What a patent does, and what it does not

Founders tend to treat a patent as a single undifferentiated good: you have one, therefore you are protected, therefore you have a moat, therefore the business is worth more. Each step in that chain is doing more work than it can support.

What follows is a practical description of what the instrument actually is, written from the position of having taken inventions through to issued patents rather than from the position of advising on them. It is not legal advice, and nothing here is a statement about the scope or strength of any particular patent, including mine.

A right to exclude, not a right to practise

This is the distinction that surprises people most, and it is the one that matters commercially.

A patent gives its owner the right to stop others from making, using or selling what the claims cover. It does not grant the owner permission to build the thing. Those are genuinely separate questions. Your product can be covered by your own patent and still read onto somebody else's earlier one. Owning a patent is not clearance to operate, and treating it as clearance is how companies walk into problems they could have seen coming.

It does not enforce itself

A granted patent is a right that sits there until somebody acts on it. Acting on it means detecting infringement, establishing it, and being willing and able to pursue it — which costs real money and considerable time, and which the other side knows.

The practical consequence is that a patent's deterrent value is partly a function of your apparent capacity to enforce it. That is uncomfortable, and it is why patents held by a small company and patents held by a large one are not the same asset, even when the claims are identical.

The claims are the boundary, and boundaries get walked around

Only the claims matter. Not the title, not the abstract, not the drawings, not the description of how brilliant the invention is. A competitor's engineers will read the claims specifically to find the edge and step past it.

This leads to something counter-intuitive about drafting: a claim that describes your product exactly, in all its specificity, is often a weak claim, because the specificity is a map of how to avoid it. The useful question during drafting is not what did I build, it is what is the general mechanism, and what are the obvious variants somebody would reach for to get the same result.

What it does do

Having said all of that — the instrument is genuinely valuable, for reasons that are more commercial than legal.

1. 01It creates an asset that can move. A patent can be licensed, assigned, pledged or contributed to a venture. Know-how sitting in your head cannot. Converting capability into a transferable asset is most of the point.
2. 02It creates a dated public record. You disclosed this, on this date, in this detail. That has value in disputes about who got somewhere first, and it is a part of the bargain: disclosure in exchange for a limited right.
3. 03It changes the shape of a negotiation. A conversation about partnership or acquisition proceeds differently when there is a defined thing being discussed rather than a general claim of expertise. It makes the discussion concrete, which usually makes it faster.
4. 04It forces clarity. Drafting claims demands that you articulate what is actually novel, in language that survives examination. A surprising number of founders discover during this process that the thing they thought was their innovation is not, and that something adjacent is.

Sequencing

The common mistake is filing around the demo — the specific configuration that happens to exist on the day you decide to file. The more useful approach is filing around the roadmap: the mechanism you intend to be using in two years, described broadly enough to cover the versions you have not built yet.

That requires knowing your roadmap, which is why intellectual property strategy is a product-strategy activity that happens to involve attorneys, rather than a legal activity that happens to involve your product.

The second mistake is treating filing as the finish line. Issuance is where the commercial work starts, not where it concludes.

Trademark and patent ownership protect the brand and the invention. Commercial agreements create the revenue. The first does not produce the second on its own, and confusing them is how a portfolio becomes an expense rather than an asset.

The honest summary

A patent is a defensive right of defined scope, which costs money to obtain and money to enforce, and which becomes an asset when it is attached to a business doing something real.

It is not a moat. It is not a valuation. It is not a substitute for a product people want. It is a well-defined piece of property — and like any property, what it is worth depends almost entirely on what you do with it.