Patent Valuation: Why the Number Is Not the Price

A patent valuation tells you what an asset is worth under a set of assumptions. A price is what one specific buyer will pay for it on one specific day. For most patents, those two numbers are not in the same postcode, and the distance between them is where owners lose years.

This is not a criticism of valuation work. The methods are sound and there are good reasons to run them, from balance sheet treatment to litigation support. The problem starts when a report built to answer one question gets used to answer a different one. As a broker who has run thousands of these transactions puts it, owners regularly arrive holding valuation reports produced by firms that have never sold a patent, setting expectations far outside market constraints.

So: the methods first, then what the market actually pays, then the four things that close the gap.

The three patent valuation methods

Income based

The most common method when there is data to work with. It assumes the owner sells products practising its own patents and that those patents protect the sales from copycats, with the assumed contribution often set at 25 percent of revenue. Build a model from past sales and forecasts, add market share expansion, subtract obsolescence toward the end of the patent life, discount to present value.

The output is usually millions, or tens of millions, even for a small portfolio. It also depends on the company hitting its forecasts, on the patents still mattering ten years out, and on a chain of assumptions that seldom all hold. It measures value to the owner, and only if everything in the model comes true.

Cost based

What it would cost to acquire the same protection. The key move here is decoupling the patents from the technology they cover. If a technology took 10 million USD to develop and the patents cost roughly 100,000 to draft, file, prosecute and maintain, the cost-based valuation sits nearer 100,000 than anywhere near the development spend. It is rarely used on its own, except where the owner bought the patents and carries them at what was paid.

Market based

Comparable transactions. This is the truest measure of what an asset is worth today, because it is the only one that reflects a real counterparty. It is also the hardest to run, since the vast majority of patent transactions are never reported publicly, so comparables barely exist and each patent is unique in ways that move value sharply.

Buyers price on this one. If your report did not use it, your report is not describing your sale.

What the market data actually says

Here is the part most valuation content leaves out.

Start with liquidity. Over 90 percent of properties find buyers in the real estate secondary market, and more than 90 percent of patents do not transact at all. That is close to the inverse of the analogy everyone reaches for when they call a broker a realtor for patents.

Then the harder number. On a market basis, the value of roughly 99 percent of issued patents is simply zero, because an acquirer has no reason to take on maintenance and annuity costs for an asset nobody is infringing. A patent grants the right to exclude. If nobody needs excluding, there is nothing to buy.

When a patent does sell, the tiers are reasonably well established: 5,000 to 50,000 USD for narrow claims or assets near expiry, 100,000 to 500,000 for solid claims that map to products already on the market with eight to ten years of life remaining, and above 500,000 for foundational technology in fields like semiconductors or wireless. Asking prices for issued US patents with demonstrable commercial use cluster at 200,000 to 350,000. One analysis of more than 43,000 telecom and IT transactions found an average near 484,000 per document, though averages in this market are pulled upward by a small number of very large deals.

And underneath all of it sits validity risk: recent studies show more than half of the patents in circulation would be invalidated if challenged. A buyer prices that in. Your report probably did not.

What actually moves a patent from valuation to price

Four things, and none of them is the report.

An identifiable infringer. An uninfringed patent has little or no exclusion value until the industry catches up with the invention. The question is not whether the technology is good. It is whether someone is already selling a product that reads on your claims, in volumes you can build a damages model around.

Claim breadth. Broad, well-drafted independent claims covering many implementations are worth far more than claims tied to one specific configuration. The difference between a claim covering any device that performs a function and one covering a device manufactured by a particular process to perform it can be a factor of ten.

Remaining life. Patents run 20 years from the earliest effective filing date. Twelve years left gives a buyer over a decade of licensing or protection. Three years left is worth a fraction of that.

Validity risk. Family size helps, an open family helps more, and the presence or absence of relevant prior art moves the number harder than any assumption in a discounted cash flow model.

How we approach it, and what we charge for

Our own answer to this problem is a validation gate that runs before an asset goes anywhere near a buyer. It uses physics-grade claim reconstruction, detectability analysis and validity probability, benchmarked at 0.76 percent mean absolute percentage error against real-world outcomes. Detectability matters more than it sounds: a claim you cannot prove is being practised is a claim you cannot price.

The incentive structure matters too, and it is worth being direct about it. Our brokerage is success-based: the seller pays commission on completion, with a minimum fee of EUR 5,000 per completed transaction, and no sale means no fee. A firm paid to produce a valuation report is paid whether or not the asset ever sells. A firm paid on completion is not. Neither model is dishonest, but they point at different numbers, and you should know which one you are buying.

Two limits, stated plainly: EX IX is not a firm of patent attorneys, and a novelty search does not guarantee grant or patentability. Drafting work is filed by an independent licensed attorney of record.

A realistic sequence before you commission anything

1. Check the remaining life from the earliest effective filing date. Under five years changes every other answer.
2. Read your own independent claims and ask honestly how easily a competitor designs around them.
3. Search for products already on the market that read on those claims, and name the companies selling them.
4. Check the family: how many members, and is it still open?
5. Assess prior-art risk before a buyer does it for you.
6. Only then choose between sale, licensing and letting it lapse.

Most of that costs time rather than money, and it tells you more about the price than a report will. If step one or step three comes back badly, the live question is not valuation at all: it becomes patent renewal fees and the renew, abandon or monetise decision. And if you are earlier than that, still deciding whether to file, the arithmetic starts with what it costs to file a patent in the first place.

FAQ

How much is my patent worth? If nobody is practising the claims, most likely nothing on the open market. When patents do sell, narrow or near-expiry assets go for 5,000 to 50,000 USD, and solid claims mapping to marketed products for 100,000 to 500,000.

Why is my valuation report so much higher than any offer? Income-method reports model value to you, assuming your forecasts hold and the patent stays relevant for a decade. A buyer takes on all the enforcement risk and cost, so it prices on comparable transactions instead. The two methods are answering different questions.

Which patent valuation method do buyers use? The market approach, almost always. Its weakness is that most patent transactions are private, so a credible market view depends on someone who sees deal flow rather than on public data.

What percentage of patents actually sell? Fewer than one in ten. More than 90 percent never transact on the secondary market.

Can a high valuation hurt a sale? Yes, and it commonly does. An owner anchored to an income-method number will decline offers that reflect the real market, and portfolios sit unsold for years while the remaining life, which is the one input that only moves downward, runs out.