The whole picture
Patent strategy, decision by decision.
A patent strategy isn't one choice — it's a sequence of them: whether to file at all, what to protect, when, for how much, and how to turn the result into leverage. Here is the whole sequence in plain English, with a free tool and a deep-dive for every decision.
New to all this? Start with what a patent actually is.
Decision by decision
Six decisions behind every patent strategy
Work them in order. Each one narrows the next — and skipping one is how inventors spend thousands protecting the wrong thing, or lose the right to protect it at all.
- 01
Should you file at all?
A patent is not always the right move. It trades secrecy for a time-limited monopoly and real money. Before anything else, decide whether the invention is even worth protecting — and whether a patent (versus a trade secrettrade secretConfidential business information that gains competitive value from being kept secret. Unlike patents, lasts as long as the secret is maintained.Read more →) is the right instrument for it.
Is my idea patentable?Free check: novelty, obviousness, eligibility — in two minutes.Patent vs trade secretDisclose for a 20-year monopoly, or keep it secret forever?How strong is a patent?The ten signals that separate a real asset from a wall plaque.What can be patented?Software, algorithms, business methods — what actually qualifies. - 02
What, exactly, should you protect?
You rarely patent a whole product — you patent the specific, non-obvious mechanism that makes it work. Scope it too narrow and a competitor designs around it; too broad and the prior artprior artEarlier patents, publications, or products that existed before this patent's filing date. Patent claims must be novel over the prior art.Read more → sinks it. This is where most of the value (or waste) is decided.
Prior-art searchFind what already exists before you spend on a filing.Freedom to operateCan you even sell this without infringing someone else's patent?Novelty (§102)What counts as new — and the grace period that protects you.Non-obviousness (§103)The bar most rejections cite: the Graham factors and KSR. - 03
When should you file — and which application?
The US is first-to-file: the date you file can decide who owns the invention. But filing too early — before you know the idea has legs — burns the clock and the budget. The provisionalprovisionalA simplified, lower-cost patent application that locks in a filing date for 12 months while the inventor refines or pitches.Read more → exists precisely to manage that tension.
Provisional vs non-provisionalBuy 12 months and a priority date for a fraction of the cost.Patent strategy for startupsThe founder's playbook: timing, what to protect, what investors want.The full patent processIdea to granted patent in eight stages, with every deadline.Patents for solo inventorsOn a budget — and how to dodge the invention-promotion scams. - 04
How do you patent your kind of invention?
The strategy bends to the technology. Software has to survive the Alice eligibility test; hardware needs the broadest mechanism claimsclaimsThe numbered statements at the end of a patent that legally define what the inventor owns.Read more →; biotech fights the §101 minefield. Pick the path that matches what you actually built.
How to patent softwareWhat survives Alice, and how to draft claims for it.How to patent hardwareWhy physical products need patents most — and the traps.How to patent biotechComposition-of-matter, §101, and publish-or-patent.How to patent a business methodThe §101-hardest category — and how fintech does it. - 05
What should it cost — and is it worth it?
A single US utility patentutility patentThe most common type of patent — covers functional inventions. 20-year term from filing.Read more → realistically runs five figures over its life, before you file a second one or go international. A strategy that ignores the cost curve stalls at the first maintenance fee. Model the spend against the return before you commit.
Patent cost estimatorFree: filing, attorney, and maintenance fees for your entity size.Patent ROI calculatorIs the monopoly worth more than the spend? Run the numbers.What a patent costsUSPTO fees, attorney rates, and the realistic all-in total.Attorney or DIY?When professional drafting pays for itself — and when it doesn't. - 06
How do you turn patents into leverage?
One patent is a fence post; a portfolio is a fence. Mature strategy is about building coverage that's expensive to design around, then extracting value — through licensing, valuation in a raise or sale, or deterrence — and defending it when challenged.
Building a patent portfolioFrom a single filing to a defensible thicket.Patent valuationWhat a patent is worth in a raise, an acquisition, or a lawsuit.Licensing a patentExclusive, non-exclusive, cross-license — how value gets extracted.Designing around a patentHow competitors avoid yours — so you can claim to block them.
Where to begin
Two ways in
If you don't yet know whether your idea is worth protecting, start with the patentability check. If you're already committed and want the founder-specific playbook, jump straight to the startup strategy guide.
FAQ
Patent strategy — common questions
What is patent strategy?
Patent strategy is the set of decisions about whether, what, when, and how much to patent — and how to use the resulting rights. It runs from the first question (is this even worth protecting, or better kept as a trade secrettrade secretConfidential business information that gains competitive value from being kept secret. Unlike patents, lasts as long as the secret is maintained.Read more →?) through scoping what to claimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more →, timing the filing, budgeting the spend, and ultimately building a portfolio you can licenselicensePermission from the patent owner to make, use, or sell the invention — usually in exchange for payment. Doesn't transfer ownership.Read more →, value, or enforce. A good strategy ties every filing to a business reason; a bad one files reflexively and runs out of money at the first maintenance fee.
When should a startup file its first patent?
Most startups file a provisional applicationprovisional applicationA simplified, lower-cost patent application that establishes a filing date. Must be converted within 12 months.Read more → as soon as the core invention is concrete enough to describe in detail — often before launch, because the US is first-to-file and any public disclosure or sale starts a clock. The provisionalprovisionalA simplified, lower-cost patent application that locks in a filing date for 12 months while the inventor refines or pitches.Read more → secures a priority datepriority dateThe earliest date used to compare the patent against prior art. Usually equals the filing date.Read more → for $65–$325 in USPTO fees (micro to large entity) and buys 12 months to validate the market before committing to the far more expensive non-provisional. The deeper playbook is in our startup patent strategy guide.
How much should a startup budget for patents?
A single US utility patentutility patentThe most common type of patent — covers functional inventions. 20-year term from filing.Read more → typically costs $15,000–$30,000 all-in over its life (attorney drafting, USPTO fees, and three maintenance feesmaintenance feesPeriodic fees the USPTO charges to keep a granted utility patent in force (3.5, 7.5, 11.5 years post-grant). Miss one and the patent expires early.Read more →), with complex fields like biotech running higher. A provisionalprovisionalA simplified, lower-cost patent application that locks in a filing date for 12 months while the inventor refines or pitches.Read more → alone can be under $3,000. Budget against the return, not the wish list — our patent cost estimator and ROI calculator model both sides before you commit.
Do you always need a patent?
No. A patent requires public disclosure of how your invention works in exchange for a 20-year monopoly. If your edge is something competitors can't easily reverse-engineer — a process, a recipe, an algorithm kept server-side — a trade secrettrade secretConfidential business information that gains competitive value from being kept secret. Unlike patents, lasts as long as the secret is maintained.Read more → can protect it indefinitely and for free. The right instrument depends on detectability and lifespan; our patent vs trade secret guide walks through the trade-off.
What do investors look for in a patent strategy?
Investors look for filings that map to the company's actual moat — protection on the specific mechanism that makes the product hard to copy — plus freedom to operate (you're not infringing someone else) and a credible plan for international coverage if the market is global. A thick portfolio of vague filings impresses no one; a few well-scoped patents on the core invention does.
This is plain-English education, not legal advice. The right strategy depends on your technology, market, and budget — the linked tools give specifics, but for a real plan work with a registered patent attorney. PatentBrief is not a law firm.