European Patent Convention, in Article 52, excludes discoveries, mathematical methods and schemes for doing business "as such". China's Patent Law, Article 2, defines an invention as a new technical solution for a product or a process.

On top of that, every system requires the application to describe the invention well enough for a skilled person to carry it out (35 U.S.C. §112(a) in the US, Article 83 EPC in Europe). An idea tells people what you want. A patent has to tell them how it is done.

How do I judge whether your product can be patented?

My working method has two steps.

First, is there a real difference from peers? I put your product next to what competitors sell and what is already published, and look for the concrete difference: a structure, a step, a material, a control sequence. "Ours is better" is not a difference. "Ours places the sensor here and switches at this condition" is.

Second, would that difference be easy to think of? Here I deliberately take the view of an outsider simulating an intellectual property judge. I am not an engineer in your field, and neither is the judge who may one day decide whether your patent is valid. If, after reading the prior art, I feel that anyone would arrive at your solution, the case is weak. If I cannot see how someone would get from the existing products to yours, there is something worth arguing for.

To be clear, the legal yardstick is formally the "person skilled in the art": the PHOSITA in US non-obviousness analysis, the skilled person in the EPO's problem-solution approach, and the 所属技术领域的技术人员 in Chinese examination. My outsider-judge view is a practical stress test, not a replacement for that standard. It works because, in the end, the people who decide are lawyers and judges reading documents, not your engineers.

My two-question test: is your product concretely different from what peers make, and would that difference be easy to think of for an outsider reading the prior art the way a judge would? Fail either one and the case is weak.

When is a good idea not yet ready to patent?

A real situation I meet regularly: the client has an idea that sounds promising, but it is still only a framework, and they cannot produce a sample in the short term.

Strictly speaking, most patent laws do not require you to build a prototype before filing. The problem is practical. When a client cannot make a sample any time soon, it usually means the key technical details have not been worked out. An application written at that stage tends to describe a goal rather than a mechanism. That creates two risks: the application may not disclose enough for anyone to carry it out, and once it is filed, the details you work out later cannot simply be added to it, because you cannot add new matter after filing.

So my advice in these cases is usually to keep developing until you can state, concretely, how the thing works, and then talk about filing.

Can something that looks ordinary still be patented?

Yes, and this is the case clients most often underestimate. A real pattern I have seen: the product is something any competitor could easily manufacture, yet none of them made it. When a sample was built according to the patented technical solution, its performance was far better than the competitors' products.

The lesson is that "easy to make" and "easy to think of" are different things. Inventive step is not about manufacturing difficulty. If a solution was within everyone's reach and still nobody used it, and using it produces a clearly better result, that combination is exactly what an applicant can argue. Chinese examination practice recognises unexpected technical effects as a factor in inventive step, and US practice treats unexpected results as one of the secondary considerations of non-obviousness.

What carries this argument is evidence. The performance comparison must come from real tests on real samples. A patent application has to reflect genuine R&D; numbers that were estimated or invented do not help you and can seriously harm you.

Simple to build is not the same as obvious. If every competitor could have made it but none did, and your version performs far better in real tests, that is a patentable story worth taking seriously.

What should you prepare before talking to a patent attorney?

You will get a much more useful answer if you bring:

• The competing products you know of, and any patents or publications you have already seen.
• A concrete description of the difference: what is placed, connected, ordered or controlled differently, not just what it achieves.
• Your explanation of why peers have not done it, in your own words. This often becomes the core of the inventive-step argument.
• Real test records comparing your version with the alternatives, if you have them.
• Whether you can build a sample now, and if not, what is still unresolved.

With that in hand, an attorney can tell you whether there is something worth protecting, and can then turn to the second question: how to claim it so the product is protected rather than just the prototype.