Patent Portfolio Strategy for a Company That Is Not Yet Big

A single patent is a document. A portfolio is a position. The difference shows up the first time a competitor's engineers sit down to design around you, or an acquirer's counsel opens your IP schedule during diligence. One patent with one set of claims can usually be avoided by a competent engineering team in a few weeks. A family of filings that covers the core mechanism, the obvious substitutes, the manufacturing method, and the consumable takes far longer to route around — often long enough that copying stops being worth it. Yet most small companies never make a portfolio decision at all. They file when something feels novel, forget about it, and get a maintenance fee notice years later for a patent that no longer maps to anything they sell.

Decide what the portfolio is for before you file anything

Patents do three commercially distinct jobs, and a filing that is good at one may be useless at another. Being explicit about which job you are buying keeps the budget honest.

• Exclusion. Stopping a specific competitor from selling a specific thing. This demands broad, enforceable, and detectable claims on the feature customers actually pay for.
• Deterrence and leverage. Making a copy expensive or risky enough that a rational competitor picks a different fight, or giving you something to trade when someone asserts a patent against you. Breadth matters less; coverage across the product's several subsystems matters more.
• Asset value. Satisfying investors, licensees, and acquirers that the technology is owned and defensible. Here consistency of ownership and clean chain of title matter as much as claim scope — the details are in what investors check in your intellectual property.

Map the invention space, not just the invention

Before drafting, sit down with your engineers and write out every place value hides in the product. A useful pass looks like this:

1. The core function. What the product does that nothing else does. This is the anchor filing.
2. Alternate embodiments. The three or four other ways a competent engineer could achieve the same result — a different actuator, a different sensing principle, a different geometry. If your specification does not describe them, a competitor can practice them freely.
3. The manufacturing method. Sometimes the process is the real barrier, and patenting a manufacturing process protects something a product claim cannot reach. Be careful: process claims are hard to police inside someone else's factory.
4. The consumable or accessory. Cartridges, tips, blades, filters, and mounts often carry the recurring revenue and are much easier to detect in the market than an internal algorithm.
5. The appearance. A design patent on the housing or a distinctive surface is fast and cheap relative to a utility filing, and it catches the direct visual knockoffs that utility claims often miss.
6. What should stay secret. Anything invisible in the shipped product and impossible to reverse engineer may be worth more unpublished; weigh it using trade secret vs patent.

Sequence the filings against your development calendar

Portfolio strategy is largely a scheduling problem. Filing everything on day one wastes money on claims describing a design you will abandon; filing nothing until the product is finished risks losing priority to someone else. The workable rhythm for most hardware companies:

The continuation habit is the cheapest strategic move on that list and the one small companies skip most often. As long as one application in a family stays pending, you can draft new claims aimed at a competitor's actual product, supported by the specification you wrote years earlier. Once every application in the family issues, that option is gone permanently.

Budget for the whole life, not the filing

The filing invoice is the small part. A single US utility family typically runs into five figures across drafting, prosecution, and issuance, and our breakdown of what a patent costs in the US gives realistic ranges. Foreign filing multiplies it: translations, local agents, and separate examination in each country, which is why choosing which countries to patent in deserves more thought than picking the biggest economies. If you go the international route, the cost structure of a PCT application is the first thing to model.

Then comes the tail. US utility patents carry maintenance fees at 3.5, 7.5, and 11.5 years after grant, each larger than the last, and most other countries charge annuities every single year. A portfolio of six families across four countries can quietly reach a five-figure annual bill before you have licensed anything. Build that number into your operating plan from the start, at entity-discounted rates if you qualify, and revisit it every year.

Prune deliberately

Portfolios rot. Products change, markets close, and a patent covering a feature you removed two generations ago is a line item with no return. Run a review once a year and put every family in one of three buckets: keep because it reads on something you sell or something a competitor sells, convert because it is worth more licensed or sold than held, or drop because nothing in your roadmap or the market touches it. Letting a patent lapse feels like failure; paying eleven more years of fees on a dead one is worse. If a family lands in the middle bucket, get it valued before you negotiate, and read up on how licensing deals are structured before anyone sends you a term sheet.

Fit the portfolio to the rest of your IP

Patents are one instrument among several. Trademarks protect the name customers ask for and never expire as long as you use and renew them. Copyright covers firmware source, documentation, and marketing assets automatically. Trade secrets cover the process knowledge you never publish. Contracts — employment agreements, contractor assignments, supplier terms — are what actually keep ownership in your company's name, and a gap there can undo an entire portfolio. A portfolio plan that ignores those contracts is building on sand.

The test of a portfolio strategy is simple: could you explain, in one page, what each family covers, which competitor it is aimed at, and what you would do if that competitor launched tomorrow? If you cannot, you have a collection of filings rather than a strategy.

Projects House is a product development engineering firm, not a law firm. We help companies map where the defensible engineering actually sits in a product, produce the technical disclosure, drawings, and test data a patent attorney needs to draft strong claims, and design around competitor patents when that is the cheaper path. Filing and prosecution belong with a registered patent attorney. To talk through the engineering side of your IP position, use the contact form.