I think part of the reason why patents are undervalued these days is the growing myth that you can avoid infringing a patent by changing your product by, say, 20% to 30% (what does a 20-30% product modification even mean?). That is not how much patent infringement works.
Let’s the set the record straight. There is no rule about any percentage modification that would avoid patent infringement. Infringement is not about how different your product is from the patent. In fact, your product may seem very different, yet still infringe. Infringement is all about the claims, and avoiding infringement is about omitting what’s in the claims (not adding or changing).
Utility patent infringement must be analyzed differently than design patent infringement.
How to analyze utility patent infringement
To analyze utility patent infringement, you must check each independent claim in the patent and see if each claim feature is found in your product. If an independent is not infringed, then logically a dependent cannot be infringed. In contrast, invalidating an independent claim does not necessarily invalidate a dependent claim.
Suppose a utility patent (not a design patent or a pending application) contains two independent claims: Independent Claim 1 includes a combination of features, or claim limitations, AB while Independent Claim 2 recites a combination of BC. Let’s assume you’re thinking of launching a product with features ABCD.
In the above example, your product would infringe both Claim 1 because it has A and B, and Claim 2 because it has C and D. Notice how you do not avoid infringing Claim 1 by adding features on top of what is claimed. Notice also that percentage changes in the product have nothing to do with the infringement analysis. You can add a limitless number of additional features that will not help you avoid infringement.
Let’s assume you redesign your product so that it contains features A and C only. In this case, you would avoid infringing both Claim 1 – because B is missing – and Claim 2 – because D is missing.
How to analyze design patent infringement
The claims of a design patent consist of the figures. Therefore, the standard for determining design patent infringement is whether an ordinary observer would consider the accused product to substantially the same as the patented design when the two designs are compared.
However, preexisting designs (aka “prior art“) must also be taken in consideration in this analysis. This basically means that the product of accused of infringement must include at least a novel feature of the design patent that makes the patented design unique.
Literal Infringement vs. Doctrine of Equivalents (DOE)
So far, we have been discussing literal infringement, but the analysis does not stop there. The doctrine of equivalents (DOE) must also be considered even if your product does not literally infringe. If a claim limitation is missing from your product (thus, no literal infringement), you still have to consider whether your product might include a feature that is equivalent to that missing claim limitation. The DOE can be a tricky gray area which deserves a separate discussion.