IP Valuation Beyond Patents: Valuing Trade Secrets, Copyrights, and Trademarks

IP valuation is often treated as synonymous with patent valuation, because patents are the most commercially active asset in most university IP portfolios and the only IP type most technology transfer offices (TTOs) have a documented methodology for. But a research institution’s IP portfolio is rarely patents alone. Trade secrets protect the know-how, protocols, and unpublished methods behind a patented invention; copyright covers the research software, datasets, and courseware a lab produces; and trademark increasingly matters the moment a lab spins out a company that needs a defensible brand. Each of these asset types resists the standard patent-valuation toolkit for a specific, structural reason — and each has its own accepted valuation approach that a TTO needs when the question isn’t “how do we value our patents” but “how do we value everything else.”

See CASRAI’s Patent Valuation Methods guide for the income, market, and cost approaches applied specifically to patents. This guide covers the same three-approach framework — income, market, and cost — but applied to the three IP types that framework doesn’t transfer cleanly to: trade secrets, copyright, and trademark.

Why non-patent IP resists the same valuation shortcuts

Patent valuation has one structural advantage the other three IP types lack: a patent’s claims are public. An appraiser can read the claims, search for licensing comparables in the same technology class, and benchmark against a visible universe of similar assets. That public-disclosure requirement is exactly what a trade secret does not have — its entire legal protection depends on staying undisclosed, which means there is no claim set to benchmark and no public license registry to search. Copyright in research software or courseware has the opposite problem: registration is trivial and near-automatic, so the existence of a copyright says almost nothing about value — unlike a patent claim, a copyright registration doesn’t describe a bounded, examined scope of protection. And a trademark tied to an early-stage spinout has no operating history to draw comparables from at all; its value is almost entirely prospective.

Each of these gaps has a standard workaround in valuation practice, adapted from the same three-approach structure (income, market, cost) used across asset appraisal generally and for patents specifically — but the adaptation differs meaningfully by IP type, which is why treating “IP valuation” as one undifferentiated exercise produces bad estimates.

Valuing trade secrets

A trade secret’s value estimate has to substitute for the market comparable a patent search would otherwise supply. Three approaches are used, roughly in order of how often TTOs can actually apply them:

• Cost approach (most tractable for early-stage secrets). Estimate what it would cost a competitor to independently develop the same know-how — R&D hours, materials, failed-attempt costs, and time-to-develop. This is the most defensible approach for process know-how, unpublished protocols, and formulation details that have no revenue history yet, because it doesn’t require forecasting a market that doesn’t exist.
• Income approach (once the secret is generating or clearly attributable to revenue). Capitalize the excess earnings the secret produces relative to what the same product or process would earn without it — for example, the margin advantage a proprietary manufacturing process gives over the best publicly known alternative. This requires isolating the secret’s specific contribution from everything else driving the product’s revenue, which is the hardest part of the calculation in practice.
• Market approach (rarely usable directly). Comparable trade-secret licenses are, almost by design, not public — the whole point of the underlying agreement is confidentiality. Where a market approach is used at all, it typically borrows royalty-rate ranges from adjacent patent or know-how licenses in the same field rather than from directly comparable trade-secret transactions.

Trade secret valuation has one additional, distinctly legal reference point that patents don’t: litigation damages. The federal Defend Trade Secrets Act sets out how a court values a misappropriated trade secret after the fact, and that statutory framework doubles as a useful mental model even outside litigation. Under 18 U.S.C. § 1836(b)(3)(B), a court may award damages for actual loss caused by the misappropriation plus any unjust enrichment not already captured in the actual-loss figure, or — “in lieu of damages measured by any other methods” — a reasonable royalty for the misappropriator’s unauthorized use. That statutory choice between (actual loss + unjust enrichment) and (reasonable royalty) tracks closely with the income-approach and market-approach logic above, which is part of why the three-approach framework and trade secret law converge on similar answers.

Valuing copyright in research software, materials, and courseware

Copyright protects the specific expression — the actual code, the actual written manual, the actual dataset structure — not the underlying idea, and it attaches automatically on creation without the examination process a patent goes through. That automatic, unexamined nature means a copyright’s existence tells a TTO almost nothing about value; the valuation work has to come entirely from the asset’s actual use and market position, not from the registration itself.

• Market approach. For research software with a plausible commercial licensee, comparable software-licensing royalty rates (by field and by exclusivity) are usually more available than trade-secret comparables, because software licensing terms surface more often in public disclosures, industry surveys, and prior university deals.
• Income approach. For software or datasets already generating license revenue or clear cost-savings for users, a discounted-cash-flow projection of that revenue stream is the standard method — the same logic used for patents that are already licensed and earning royalties.
• Cost approach. Development-cost estimation (programmer time, testing, validation) is common for early-stage research software with no revenue history yet, similar to the cost approach for an early-stage trade secret.

A practical complication specific to university-authored software: much of it is built with open-source components, which constrains what can actually be commercialized or exclusively licensed regardless of the copyright’s nominal value. See CASRAI’s Open Source Software Licensing in University Technology Transfer guide before valuing a specific piece of research software for licensing — the license terms on the open-source components it depends on often determine the addressable value more than the valuation method does.

Valuing trademarks and brand — the spinout-formation case

Trademark valuation matters to a TTO at a specific, predictable moment: when a university spinout forms and needs a brand, or when an existing university-affiliated mark (a lab name, a platform name, a program name) is licensed to or transferred into that new company. Unlike patents and copyrights, a trademark’s value is almost entirely about market recognition and goodwill rather than about the underlying invention — which makes the standard patent/copyright approaches a poor fit.

• Relief-from-royalty method (the standard approach for trademarks). Estimate the royalty rate the company would have to pay a third party to license an equivalent mark, then capitalize that hypothetical royalty stream over the mark’s expected useful life. This is the most commonly used trademark-valuation method in professional appraisal practice because it doesn’t require isolating brand-specific revenue from everything else the company does — it only requires a defensible royalty-rate benchmark for licensing marks in the relevant sector.
• Income approach (premium-profit variant). Where data allows it, compare the branded product’s margin or price premium against an unbranded or generically-branded equivalent, and capitalize that premium. This is data-intensive and rarely feasible for an early-stage spinout with no unbranded comparable to measure against.
• Cost approach (weakest fit, used mainly as a floor). Summing trademark registration, prosecution, and initial marketing costs understates true brand value almost every time, because it captures money spent, not market recognition achieved — it’s most useful as a sanity-check floor rather than a real estimate.

In practice, most university-spinout trademark situations don’t need a formal valuation exercise at all — the mark is transferred or licensed to the new company as part of the broader spinout equity and IP-assignment package, valued implicitly as part of that larger deal rather than priced on its own. A standalone trademark valuation becomes necessary mainly when the mark is being licensed separately from the underlying technology, or when a dispute or acquisition requires a specific number.

Where this connects back to patent valuation

The three-approach structure — income, market, cost — is the same skeleton across all four IP types, which is deliberate: it’s the general asset-appraisal framework adapted to intangibles rather than four unrelated methodologies. What changes by IP type is which approach is actually usable and what has to substitute for the market comparables a patent search would supply. For the patent-specific application of income, market, and cost — including where each approach breaks down for early-stage academic patents — see CASRAI’s Patent Valuation Methods guide. For how a valuation estimate (of any IP type) actually gets used once a licensing negotiation starts, see How University Tech Transfer Offices Evaluate and Price a License.

Frequently asked questions

Does a university need a formal valuation for every trade secret, copyright, or trademark it holds?

No. Most non-patent IP is valued informally, if at all, until a specific transaction — a license negotiation, a spinout equity allocation, an IP audit, or litigation — forces the question. Formal valuation is a transaction-driven exercise, not a portfolio-wide default.

Which approach should a TTO use first when a specific valuation question comes up?

Start with whichever approach has usable data. For most early-stage academic IP without revenue history, that means the cost approach for trade secrets and software, and the relief-from-royalty method for trademarks — the income and market approaches generally require data (revenue history, public comparables) that early-stage university IP doesn’t yet have.

How is this different from setting a royalty rate in a license agreement?

Valuation estimates what an asset is worth; a royalty rate is a negotiated price within an already-structured deal. The same three-approach logic informs both, but a valuation exercise happens before or independent of a specific negotiation, while a royalty rate is one negotiated term inside one specific agreement. See the patent valuation guide’s discussion of this distinction, which applies identically here.