If you are reading this, there is probably a document with a patent number on it somewhere you can see it, and you have read the same paragraph four times.
That is a reasonable way to feel about it. The document was written in a register meant for lawyers and patent examiners, not for the person who has to decide what to do about it by Friday.
Almost none of it answers the three questions you actually have, which are whether this is serious, what it is going to cost, and whether you did something wrong.
What just happened
Two different pieces of paper are usually what starts this.
The first is a letter. Someone's counsel writes to say that a patent exists, that your product appears to fall inside it, and that they would like to hear from you. A letter is not a lawsuit. It is also not nothing. It opens a range of outcomes that runs from a license negotiation through to litigation, and it starts real obligations on the day it arrives, the most immediate being that you stop letting documents disappear.
The second is a complaint, which means a case has been filed. That is louder. A complaint is one side's allegation, written by the side that wants to win. It is not a finding of anything. Companies of every size receive them, including companies that turn out to have done nothing wrong.
In federal court a response to a complaint is generally due twenty one days after service, though waivers and extensions move that date, and it is specific to your court and your case. Confirming the real date, from the real document, is the first thing to do rather than the second.
The picture in your head right now is probably a courtroom. Most patent disputes end in a negotiated resolution rather than a trial verdict. That is not a promise about yours. It is a correction to the scale of the thing you are imagining.
What the document actually says
Patents are less mysterious than they look, and the misunderstanding is nearly always in the same place.
A patent is a time limited right to stop other people from making, using, or selling what it describes. It is not permission to build the thing yourself, and it is not a prize for having had an idea first. It is a fence. The only part of the document that draws that fence is the set of numbered sentences at the very end, called the claims. Everything before them, the drawings, the background, the long description of the field, is scenery. Whether a product infringes turns almost entirely on whether it does what one of those numbered sentences describes.
So read the claims first, slowly, one clause at a time, asking of each clause whether your product actually has that thing. If a single element is missing, the analysis changes completely.
The fact that surprises people most is usually the source of the anger in the room. Copying is not required. If a product does what a claim describes, it can infringe even though the company invented it independently, in a different building, in a different decade, and had never heard of the patent or the person who owns it. You may not have taken anything from anyone. You may still have a problem. Both of those are true at once, and holding them together is most of the work of the first week.
The part that quietly decides cases is not in the patent at all. Every patent carries a public record of the argument that produced it, the back and forth between the applicant's lawyers and the patent office. It is called the prosecution history, or the file wrapper. To get a patent granted, applicants very often narrow what they are claiming. Those narrowings are permanent, they are public, and they limit what the patent can be argued to mean years afterward. Reading all of it is slow, unglamorous work that a great many people skip. It is also where a good deal of leverage lives.
None of that is beyond you. It is a fence, a set of sentences that draw it, and a public record of the arguments that moved it. Anyone who tells you the whole thing is too technical to explain is telling you about themselves.
What is worth doing this week
Nothing you do in the next seven days will win this or lose it. A few things can make it materially worse, and all of them are avoidable.
Confirm the deadline in writing, taken from the document itself rather than from memory or from whoever forwarded it to you. Then suspend automatic deletion. Most companies quietly destroy email and files on a schedule, and that schedule has to stop for anything touching the product being accused. Courts treat this seriously, and it is one of the few early mistakes that can hurt a company independently of whether it infringed at all.
Call your insurance broker in the same week. Some business policies cover intellectual property defense and some do not, and late notice can forfeit coverage you have already been paying for. Asking costs nothing.
Keep the speculation off email. The instinct is to circulate the document with a paragraph of opinion attached, and written guesses about whether you infringe, made by people who have not read the claims, have a way of reappearing later in a setting you would not have chosen. Have that conversation out loud, in a room.
Then gather the dull material: when the product first shipped, what has changed in it since, who built it, and anything already on hand about others working in the same area. Whoever you hire will ask for all of it in the first conversation, and having it ready is worth a week of calendar.
Choosing the person who reads it
The hardest decision of the first month is not a legal one. It is deciding who to trust with something you built.
A patent case is a technical argument that eventually has to be made comprehensible to people who are not technical: a judge, and possibly a jury of people who have never seen your product and never will outside a courtroom. If the lawyer cannot understand the technology at the level it was actually built, that translation has to happen twice, first from your engineers to a hired expert and then from the expert to the court, and something is lost at both joints.
Which means the useful questions in an interview are not about credentials. Ask whether they can understand your technology themselves, or whether they will depend on an expert to translate it for them. Ask whether they will read the full prosecution history personally, or hand it to someone else. Ask how they would explain your product to a judge who has never seen one.
Ask what they would decline to spend your money on, and listen for whether the answer is specific. Ask how they intend to create pressure without letting the cost run away, since those two goals genuinely pull against each other and anyone who says otherwise is selling. And ask them to explain the matter back to you in language your board can use. If they cannot do that in the first meeting, when they are trying to win your business, it will not improve later.
Interview more than one. The comparison teaches you more about the case than any single conversation will, and the differences between the answers are usually the most honest information available to you.
Read these recommendations in full on LinkedIn, where they were written.
Being early
Being early to this is worth more than being right about it. The people who come out of a patent dispute in decent shape are usually not the ones who guessed correctly in the first week. They are the ones who confirmed the date, stopped the deletion, made the insurance call, and put a competent person on it before the calendar started making decisions on their behalf.
If a conversation today would be useful, the phone number is answered directly.
Jesse J. Camacho