Having an invention is the point at which most people spend money on the wrong thing.
Three things come first, and all three are free.
Write it down and date it. Search what already exists. Tell nobody publicly.
The search is the one that changes everything — most inventions stop there, and the ones that do not usually change shape because of what it found.
Do these three first
| Step | Cost | Why |
|---|---|---|
| 1 | Write it down in detail, dated | $0 |
| 2 | Free prior art search | $0 |
| 3 | Do not disclose publicly | $0 |
Nothing else should happen before these. Not prototypes, not drawings, not attorneys, not submission services.
Writing it down
| Capture | Why |
|---|---|
| The problem it solves | Frames everything |
| How it works, in detail | Enablement |
| Every alternative you can think of | Cannot be added to a filing later |
| Ranges, materials, dimensions | Amendment room |
| What you tried that failed | Supports non-obviousness |
| The date | Evidence of conception |
| Who contributed what | Inventorship |
Alternatives matter more than people expect. Nothing can be added to a patent application after filing, so a variation you did not write down is permanently unavailable.
The free search
| Tool | Covers |
|---|---|
| Google Patents | Full text worldwide, translations |
| USPTO Patent Public Search | US records, classification |
| Espacenet | Foreign families |
| Product listings and marketplaces | Anything on sale is prior art |
| Trade publications | Where examiners search least |
Search by function, in several vocabularies. What you call one thing, the prior art calls another, and one vocabulary finds one slice.
Search products as well as patents. Something sold but never patented is still prior art and no patent database holds it.
If you find something close
| Reaction | Better reaction |
|---|---|
| Give up | Read it properly |
| Assume it is the same | Read the claims, not the title |
| Assume it blocks you | Check if it is still in force |
| Move on | Find where it fell short |
The reformulation usually comes from here. Understanding exactly what the closest reference discloses shows where the remaining space is.
Check whether it is in force. Only 41.4% of US utility patents reach full term, so a substantial share of what you find is already expired — which blocks patenting but not selling.
Do not disclose
| Action | Effect |
|---|---|
| Public disclosure before filing | Forfeits rights in most countries |
| US grace period | 1 year — US only |
| Trade show, conference, published paper | Public disclosure |
| Crowdfunding page | Public disclosure |
| Social post describing it | Public disclosure |
| Offer for sale, even one unit | Public disclosure |
| Under a signed NDA | Generally protected |
The US grace period does not travel. Relying on it preserves US rights and destroys foreign ones in the same act.
Sign NDAs before technical conversations, not after a quote request. See manufacturing a prototype.
Free help that is not selling you anything
| Resource | Provides |
|---|---|
| USPTO Pro Se Assistance Program | Guidance for unrepresented filers |
| Law school IP clinics | Supervised drafting and prosecution, free |
| Bar association inventor clinics | Reduced or no-fee consultations |
| Patent and Trademark Resource Centers | In-person help at partner libraries |
| Free search tools | Google Patents, Patent Public Search, Espacenet |
Law school clinics do real work under the USPTO's Law School Clinic Certification Program. Capacity is limited and worth applying for early.
None of these has a product to sell you, which is what makes them the right first call.
What "protecting" actually means
| Belief | Reality |
|---|---|
| "A patent stops people copying" | It gives you the right to sue |
| Enforcement is automatic | You must act, and fund it |
| The USPTO polices infringement | It does not |
| A patent means you can sell your product | No — that is freedom to operate |
A patent is a right to exclude, not permission to practise. Whether you can sell depends on other people's claims, which is a separate analysis. See fto meaning.
Is it worth pursuing?
| Question | Why |
|---|---|
| Does the problem cost someone real money? | Determines whether anyone pays |
| How often does it occur? | Frequency drives value |
| What do people do instead now? | The workaround is your competition |
| Was the prior art crowded or empty? | From your search |
| Would you fund it through to grant? | $8,000–$20,000 |
| Would you fund enforcement? | Millions, if it came to that |
Be honest about the last two. A patent you cannot afford to obtain or enforce is a different proposition from one you can.
What it costs
| Stage | Micro entity |
|---|---|
| Free search | $0 |
| Provisional USPTO fee | ~$60 |
| Provisional drafting | $1,500–$4,000 |
| Non-provisional USPTO fee | ~$400 |
| Non-provisional drafting | $5,000–$15,000 |
| Prosecution | $2,400–$8,000 |
| Maintenance, lifetime | $2,894 |
Micro entity status cuts USPTO fees by 80% for the whole life of the patent. Check eligibility before paying anything. See small entity status.
Be careful who you pay
Federal law requires invention promoters to disclose 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
- How many obtained licence agreements
- Any other trading names
| Red flag | Meaning |
|---|---|
| They contacted you first | Enthusiasm preceded evaluation |
| A paid evaluation comes back positive | Positive evaluations are the product |
| Will not give the §297 numbers | The numbers are bad |
| Guaranteed outcomes | Nobody can guarantee a grant or a sale |
| Pressure to decide today | Patent deadlines are real; sales urgency is not |
| Rights transferred to them | Never sign this |
Read the net-profit figure, not the customer count. See invention companies.
If a patent is not affordable
| Alternative | Gives you |
|---|---|
| Defensive publication | Stops anyone else patenting it, very cheaply |
| Trade secret | No expiry — while it stays secret |
| Trademark on the name | Indefinite with renewal, much cheaper |
| Copyright | Automatic for code and content |
| Execution and first-mover position | No legal right, real commercial value |
Defensive publication is underused where the worry is being blocked rather than blocking others.
Filing forecloses trade secret permanently, because the application publishes at eighteen months whether or not it grants — so that fork is decided before filing, not after. See can you patent something and make it free.
What usually happens
| Ipiry Patent Survival Curve v1.0 | Rate |
|---|---|
| Survive the 3.5-year fee (2022 cohort) | 85.8% |
| Survive the 7.5-year fee (2018 cohort) | 64.6% |
| Reach full term (2014 cohort) | 41.4% |
| Abandoned before full term | 58.6% |
Computed from 27,273,654 USPTO maintenance fee records covering 8,262,336 US utility patents — see the patent survival curve.
Nearly three in five granted patents are released early by their owners, usually because nobody practises the claims. That is the honest base rate, and the free search is what keeps you out of it.
I have an invention: the checklist
- Write it down in detail and date it, including every alternative.
- Run the free prior art search before spending anything.
- Search products as well as patents. Anything on sale is prior art.
- Read close references properly. The reformulation is usually in them.
- Do not disclose publicly. Most countries have no grace period.
- Sign NDAs before technical conversations, not after.
- Use free help first — Pro Se Assistance, law school clinics, PTRCs.
- Check micro entity eligibility before paying any USPTO fee.
- Demand the §297 disclosure from anyone selling evaluation or submission.
- Never transfer rights to a firm you are paying. You are buying services.