Next | Previous | Contents

The Provisional Application

1996 TEN article by Ed Zimmer, The Entrepreneur Network.

As we've discussed previously, for the inventor with a new product idea, there are basically just two paths to get that product to the market: Venturing or Licensing.  There are many strategies for pursuing either path.  There is no "right" way!  It is literally true that there are as many ways of successfully "doing business" as there are people doing it.

The patent law change in June 1995 introducing the provisional application opened up a new cost-effective strategy for those interested in pursuing Licensing.  Readers may wish to consider this new option.  It is outlined below. 

Provisional Application

A provisional application is something you write up yourself.  Its only statutory requirement is that it disclose your invention in sufficient detail that anyone skilled in the applicable art can make and use the invention.  You may attach any sketches, drawings, photos, etc.  that may be helpful in describing how the invention can be made and used.  There are no format requirements on either the text or drawings, although the sheets should be (or be foldable to) 8.5x11.

The provisional application can be filed with the Patent Office for $75 and will buy you one year of protection.  Within that year you must either file a formal patent application (average cost about $5,000) or abandon the idea.

Prospect List

Before filing the provisional, you should compile a complete list of potential licensees.  By complete, I mean a sufficient list that after you've contacted them all, if you've found no interest, you'll be satisfied that your product isn't licensable — and your only alternatives are Venturing or going on to your next idea.

Your prime candidates are companies who are already selling into your target market.  If your invention is in your trade, you already know many of these companies.  Pull the rest out of your trade journals and trade directories.  If it's outside your trade, try to find someone in the trade (preferably a buyer) who'll loan you, or at least let you look at, their trade journals and directories.

If it's a consumer product that would sell through stores, go to stores that would be likely to carry your product, determine where in the store your product would likely be located, and collect all the brand names and company names you can find in that vicinity.  Then find yourself a good reference librarian.  A reference librarian is one who knows all the directories and databases and how to use them.

A good reference librarian takes great pride in being able to show you how to find answers to any questions you might pose.  They'll show you how to associate company names with brand names, how to get address, phone number (and other useful data such as annual sales) for company names and, to the extent you can accurately define your target market, show you how to find other companies selling into that market.

Secondary candidates for your prospect list are companies already making similar products - but not currently selling into your target market.  These are "possibles but unlikelys".  It's much more difficult, costly - and risky - for a company to tackle a new market than it is to produce a new product for a market they're already servicing.  If they don't have the in-house capabilities to manufacture the product, they'll simply have it contract manufactured.

Postscript (5/97): Don't overlook existing patent records as a source for compiling your prospect list — especially now that on-line patent searching is becoming quick and easy.  Any company maintaining patents in a class/subclass related to your invention is certainly a prospect — see Patent Searching

Licensable-Quality Patent

Now you're ready to file your provisional application — almost.  One of the downsides of the Licensing path is that the odds are long.  Rules of thumb in the field are that less than 1 in 100 patents ever license and less than 1 in 500 ever make back more money than they cost the inventor.  A major reason for these long odds is that most inexperienced inventors don't take the time to determine whether they can get a licensable-quality patent.

Recognize that the only legal part of a patent is the claims section at the very end.  All the rest of the patent — the drawings, specification, etc. — are only there to help someone interpret the claims.  The claims section is a metes-and-bounds description of your intellectual property in the same sense that a deed can be a metes-and-bounds description of real property — it surrounds it.  When you hire a patent attorney to draft a formal patent application, what you're really hiring him to do is to surround as much intellectual property for you as he can without overlapping someone else's intellectual property.

Recognize also that the first thing any potential licensee who's interested in your product is going to do is to see if they can design around your patent.  And they're not doing that because they're bad guys.  They're doing it because, if they do enter into a license agreement with you, and the product sells well, all their competitors will simply design around the patent, and they'll be stuck paying you a royalty and their competitors won't — a situation they wish to avoid.

So, before you file your provisional, you really ought to determine whether you have anything to license.  From what I've been seeing, if you'll take this step, 95% of the time you'll decide not to go ahead.  You'll find that you simply can't get the protection that you need to have any chance of licensing (and for those few inventions that do pass this screen, you'll find that your odds of licensing greatly improve — to as much as maybe 1 in 5).

What you should do with that provisional application is, instead, take it to a patent attorney.  Have the patent attorney do a patent search.  The search will yield copies of all patents that surround your invention.  Once the patent attorney has studied the claims in those patents, he'll have a pretty good idea of what he can claim for you.

At that point, you must do two things.  First, make the attorney explain to you in language you can understand what it is he thinks he can claim for you.  Second, how claims can be designed around.

Then, take the time to try to design around yourself.  If you find that the proposed claims are "impossible" to design around — i.e., they protect the only economical ways of providing your intended user benefits — then go for it.  Otherwise, forget the patent — and forget Licensing.

If you decide to go ahead, have your patent attorney review your provisional.  One of the restrictions of a provisional application is that you cannot "claim" anything in a later formal patent application that you have not disclosed in your provisional.  Your patent attorney will show you where to "broaden" the language in your provisional, if necessary, in order to adequately cover what it is that he thinks he can claim.

The patent attorney's services, for the search and the two explanations, will add $500-800 to your front-end costs — but will save you time and phone bills for the 95 of 100 that aren't worth pursuing.  Your choice.


Now — finally — you can file your provisional application.  You'll need two forms: a Provisional Application For Patent Cover Sheet and a Small Entity Statement.  The former is simply a cover sheet for your application; the latter gets you the half-rate fee.  (The provisional filing fee is $150 if the Small Entity Statement is not enclosed.) If you're dealing with a patent attorney, the attorney will give you these forms; otherwise call the PTO's information number, 800-786-9199, and find out how to get them.

As soon as you get confirmation back from the Patent Office that they've accepted your provisional, your next step is to get on the phone.  Call into each company on your list.  Try to get through to the person responsible for "dealing with outside inventors".

Every company, at least those with competent legal counsel, have someone designated to deal with outside inventors.  There's sufficient exposure to "you-stole-my-idea" lawsuits downstream, that most companies will try to limit such contacts to one person.  If you contact and start dealing with the wrong person — even inadvertently — when the right person hears about it, the whole deal is likely to blow up (and the person who dealt with you is apt to get fired).

If you hit a dead-end in trying to get through to the right person, try Legal (if they have a Legal Dept.).  If not, try Purchasing; someone in Purchasing is certain to know who the right person is.  (And if that doesn't work, ask who's their outside patent counsel... or, finally, their legal counsel).

When you get through to the right person, tell them you'd like to submit a new idea to the company; do they take them?  Some companies don't.  If they do, how do you go about it?  Most companies will have a "disclosure" form that you'll need to sign.  This form is basically a "waiver of confidentiality".  It basically says that in any disclosures you make to the company you are relying solely on your patent protection.

Since you have "patent pending" — that's what your provisional application gives you — you can freely sign their form.  So you have them send you a copy.  When you get it, you sign it and send it back in with a good one-page description of your invention — to the attention of the person you talked with.  Never send an invention in "blind".  Apart from the disclosure-form problem, it's sheer chance whether it will ever get to the right person.

That one-page description is important.  It wants to be business-like, professional, no hype, no hard-sell.  It wants to clearly describe your invention and its perceived advantages in the first couple of paragraphs, followed by maybe a paragraph explaining why the company may be interested in licensing it.

The person you're sending this to will be making his decision in the first 20-30 seconds of reading your letter whether this is something the company might be interested in.  Make sure you get your best message across in that time frame.  It's ok to attach an illustration or photo if it's applicable, but don't send prototypes, supporting data, etc.  If they're interested, they'll ask for it.

Within a couple of weeks, call that person back.  Verify that they received your material.  See if there's any preliminary interest.  If there isn't, try to get a feel for why; there might be something you can do about it.  If there is, find out when you should call back next.

Go through your whole prospect list this way — just as fast as you can.  Don't stop just because someone expresses preliminary interest.  It's not at all unusual for initial interest to peter out once they've had more time to look at it.  And, besides, you'd like to have multiple companies interested to give you some negotiating leverage.

Strategically, you'd like to get through your prospect list within a couple of months.  If a company's interested, it can take them up to 6 months to decide whether they're really interested.  Your contact is typically only the initial screen.  If he thinks your invention might be of interest to the company, he'll forward it to the appropriate division or department for review.  If it gets through their review, it's then likely to go through an engineering review, a manufacturing review, a marketing review, maybe a sales review, a new product committee, etc., etc.

But if you did your pass in 2 months, and they take 6 months to come down to a licensing agreement, you're still only 8 months into the 12 months of your provisional — plenty of time to negotiate the costs of your formal patent into the licensing agreement! If you get on the ball and don't procrastinate, it's likely you'll never have to face that $5,000 formal patent cost.

If you go through your whole list and no one's interested — i.e., if you did a complete list (see earlier) — the invention's just not going to license and your only alternatives are to try Venturing it or go on to the next invention.

However, if you did your phone work diligently — including closure back with those you sent materials to — it's likely you'll have a pretty good understanding of why the invention didn't license.  Companies are quite reluctant to explain why they turned down an invention, but just a phrase here and sentence there from a large number of rejectors can give you a pretty good idea of what the problem was — which knowledge you can then use to avoid that problem with your next invention.

Licensing Agreement

If some company does come down to a licensing agreement, how do you negotiate the agreement?  For the inexperienced, the simple answer is you don't.  You take what you can get — and count your blessings.  The company will typically draft the licensing agreement — and that's their offer.  Take it or leave it.  You might be able to negotiate some fine points, but if you get too demanding, it's likely they'll just walk away.

In fact they're likely to have tested you very early on — by asking what you're looking for.  Find out from your patent attorney (or the library) what the going royalty rate is in that industry.  In the absence of better information, figure about 5%.  When they ask that question, respond with something like "a 6% royalty and acceptable minimums".

Their real reason for asking that question is to judge whether you're going to be "reasonable", i.e., business-like, to deal with.  If your response indicates you're not, it's likely they'll just walk away at that point — and not even try to determine if the product itself is of "real" interest.  A product they can't "reasonably" license is of no more interest to them than a product they can't sell.

They're expecting an exclusive agreement, i.e., that you'll be assigning your patent rights exclusively to them.  Under an exclusive, you must have minimums (or a reasonable alternative).  Otherwise, they can just `sit on' the product, doing nothing with it — and your royalty revenue will be an acceptable percentage of nothing.

During their review of your invention, long before negotiations, it's likely you'll be talking with others in the company.  With each, try to get a `feel' for their expectations of sales as a function of time.  (You're not likely to get this kind of information from your initial contact — he's the one you're going to be negotiating with and he knows why you want it.) Then when the subject of minimums comes up, use some reasonable percentage of those numbers in negotiating the minimums.

If you have detailed knowledge of the industry, use it.  But use it softly: "I've heard that..." "I believe that..." Even if you have as much knowledge of the industry as the guy your dealing with.  In any negotiation, if a deal can't be made, you want to be the one who walks away — you never want to cause the other party to walk away.

When you get an offer from a company — in the form of a licensing agreement — you have to decide whether to take it or leave it.  Look objectively at your alternatives.  If they don't look that good, in general, take it.  Even if it's much less than you'd hoped.  Something is generally better than nothing.

However, before you sign the agreement, take it back to your patent attorney.  The agreement deals with intellectual property law and that body of law is insanely complex.  The attorney will call to your attention any `hookers' you may not be aware of and, if minor, will typically negotiate their resolution with the company's patent attorney, and thereafter give you the go-ahead to sign.


The provisional application offers a new and low-cost strategy for trying out new ideas on prospective licensees.  However, be sure you clearly understand the two cautions below:

Caution: This strategy assumes that you will be writing the provisional application yourself.  If you require a patent attorney to write it for you, it will cost almost as much as a formal application.  A patent attorney will write it in the format with which he is most familiar — namely as a formal application, less only the claims section.  Given that, any cost-effectiveness benefit is lost and we would recommend that you forget the provisional and just go with a formal (i.e., non-provisional) application.

Caution: This strategy assumes that you will not procrastinate, i.e., that you will poll all potential licensees within at least a 4-5 month period.  If you have any thoughts of extending your time-to-license with a 2nd provisional, forget them.  If you let your 1st provisional lapse without filing a formal application, any contacts you made with potential licensees under that 1st provisional will constitute a "public disclosure", voiding immediately the availability of any foreign patent rights and voiding after one year (of your first such disclosure) the availability of any U.S. patent rights — and a 2nd provisional will not prevent this.

Postscript 2/98

Because of the flood of Provisional Patent Applications (PPA) being received by the Patent and Trademark Office (PTO), applicants are receiving no notice of receipt for several months.  Since the PPA only buys you 12 months, if you're using it to determine licensing interest before committing to a formal application, you can't afford to lose that time.  Some advice from David Pressman, author of the inventors' bible, Patent It Yourself, in response to a question of whether to send a PPA via registered mail:

"Never send anything by registered mail unless it is inherently valuable (like jewels or currency) or an important foreign mailing.  If you want proof of receipt, use certified mail, which is much cheaper.

However, you should never use either certified or registered mail for any correspondence to the PTO.  If you want to be sure the PTO gets something on a certain date — and have a sure way to recover it if the Postal Service or PTO loses it (not uncommon now) — use Express Mail with a certificate of Express Mailing, or at least the Express Mail number, on your cover letter.  If you use Express Mail, the PTO considers that you've filed it with the PTO when you get the Express Mail receipt from the postal clerk.

If you've filed a PPA and haven't received anything back for a long time — and didn't use Express Mail or a receipt postcard — look at your cancelled check (if you have one).  Your Serial Number might be stamped there.  If not, then call the PTO's Office of Initial Patent Examination (OIPE), 703-308-1202, and see if they can track it down.  If you have no Express Mail receipt, no receipt postcard, no cancelled check, and the OIPE can't find any record of it, then it was probably lost in the mail or the PTO."

Next | Previous | Contents