Submitting invention ideas for royalties is a real route with a narrow opening, and it is surrounded by businesses that profit from people trying to use it.
The genuine version: a company runs a submission programme, you accept its terms, and if they commercialise your idea you receive a royalty.
The terms are the part people skip. Almost every programme requires a waiver of confidentiality, and many consider only submissions with a patent application already filed.
The other version is a paid service that takes money to submit your idea somewhere. Federal law requires those firms to publish their success rate, and the number is usually very small.
What "submit invention ideas for royalties" actually promises
The phrase describes two very different transactions, and the word "royalties" is doing a lot of work in both.
| Version | Who pays | What triggers payment |
|---|---|---|
| Company programme | The manufacturer | Only if they commercialise it |
| Paid service | You pay them | Submission itself; royalties are hypothetical |
In the first, royalties are the entire consideration. No advance, no fee, and payment only if a product ships. That is a genuine deal structure and it is how outside inventors have worked with consumer-products companies for decades.
In the second, the royalty is the marketing. The revenue comes from you, at signature, and the royalty is the outcome being sold rather than the outcome being paid.
Distinguishing them takes one question: who writes the first cheque? If money flows from you before anything is submitted, you are the customer, not the supplier — and the §297 disclosure numbers below tell you what customers of that model typically receive.
Why companies require a waiver
Not to exploit you. To manage liability.
| The company's problem | Why the waiver solves it |
|---|---|
| They may already be developing something similar | Seeing your version creates a claim risk |
| Hundreds of submissions arrive | Any could later be alleged as the source |
| Their engineers work on obvious problems | Convergent ideas are common |
| A confidentiality obligation is unmanageable at volume | The waiver removes it |
Independent development is the core issue. If a company launches a product resembling something you submitted, they must be able to show their own work — and the cleanest way to avoid that fight is to accept nothing in confidence.
Which is why an NDA request usually ends the conversation. They are not being evasive; the policy exists for a reason and it is not negotiable.
What programme terms typically say
| Term | Effect |
|---|---|
| No confidential relationship | They owe you no secrecy |
| No obligation to compensate | Even if they proceed |
| Acknowledgement of parallel development | They may already have it |
| Non-return of materials | Nothing comes back |
| Their governing law and forum | Disputes on their terms |
| Patent status required or preferred | Often the real filter |
| Rate set by the programme | Usually non-negotiable |
Read the patent clause first. Programmes that consider only patented or patent-pending submissions are telling you what they are actually buying: an asset, not information.
Where royalties are paid, consumer product rates commonly run 2–5% of net sales, sometimes with a small advance. See patent royalty rates for how rates and bases work generally — the base matters more than the percentage here too.
What a genuine programme looks like
Real programmes share a set of characteristics, and the absence of them is informative.
| Characteristic | Genuine programme | Fee-extraction business |
|---|---|---|
| Who initiates contact | You do | They do |
| Cost to submit | Free | Substantial upfront fee |
| Published terms | On the company's own site | In a contract you sign |
| Patent status | Often required or preferred | Rarely mentioned |
| Evaluation outcome | Mostly declines | Almost always encouraging |
| Payment structure | Royalty on actual sales | Royalty is aspirational |
| §297 disclosure | Not applicable — they are not a promoter | Required, often resisted |
Genuine programmes decline most submissions and say so. A published programme that is candid about its acceptance rate is behaving like a company managing volume rather than one selling hope.
You should always be the one making contact. Companies with real programmes publish them and wait. Businesses that call you first are selling something.
Free submission is the clearest single signal. A manufacturer evaluating ideas has no reason to charge you; a business whose revenue is the fee has no other reason to exist.
Which industries actually run programmes
| Sector | Likely to accept? | Why |
|---|---|---|
| Consumer products, housewares | Yes | Short cycles, external ideas fit |
| Toys and games | Yes | Long tradition of outside inventors |
| Tools and hardware | Sometimes | Incremental improvements suit |
| Food and beverage | Occasionally | Usually formulation, heavily regulated |
| Automotive | Rarely | Integrated engineering, safety qualification |
| Technology and software | Rarely | Fast internal development, high claim risk |
| Pharmaceuticals | No | Regulatory pathway, enormous development cost |
The pattern follows product cycle length. Where a product can go from concept to shelf in a year, an outside idea is useful. Where it takes a decade and hundreds of millions, it is not.
What the royalty is calculated on
Where a programme does pay, the base matters as much as the rate — and submitters rarely negotiate either.
| Base | Effect on payment |
|---|---|
| Net sales of the finished product | Highest |
| Net sales of the component embodying the idea | Much lower |
| Wholesale rather than retail price | Roughly halves it |
| After "customary" deductions | Reduces it further, by an undefined amount |
A 3% royalty on wholesale of a component is a fraction of 3% on retail of the product, and both are described as "3%".
Programme terms usually set the base and the rate together, with no room to move. Read what the percentage applies to before assuming what a successful submission is worth.
Worked example: two submissions
The same kitchen tool improvement, submitted two ways.
Without a filing
| Step | Outcome |
|---|---|
| Requests NDA | Declined; submission policy sent |
| Submits under waiver | Acknowledged in 3 days |
| Response at week 7 | Declined, form letter, no reason |
| Submits to two more companies | Both decline |
| Position | Idea disclosed three times, nothing owned, no recourse |
With a provisional on file
| Step | Outcome | Cost |
|---|---|---|
| Prior art search | One close reference; claim narrowed to the locking mechanism | $0 |
| Provisional filed | Priority secured | $130 + $1,800 drafting |
| Submits, noting patent pending | Routed to the programme's IP review rather than general intake | — |
| Evaluation at week 9 | Progressed to product team | — |
| Offer at month 5 | 3% of net sales, $5,000 advance, exclusive licence | — |
| Non-provisional filed at month 10 | Funded partly by the advance | $800 + $8,500 |
The filing changed which queue it entered. Programmes commonly route patent-pending submissions differently, because those are the ones they can actually transact on.
It also changed what a decline would have cost. In the first version, three disclosures and nothing owned. In the second, an asset that could be taken elsewhere.
Note the honest part: most submissions in both queues are declined. Filing improves the odds and the position; it does not make acceptance likely.
After a decline
Most submissions are declined, and what you hold afterwards depends entirely on whether you filed first.
| Submitted without a filing | Submitted with a filing | |
|---|---|---|
| What you own now | Nothing | A priority date |
| Can you approach others? | Yes, but disclosed again each time | Yes, from the same position |
| If they launch something similar | No recourse | A patent claim may exist |
| Cost of the attempt | The disclosure | The filing fees |
A decline is not a verdict on the idea. Companies decline for roadmap reasons, capacity reasons and legal caution, and rarely explain which.
Keep a record of what you sent and when. Should a similar product appear later, the submission record is the starting point for any conversation — and with a filed application behind it, a considerably stronger one.
Paid submission services
A different business. These firms charge you to prepare and submit ideas rather than being companies that might buy them.
Federal law requires disclosure under 35 U.S.C. §297, in writing, before you contract:
- Total customers in the past five years
- How many received a net financial profit from the services
- How many received licence agreements
- Any other names the promoter has traded under
Request it in writing and read the second figure. A firm with several thousand customers and a handful who profited has answered the only question that matters.
Failure to provide the disclosure gives you a civil action for damages, and the USPTO publishes complaints it receives about invention promoters.
| Signal | Interpretation |
|---|---|
| Contacted you first, then enthusiastic | Enthusiasm preceded evaluation |
| Paid evaluation returns positive | Positive evaluations are the product |
| Fee for submitting to manufacturers | Submission costs little; the fee is the business |
| Will not provide §297 numbers | The numbers are bad |
| Urgency about signing | Filing deadlines are real; sales deadlines are not |
| No registered attorney or agent named | Nobody accountable is drafting |
Cheaper legitimate alternatives exist. The USPTO Pro Se Assistance Program, law school IP clinics, bar association inventor clinics, and registered patent agents charging by the hour all cost less and do more.
What to do instead
The sequence that works is the same one that works for any sale.
| Step | Why |
|---|---|
| 1. Search the prior art | Free; eliminates most ideas |
| 2. Reframe as a mechanism | A result is not patentable or sellable |
| 3. File a provisional | Priority, modest cost |
| 4. Identify who already makes similar products | They are the realistic buyer |
| 5. Submit or approach with the filing noted | Changes which queue you enter |
| 6. File the non-provisional within 12 months | Creates the assignable asset |
Approaching a company that already makes similar products is the highest-value step, and it is also why filing first matters — those are exactly the companies most likely to have parallel work.
See sell my idea for the broader routes and invention ideas for how to test an idea before spending anything.
Most patents never earn a royalty at all. Only 41.4% of US utility patents granted in 2014 reached full term, and 58.6% were abandoned when owners judged the maintenance fee not worth paying — see the patent survival curve. Royalties are concentrated in a small number of patents covering things other people need.
Submitting invention ideas: the checklist
- Search the prior art before submitting anything. Free, and most ideas stop here.
- File at least a provisional first. It changes which queue your submission enters and gives you something if it is declined.
- Read the submission agreement in full, particularly the confidentiality waiver and the parallel-development acknowledgement.
- Expect no NDA. Requesting one usually ends the conversation, and the policy exists for reasons unrelated to you.
- Check whether the programme requires patent status. Those that do are the ones that actually transact.
- Target industries that run genuine programmes — consumer products, toys, housewares, tools.
- Demand the §297 disclosure from any paid service, in writing, before paying anything.
- Read the net-profit number, not the customer count. That is the success rate.
- Check whether the programme requires exclusivity during review before submitting elsewhere.
- Treat a decline as normal. Most submissions are declined; with a filing in place you still own something afterwards.