I work for a software company. I have the beginnings of an idea of, what I believe will be, a patentable algorithm. I don't know yet if it will actually be accepted, but anyway, won't know until I try...

If I do come up with a working algorithm/code, I would like to let my employers know that I created this, and want them to handle patenting it (if they deem it worthy). Generally, what are the steps I need to follow to let my employer know that I have this algorithm that I want patented by the company (in the United States)? I know each company has it's own set of procedures, etc., but, I'm just looking for pointers in the right direction.

The reason I want to involve my employer, is because I'm bound by contract to do so, and also, because that takes away a lot of the expenses involved. And, of course, they have a legal team on payroll (presumably), just for this purpose.

The company is not too small, but no, I don't have any access, or even visibility, to the 'big' bosses like the CEO, etc. (yes, I've heard of email, but let's assume that the CEO probably has filters to weed out mails by lowly people such as myself :-) ) The only people I have access to are three levels of management above me (manager, manager's manager, manager's manager's manager). I suppose I could even get in touch with the HR team, if that's what's required. As you can see, I have no idea about this, and don't even know where to begin or whom to contact.

Even though I'm contractually obliged to 'immediately disclose any inventions' to the company, with 'no expectations of receiving additional compensation', I still want to monetize this as much as possible (perhaps a pay hike/bonus, promotion, etc.). In view of this, I would like to avoid handing this out to my immediate manager and the manager's manager, since I believe this would lead to this effort either being throttled at the source, or worse, presented to the 'big' bosses as some idea that they came up with ('team effort', etc.). It may be possible for me to prove otherwise, but I don't want to go down that path. I believe presenting this directly to the person three levels above me would be best in the interests of furthering my career.

Preferably, I'm looking for answers from people who have actually done this before, so that I know what (not) to do.

PS: Let's not argue about whether algorithms are patentable or not...my company has a few of them already, so I know it's possible.

  • What country are you in? Based on your reference to a work contract, I assume that it's not the USA. Commented Nov 10, 2018 at 22:03

7 Answers 7


I am listed as the inventor on a patent. It doesn't mean anything. It's actually kind of hard to do software development for any length of time and not end up with your name on a patent application.

It really isn't something your career is going to "explode" off of. Honestly, I get no joy from my name on the patent, and since I assigned it to the company I was contracting for (for a fee, of course), it doesn't bring me any income, either.

Were I you, I'd use this as a publishing opportunity. I'd write up an article to submit to trade magazines about your algorithm, and try to build up some recognition in the industry. Submit the proposed article to your company hierarchy for approval. If they decide it is something worth patenting, your name will be in front of too many people for someone to steal the credit.

FWIW: Were I a patent examiner, I would not have granted the patent I am on. The whole software patent library needs to be flushed. I seriously doubt that more than 5% of them are truly innovative. I was surprised that my customer wanted to patent what I did. All I did was take 6 different existing technologies and use them to solve a business problem. I'll bet there are 100 people over on StackOverflow that do that all the time, too.

You figure out how to make an airplane fly at supersonic speeds without fossil fuels, go get a patent. You figure out how to sort an array with 30% fewer CPU operations, write a paper.

  • I didn't even think of publishing this as an article. Thanks.
    – ArM
    Commented Mar 2, 2015 at 15:18
  • 4
    +1000 for The whole software patent library needs to be flushed. Commented Sep 23, 2016 at 23:50

I believe presenting this directly to the person three levels above me would be best in the interests of furthering my career.

You know your immediate manager and the manager's manager better than I do, of course, but in almost every conceivable situation, cutting out your manager and going over their head is a career-limiting move, not a career-furthering move.

It is very likely that this will be perceived as a hostile act - and frankly, rightly so. You say yourself that you don't want to let your manager get involved because you believe they will either throttle your idea or try to steal the credit for it. You're already gearing up to launch a hostile operation in the world of office politics, so you would have to expect retaliation. Are you confident enough in your politicking skills to defeat two levels of management above you?

Remember, the game of office politics is like the Game of Thrones - you win, or you die.

Now, you say that the company has already patented a few algorithms. If you want advice from people who have been in your situation, take a look at those algorithms. Are the inventors of them still at the company? Do you know any of them, or are they in completely separate teams? Even if they are, you could informally get in touch. Fire off an email and ask them what the process was that they followed to get their algorithms patented.


Even though you have indicated that you would rather not, you really should talk to your boss.

Beginnings-of-ideas are not patent-able. And taking a concept to the patent stage takes time, effort, and money. Your company may or may not be willing to do this with you. If not, you may be able to pursue this on your own, on your own time.

As far as monetizing, I'd assume this patent would belong to your employer, if you work with them. Don't expect much (if any) monetization to flow your way. Perhaps a token of their appreciation.

On the other hand, it might indicate your dedication to your company, and thus contribute to a promotion or raise somewhere down the road.


Almost all US companies make you sign any intellectual property agreement that clearly states that whatever you come up with is theirs.

Most likely you are required by contract to immediately disclose the invention. Many companies actually do have "invention disclosure forms" that you can use. So you should write it up, put your name as inventor on it, sign and date it, and hand it to your manager or your company's IP department if you have on. Keep one original for yourself.

From that point on it's out of your hands. It's up to the company to decide whether to pursue the patent or not. This is a complicated decision and it depends not only and the merit and value of the idea itself but also on business plans, infringement detection and enforcement strategy. In most cases that I was involved in concerning algorithms, the company decided to keep it as a trade secret and not as a patent. It's almost impossible to proof infringement with algorithms and if you can't detect and enforce, the only thing the patent does is teach other people how to do it.

In terms of what you can get out of it, you are mostly out of luck. If the patent gets filed, you are entitled to $1 (no joke), but that's it. The patent MUST be filed with you as the inventor. If you are afraid someone may steal your idea, you can have the original disclosure notarized. Having patents in your name looks good on the resume and it certainly boosts your reputation in the company. If you want to rub it people's face, have it framed and hang it in your office.

In general it's a long and tedious process. It's not unusual that the time between disclosure and grant (if you actually get it) approaches 10 years. These days the only way I notice that one of my ancient filings got granted is by getting junk mail from companies that sell patent frames.

  • The first line is why I read employment agreements and any other contract carefully. I'll agree to have any work product property of the company (It is what they pay me for) but if there is any thing that says EVERY thing I do is property of the company I will walk away. What I do in my free time is mine and mine alone. (Software developer specifically) Commented Feb 27, 2015 at 21:20
  • "These days the only way I notice that one of my ancient filings got granted is by getting junk mail from companies that sell patent frames." - Exactly how I found out. Commented Feb 17, 2017 at 18:35

Documentation first

So, you have a potentially lucrative idea. Awesome! First, document when and how the idea came to be details about the idea, etc. This is both to help demonstrate it was your idea if someone tries to steal it, as well as for taking to whomever will need to patent the idea.

Who would patent it?

The first question you'll need to get on the table is in the event this was deemed patent worthy who would actually be the one who would file the patent on behalf of the company. More often than not you have someone who's job is to be the middle man between the lawyers and engineers, typically this person is a department head, but that varies place to place.

Found the, now what?

Okay so you found who you ultimately need to sell the idea on. Let's say it's your boss's boss's boss and they are the department head of R&D. If you're able to open a direct channel send them an email, or quickly say hi and say "Hey Jaime, next time you have ten to fifteen minutes I've got something I'd like to show you that could benefit the company"

In the event you do not have direct access to the necessary individual you can request it with however far up the chain you can reach. (or HR) it may or may not be approved though. If you can't reach the intended audience without going through others that's life. (If you have to go through someone, just give them the overall idea without outlining all the details)

The pitch

Okay, so you've reached your target audience and the next ten to fifteen minutes are yours to convince them they want it. At this point you just need to explain to this person why they want to both use and protect this idea for the company. How will it make them money, this part is actually pretty straight forward.

Benefit from the idea

This isn't easy... Even if the idea is 100% your work, because it's on the company dime the idea is technically property of your employer. So you're not convincing them give you a something nice for the idea, but rather because of the idea. You could try and get a dialog started something like "If I come up with any other ideas is there anyway to get some kind of kickback?"

Or you could wait until the next time you're going for a raise or promotion and use this as a pretty solid card to play.

Other options might be available, but just keep the context in mind that this already belongs to your employer, you need to sell them on how these ideas make you an asset worth investing a little more into.


If the idea is a good one (where good relates to the ability to get money from it), maybe instead your spouse or sibling/parent (someone you trust) might have come up from it and you are just remembering it wrong?

Then maybe they can start their own company, apply for a provisional patent, and then maybe that company can approach your current company and sell the idea to them? With the rider that you have some responsibility for it?

This of course is only if you've remembered how things went differently from what you have described in the question. Obviously if you were responsible, then you should tell your company. And it is hard to manage how senior management take any decision, so there is little guarantee you will gain any of the outcomes you wish.


The status of whether software is patentable, in the US, is ambiguous.

No law has specifically addressed the software patents issue, and the courts have been mixed.

Most professional software engineers, and professional electrical engineers are against software patents. Most business firms don't think software patents are useful. The following paper quantifies this:

Graham, Stuart J. H., Merges, Robert P., Samuelson, Pamela and Sichelman, Ted M., High Technology Entrepreneurs and the Patent System: Results of the 2008 Berkeley Patent Survey (June 30, 2009). Berkeley Technology Law Journal, Vol. 24, No. 4, pp. 255-327, 2009; CELS 2009 4th Annual Conference on Empirical Legal Studies Paper.

You should not attempt to patent this work. It is very unlikely that you will realize any monetary compensation from the company.

If you want recognition, write a paper about it. If you decide to write a paper, you should, of course, get permission from your employer, unless the algorithm was developed on your own time.

  • 2
    Disclose it to the company. Let their lawyers decide whether they want to patent, publish, retain it as a trade secret until they can use it in a product, or let you own it and decide whether/how to invest your own money in it.. That decision is theirs on behalf of the company, not yours, unless your employment contract says otherwise.
    – keshlam
    Commented Feb 27, 2015 at 23:50
  • He doesn't have to disclose it if he doesn't want to, and if he does, I stated that he should consult with them first.
    – daaxix
    Commented Mar 3, 2015 at 17:28
  • If he doesn't disclose it, his contract may keep him from legally being able to do anything else with it. And I hesitate to suggest breaking the contract; "the object of the game is to be as honest as the law allows."
    – keshlam
    Commented Mar 3, 2015 at 17:37
  • He may not want to do anything else with it, and if he keeps it in his brain, they can't know he had the idea while he worked for them, it is ridiculous that a company can claim IP rights to work done on our own time, and they have no claim to ideas in our heads...
    – daaxix
    Commented Mar 3, 2015 at 17:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .