12960 Linden Church Road
Clarksville, Maryland 21029

What Should You Write Down Before Talking to a Patent Attorney?

Oct 29, 2026

You built something—a piece of software that solves a problem your team used to handle by hand, or a process that runs faster than the approach you used before. At some point, someone says the word “patent,” and you have no idea what to do next.

Here is the good news: you do not need to learn patent law before that first conversation. You need to get clear on what your software actually does. That is the homework.

I have spent more than 20 years drafting patent claims for mechanical and electromechanical systems, including automated packaging equipment and automotive components. The questions that help explain a conveyor system are similar to the questions that matter for software: What problem does it address? How does it work? What is different about the approach? Where and how is it used?

If you can answer those questions clearly, you have done much of the useful preparation work.

Start With the Problem, Not the Code

Before describing what your software does, describe what it was built to fix.

Write down the specific problem your team faced before the solution existed. We needed better data processing is difficult to evaluate. A concrete explanation of how the old process operated, where it failed, or why it was too slow gives the attorney a much better starting point.

A clear problem statement can help identify whether the software reflects a potentially meaningful technological improvement or merely applies a familiar idea using conventional computer functions. But solving a real problem does not, by itself, establish patentability. Federal law recognizes new and useful processes, machines, manufactures, compositions of matter, and improvements as categories of potentially patentable subject matter, subject to the other requirements of the Patent Act. 35 USCS § 101.

Software claims also must do more than state an abstract idea and add generic instructions to carry it out. The eligibility analysis asks whether the claimed elements or their combination amount to significantly more than the ineligible concept itself. Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66, 73 (2012).

The problem statement matters because it helps frame the invention. It is not a substitute for analyzing eligibility, novelty, nonobviousness, and the other legal requirements.

Explain How It Works, Not Just What It Does

This is where many founders get stuck. It uses artificial intelligence to optimize scheduling tells an attorney very little. Artificial intelligence is not a magic phrase that makes something patentable, and it is not an automatic disqualifier. What matters is what sits underneath the description.

Start with the information the system receives. Describe the data or other inputs and where they come from.

Then explain how the system processes that information. If the order of operations matters, describe it. If the system uses an unusual data structure, validation method, decision rule, or interaction among components, explain that in ordinary language.

Finally, describe what the system produces and how that output is used. The practical result may help reveal whether the claimed process is tied to a particular implementation or simply states a desired outcome.

You do not necessarily need engineering diagrams for the first conversation, although they can help. You need a description detailed enough that someone unfamiliar with the codebase can understand the mechanics rather than only the marketing pitch.

Identify What Is Different From the Usual Approach

An attorney evaluating the idea will want to understand what people in the field did before your solution and how your approach differs.

Perhaps the system structures data in a new way. Perhaps it connects processes that were previously handled separately. Perhaps it uses an unconventional sequence that improves speed, accuracy, security, or resource use. Write down what you believe is different and why the difference matters, even if you do not know whether it is legally significant.

That information helps frame the patentability analysis, but there is no single mechanical test that resolves whether a process is patent eligible. The Supreme Court has explained, for example, that the machine-or-transformation test can provide a useful clue for process claims but is not the exclusive test. Bilski v. Kappos, 561 U.S. 593, 604 (2010).

The attorney will still need to consider what the patent claims might cover and how the invention compares with the relevant prior art. Your job before the first conversation is not to reach that legal conclusion. It is to explain the difference with enough precision that the analysis can begin.

Document Where and How It Is Used

Describe who uses the software, the setting in which it operates, and what happens as a result. An internal scheduling tool used by a logistics team presents different business considerations from a similar tool licensed to customers as a standalone product.

The deployment model does not automatically determine whether the technology is patentable. It can, however, affect filing strategy, disclosure concerns, ownership questions, and whether patent protection is the most practical form of protection.

Include any important timing information. Note whether the system has been demonstrated, offered for sale, released to users, described in a publication, or shared outside the company. Also identify who contributed to the underlying concept and whether employees, contractors, customers, or business partners participated in its development.

Please note that if the invention has already been disclosed or a public release is approaching, say so at the beginning of the conversation. Filing options can depend on what was disclosed, when it happened, who made the disclosure, and where protection may be sought.

What This Preparation Gives You

None of this requires legal training. It requires you to step back and explain your own product clearly, often for the first time in a while. Founders live so close to their software that they may stop noticing what is unusual about it.

Writing down the problem, the mechanism, the difference, and the practical use creates enough distance to make the first conversation more productive. It also helps separate the actual engineering contribution from the broad result the company hopes to achieve.

Whether a particular software or artificial-intelligence system is patentable depends on the claimed invention, the prior art, and the applicable legal requirements. But an attorney cannot evaluate a technical contribution that has not been explained clearly.

If you have been sitting on software that your team built out of necessity and never considered protecting, begin with a plain-language description. The part that is hardest to explain may be the part that needs the most attention—not necessarily because it is patentable, but because it may contain the technical detail that distinguishes the system from the idea behind it.

Garcia-Zamor serves as fractional general counsel for innovators, protecting business operations and intellectual property. Ruy Garcia-Zamor leads business-growth strategy as a registered United States patent attorney. Elliott Alderman handles intellectual-property matters, Claudia Castillo focuses on employment law, and registered patent attorney Amulya Annasamudram focuses on patents and intellectual property. Contact the firm through its website or at (410) 531-9853.