Designing around a patent means changing your product so it no longer falls within the scope of someone else's claims, while still delivering the function your customer wants. It is entirely legal, it is standard practice across every hardware industry, and it is one of the intended effects of the patent system: publishing an invention in exchange for a limited monopoly invites everyone else to invent a different way. The work is precise, though, because infringement is decided on claim language, not on how similar two products look.

Projects House is an engineering firm, not a law firm. This article is educational only and is not legal advice. Any real design-around should be reviewed by a licensed patent attorney.

Why this is a legitimate engineering activity

A patent does not grant ownership of a problem or of a market. It grants the right to exclude others from practicing what the claims cover, for a limited time, in the countries where it was granted. Everything outside the claims is open. A patent also teaches: the specification is a published technical disclosure you are free to read, learn from, and improve upon — which is precisely why competitors read each other's filings. Working around a blocking patent is a normal design constraint, in the same category as a cost target or a size limit.

It all starts with the claims, not the description

The single most common mistake is reading the abstract, the drawings, or the description and concluding "this covers our idea." The description is usually far broader than the protection. Only the numbered claims define the boundary.

How claims work in practice:

  • Every element must be present. An independent claim is a list of elements joined by "comprising." To infringe literally, your product must include every element in that list. Omit one required element and the literal claim is not met.
  • Independent claims are the fence; dependent claims narrow it. Focus first on the broadest independent claims — if you are clear of those, the dependents follow.
  • Claim terms have specific meanings. Words are interpreted in light of the specification and the prosecution record, so a term that looks generic may be much narrower than everyday usage suggests.
  • Equivalents matter. Substituting a trivially interchangeable part for a claimed element may still be treated as infringement under the doctrine of equivalents. A cosmetic swap is not a design-around.
  • Prosecution history constrains the owner. If the applicant narrowed a claim during examination to get it granted, they generally cannot reclaim that territory later. Reading the file history often reveals where the real, defensible boundary sits.

Practical starting point: our guide to reading a patent and the walkthrough in patent claims explained cover the mechanics. Before any of it, check whether the patent is even a live obstacle — expired, lapsed for unpaid maintenance fees, or never granted in your market means there is nothing to design around. That check is described in how to check if a patent is still in force.

Four engineering moves that work

1. Eliminate a claimed element

The cleanest approach. If the claim requires a spring, a sensor, a second chamber, or a controller performing a named step, and your design achieves the result without that element at all, the claim is not literally met. This is genuine invention, not evasion — and it often produces a simpler, cheaper product.

2. Change the mechanism, keep the function

Deliver the same user outcome by a physically different principle: a magnetic latch instead of a mechanical one, a capacitive measurement instead of an optical one, a compliant flexure instead of a hinge and spring, software processing instead of a hardware filter. The more the underlying physics differs, the further you are from the doctrine of equivalents.

3. Rearrange the architecture

Move a claimed function to a different location in the system — from the device to the cloud, from hardware to firmware, from one subassembly to another. Claims that recite structural relationships ("a X positioned between Y and Z") can be avoided by changing the relationship itself.

4. Operate outside a claimed range or condition

Some claims specify numeric ranges, materials, temperatures, or sequences. Working deliberately outside the stated range can put you clear — but only if the product still works well there, and only if the range is a genuine claim limitation rather than a stated preference.

The traps that catch teams

  • Looking at one patent. Families matter. The same invention may exist as several granted patents plus pending continuations in multiple countries, and a pending application can still change shape. Clearing one document clears one document.
  • Ignoring the doctrine of equivalents. If your substitution performs substantially the same function in substantially the same way to achieve substantially the same result, you may not be clear at all.
  • Designing around the product instead of the claims. The competitor's shipping product may be narrower or broader than what they actually own.
  • Country confusion. Patents are territorial. A US patent does not stop manufacture or sale in a country where no counterpart was granted — but importing into the US does infringe. Map protection against your actual markets and supply chain.
  • Documenting the wrong way. Internal notes that read like an admission are unhelpful later. This is a good reason to run the exercise with counsel involved rather than purely as an engineering side project.
  • Leaving it too late. A design-around after tooling is cut costs many times what the same change costs at the concept stage.

When a freedom-to-operate opinion is worth it

A formal opinion from patent counsel — a written analysis of whether your design falls within identified claims — costs real money and is worth it when the stakes justify it: significant tooling investment, a fundraise or acquisition where diligence will ask, a crowded field with active enforcers, or a known patent holder who has already made contact. For an early prototype with no revenue, a structured search plus an informal read usually suffices. The scope and process are covered in freedom-to-operate search. If a letter has already arrived, that is a different situation with its own timeline — see what to do about a patent cease and desist letter.

The upside: a design-around can be its own invention

A genuine alternative mechanism is, by definition, different from what was published. If it is also non-obvious and useful, it may be patentable in its own right — which turns a defensive exercise into an asset. Teams often find that the constraint forced a simpler architecture, fewer parts, or a lower cost than the original approach. That is the same logic behind patenting an improvement to an existing product: the blocked path is frequently not the best one.

Projects House does the engineering half of this work — taking a claim set as a design constraint and generating architectures that deliver the function by a different route, alongside your patent attorney. Get in touch through our contact form to discuss a design-around for your product.