Having a patent does not guarantee that you will make money -- 99+% of patented inventions never make it in the marketplace -- or even that you will not infringe an unexpired broader patent of another.

It's expensive to get a patent.  $1000 in government fees to get it issued, and another $3000 in government fees after it issues, just to keep it alive for its full term -- it expires 20 years after the application date and is not renewable.  And it's not a do-it-yourself undertaking.  You'll spend as much or more in attorney fees and search costs, so by the time you're finished, you'll have spent up to $10,000 even if there are no complications.  You can invest $10,000 in mutual funds for 20 years and turn it into $50,000 or more.  So you should be able to see profits comfortably over and above that amount to make the investment worthwhile. 

And patents do not enforce themselves either.  If you have to sue an infringer, that can cost even hundreds of thousands of dollars even if you win.  So even if you can come up with the litigation costs, or find an attorney to represent you on a contingent fee basis, as a practical matter you can only go after infringers with big sales to recover damages greater than your costs.

The test for patentability is threefold -- it must be

• new (not on sale or in public use or shown or described in any patent or printed publication anywhere in the world)

• useful (not usually a problem), and

• unobvious to persons of ordinary skill in the field of the invention.  Here's the rub.  Routine, or ho hum, improvements or variations are not patentable.  There must be an element of invention -- something that goes beyond ordinary product improvement evolution. Perhaps it provides a new result or a new function, or achieves an old result in a much better (e.g., faster, less expensive, quieter, cleaner, etc.) way.  But something over and above what a good engineer would be expected to do in the ordinary course of business.

........ what is the best way to get started?

I suggest that you start by doing some serious reading.  There is a book entitled The Inventor's Desktop Companion, by Richard Levy, a successful private inventor.  It's available at many public libraries. You can browse its contents at www.aspensite.com/ibc/inventors-desktop-comp.html.

Go to the US Patent and Trademark Office (PTO) website at www.uspto.gov and read "Basic Facts about Patents".  You can also search the PTO database at that web site for prior patents to see if your idea may have been previously patented by someone else.

Most of all, before deciding whether to make a major investment in seeking a patent, spend some quality time at the web site of Jim White to see if your invention could be a marketplace winner.  His web site is at www.willitsell.com.

Finally, there's lots more to making money with a product idea, even if it's new, than most people realize.  And there are lots of scam operators out there, trying to mislead you into thinking otherwise.  For heaven's sake, whatever you decide to do, don't deal with the invention promotion companies that advertise to new inventors -- they will tell you that your idea (whatever it is) is the greatest thing since sliced bread, write it up in a brochure that promises sales in the millions or even more, and send copies of the brochure to hundreds of big companies who will never even read it.  They will charge you a thousand bucks, more or less, up front, and then there will be no responses from the companies.  Their batting average in making even a nickel for the inventor is a very, very small fraction of one percent.  See the warning bulletin from the US Federal Trade Commission at www.ftc.gov/bcp/conline/pubs/services/invent.htm

See also his "10 Common Myths & Misundestandings about Patents http://www.score.org/myths_patents.html