What's the first step?

Sometimes the first step is realizing that an idea is not an invention – although it may lead to one. A lot of would-be inventors start out with an idea for a product, without realizing that there’s no invention until one figures out how to make the product.. 

One definition of “invention” is: a completed, new and unusual solution to a technical problem that has economic value. 

There isn't one first step that suits every situation. The best choice for a first step depends on your own knowledge of the market and of the technology.. Often the best approach is to try to get partial answers to some of these questions:


Picking one of these questions and working it to death is usually not a good approach. It's better to get rough estimates in several areas and then go back and refine the estimates later.


Skipping one of those questions because you’re sure you already have a positive answer is dangerous. There are lots of inventors who are sure it will work. There are lots of inventors who are sure everybody will want it. There are lots of gun-shot victims who were sure it wasn’t loaded. There are lots of assumptions that can get one into trouble.

Getting answers when the idea is a total secret:

No idea is a total secret. A good idea is usually an improvement on something well-known. A good idea is almost always a solution to a widely known problem. You can set up questions about what people don't like about the old thing and what kinds of solutions other people have proposed to solve the known problem without letting the cat out of the bag. Take a look at my recommended six-step process for being practical about secrecy.

Expect to spend some time defining the market

The enthusiastic feeling that "there's nothing like it out there and everybody will want it" may be a good starting point, but it's no place to stop. If there's really nothing at all like it out there, you've got a big problem -- there's no market for it either. On the other hand, if everybody will want one as soon as they see it, you've got a different big problem -- the only things everybody wants are things that are already known and therefore not usefully protectable..

Inventors working on an industrial product, a new tool, or some similar item generally have an easier job of market research. They can identify what the invention is going to replace, and what the costs and benefits of the replacement are going to be. Not surprisingly, individual inventors in these areas tend to do a lot better than average.

Expect to spend some time defining what the innovation really is

Finding out what is "kind of like it" is important. The scope of protection that you can hope to get is tightly bound up with the degree to which your invention differs from the item it replaces. Some things to do in this area are:

Do I need to make a prototype?

If you can make an actual working product, that should be one of the first steps. There's always room for doubt about how well something will work until it actually does work. Even simple products commonly take two or three tries before they come out just right. In order to get patent protection, the inventor must be able to describe the invention well enough that someone "ordinarily skilled in the art" can practice the invention. The best way to know that the description is good enough is to make and test the product before writing that description. On the other hand, if making the first article is going to be a long expensive process requiring outside investment, you may be better off looking into patent protection on the invention before actually making something.

David A. Kiewit
Registered Patent Agent
5901 Third St. South
St. Petersburg FL 33705-5305
+1 (727) 656-0669 voice

+1 (760) 841 0989 fax
questions to: [email protected]
Copyright 2002-2017  by David A. Kiewit
All rights reserved



ABCs of Patents

Practical Secrecy

Patent Searches


Author! Author!


Design Patents

Utility Patents


Types of Intellectual Property



Intellectual Property Issues for New Companies

Innovation in Production

Steps to getting a US patent