Self-created intangible assets — the software, patents, trademarks, trade secrets, customer lists, and brand value your business builds internally rather than buys — get harsher treatment under both accounting rules and tax law than intangibles you acquire from someone else. Most of the money you spend creating them is expensed as you go rather than recorded as an asset. Since 2022, research and development costs must be capitalized and amortized over five years (or fifteen for foreign research) for tax purposes. And when you sell one, the gain is generally taxed as ordinary income, not at the lower capital gains rate. Understanding these rules matters at tax time, during fundraising, and especially when selling the company, because the gap between what these assets are actually worth and what they show on the books catches most owners off guard.
What Counts as Self-Created
The defining feature is origin. A self-created intangible comes from your own employees, spending, and effort rather than from a purchase order. Common examples:
- Internally developed software written by your engineers for internal use or for sale.
- Patents and inventions produced through your R&D activities.
- Trademarks, logos, and trade dress your team designed and registered.
- Trade secrets: recipes, algorithms, manufacturing methods, pricing models kept confidential.
- Customer lists and relationships built through direct marketing and sales.
- Domain names you registered directly.
Many of these never appear on a standard balance sheet even when they carry most of the company’s competitive value. That invisibility has consequences: investors undervalue the business, owners underinsure it, and acquirers don’t know what they’re paying for until a proper valuation is run.
How GAAP Treats Self-Created Intangibles
Generally Accepted Accounting Principles take a conservative approach. The core rule is blunt: most costs of internally developing an intangible get expensed immediately rather than recorded as a long-term asset. ASC 350-30 states that costs of internally developing, maintaining, or restoring intangible assets are expensed as incurred. ASC 730 does the same for research and development: wages for R&D staff, lab supplies, and testing costs all flow straight to the income statement. The only carve-out is for materials, equipment, or facilities acquired for R&D that have a clear alternative future use.
The practical effect is that R&D-intensive companies often look less valuable on paper than they actually are. Investors looking for future growth need to read the R&D line on the income statement, not the asset side of the balance sheet.
The Internal-Use Software Exception
ASC 350-40 is the major exception. It splits software development into three phases, each treated differently:
- Preliminary project stage — scoping the requirements, evaluating vendors, deciding whether to proceed. All costs are expensed.
- Application development stage — once management authorizes the project and coding begins, direct costs like programming, testing, and installation are capitalized. Training and general overhead stay expensed even during this phase.
- Post-implementation stage — the software is live. Maintenance and training are expensed. Upgrades that add genuine new functionality can be capitalized if they meet the development-stage criteria.
Capitalized software gets amortized over its useful life, typically three to five years. The transition from preliminary to development phase is what auditors probe hardest: you need documented management authorization and a clear project plan before you start capitalizing anything.
Capitalized software also has to be tested for impairment when circumstances suggest the carrying value may not be recoverable. Triggers include the software no longer being expected to provide meaningful service, a significant change in how it’s used, a major planned modification, or development costs running well over budget. When any trigger fires, you compare carrying amount to expected future cash flows and write the asset down to fair value if the cash flows fall short. Abandoned projects usually mean a full write-off of whatever was capitalized.
How Federal Tax Law Treats Them
The IRS treats self-created intangibles very differently from acquired ones, and the rules got harsher starting in 2022. Three statutes drive the outcomes: Section 197 (amortization of acquired intangibles), Section 174 (R&D expenditures), and Section 41 (the R&D tax credit).
Section 197 Mostly Excludes What You Built Yourself
When a business buys intangibles like goodwill, customer lists, or non-compete agreements, those acquired intangibles are amortized over 15 years under Section 197. Self-created intangibles are mostly excluded from this schedule. The statute keeps three narrow categories inside Section 197 even when self-created: government-granted licenses or permits, covenants not to compete entered in connection with an acquisition, and franchises, trademarks, and trade names.1Office of the Law Revision Counsel. 26 USC 197 – Amortization of Goodwill and Certain Other Intangibles
So self-created trademarks stay subject to Section 197, but in practice most of the money spent building a brand — advertising, marketing campaigns, design work — gets expensed as ordinary business costs. What actually gets capitalized are the legal fees and government registration costs tied to securing the trademark.2Internal Revenue Service. Chief Counsel Advice 201536006 Those go on the 15-year schedule. The brand’s underlying market value, which may be enormous, has no tax basis and generates no amortization deductions.
Section 174 Now Forces Capitalization of R&D
Before 2022, businesses could deduct research and experimental expenditures in the year they were incurred. The Tax Cuts and Jobs Act changed that. For taxable years beginning after December 31, 2021, Section 174 requires that R&D expenditures be capitalized and amortized over five years for domestic research or fifteen years for foreign research, using a midpoint-of-year convention.3Office of the Law Revision Counsel. 26 USC 174 – Amortization of Research and Experimental Expenditures
The midpoint convention starts your amortization clock at the middle of the tax year you paid the expense — typically July 1 for calendar-year filers — regardless of when in the year the money went out. The cash flow hit is real. A company spending $1 million on domestic research can no longer deduct the full amount that year; it writes off roughly $100,000 in year one (half a year of a five-year schedule) and $200,000 per year for the following four years.
The R&D Credit Under Section 41
The forced amortization under Section 174 makes the R&D tax credit under Section 41 more valuable than before. This credit reduces your tax liability dollar for dollar based on qualified research expenses, which include wages paid to employees for qualified research, supplies consumed in the research, and 65% of amounts paid to outside contractors for qualified research.4Office of the Law Revision Counsel. 26 USC 41 – Credit for Increasing Research Activities
Qualifying research has to be technological in nature, aimed at developing or improving a product, process, software, formula, or technique, and substantially all of the activity has to involve evaluating alternatives where the method or design was uncertain at the outset.5Office of the Law Revision Counsel. 26 USC 41 – Credit for Increasing Research Activities The IRS has flagged this credit as prone to abuse, particularly when businesses claim it for routine day-to-day activities. Research conducted after commercial production begins, customer-funded research, and activities with no genuine uncertainty do not qualify.
Startups and small businesses with gross receipts under $5 million can elect to apply up to $500,000 of the credit against their share of payroll taxes instead of income taxes, an option expanded by the Inflation Reduction Act of 2022.6Internal Revenue Service. Research Credit Against Payroll Tax for Small Businesses The election is made on Form 6765 with the original income tax return (including extensions), and the payroll tax offset takes effect the first calendar quarter after the return is filed.
Selling a Self-Created Intangible: Ordinary Income
This is where many owners get an unpleasant surprise. Under Section 1221, self-created patents, inventions, secret formulas, copyrights, and similar property are explicitly excluded from the definition of a “capital asset” when held by the person whose efforts created them.7Office of the Law Revision Counsel. 26 USC 1221 – Capital Asset Defined The gain from selling one is taxed as ordinary income, not at the lower long-term capital gains rate. Holding period doesn’t help. Develop a proprietary process in-house, sell it years later for a large profit, and the entire gain is taxed at your ordinary income rate. The single statutory exception covers musical compositions and copyrights in musical works, where the creator can elect capital gains treatment.
Protecting the Asset So You Don’t Lose It
Building an intangible is only half the job. If you don’t protect it properly, you can lose the legal rights entirely, sometimes permanently.
Patents and the One-Year Clock
Federal patent law gives inventors a one-year grace period after the first public disclosure, sale, offer for sale, or publication describing the invention. Miss that window and you’re permanently barred from obtaining a patent.8Office of the Law Revision Counsel. 35 USC 102 – Conditions for Patentability The deadline catches businesses that demonstrate a product at a trade show or start selling before the patent attorney has filed. The clock runs from the earliest public event, and no amount of money buys an extension.
The Contractor Copyright Trap
When an employee creates a work within the scope of their employment, federal copyright law automatically assigns ownership to the employer as a work made for hire.9Office of the Law Revision Counsel. 17 USC 101 – Definitions No written agreement needed. This covers software code, marketing materials, technical documentation, and anything else copyrightable your employees produce as part of the job.
Independent contractors are a different story. A contractor’s work qualifies as work for hire only if it falls into one of a handful of specific categories (contributions to collective works, translations, compilations, instructional texts, and a few others) and both parties have signed a written agreement stating the work is a work for hire.9Office of the Law Revision Counsel. 17 USC 101 – Definitions Without that signed agreement, the contractor may own the copyright to code, designs, or content they created for you, even if you paid for every hour. This is one of the most common and expensive intellectual property mistakes small businesses make.
Trade Secrets and Reasonable Measures
Trade secret protection under both the federal Defend Trade Secrets Act and the Uniform Trade Secrets Act (adopted in most states) requires that the information derive economic value from being kept secret and that the owner take reasonable measures to maintain that secrecy. What counts as reasonable depends on context, but courts consistently look for concrete steps: non-disclosure agreements that specifically identify the protected information, password protection and access on a need-to-know basis, physical security for proprietary areas, clear communication to employees about what is confidential, and training on protection procedures.
Notice is the point courts return to. Even robust security can fail the legal test if you never told the people with access that the information was confidential. Onboarding agreements, exit interviews, and specific labeling of confidential documents all strengthen the position. Vague or boilerplate NDAs that don’t identify what’s actually being protected are frequently ruled insufficient.
Putting a Value on Them
Pricing an asset that has never been sold requires specialized appraisal. The question comes up during business sales, litigation, tax disputes, and financial reporting for acquisitions. Three approaches dominate.
The cost approach asks what it would take to recreate the asset from scratch today: labor, materials, overhead, time. It provides a useful floor but tends to understate value because it ignores competitive advantage and revenue generation. A customer list that took $200,000 in employee time to build might generate $2 million a year in revenue; the cost approach captures only the first number.
The market approach compares your asset to similar intangibles recently sold in arm’s-length transactions. In theory it’s clean; in practice, self-created intangibles like trade secrets and proprietary algorithms are unique by nature, and comparable sales data is either nonexistent or not public. The method works best for assets with active licensing markets, such as certain technology patents where royalty rates are reported.
The income approach projects the future cash flows attributable to the intangible and discounts them to present value. It is the most commonly used method for high-value intangibles because it captures what the asset is actually worth to the business. Analysts project incremental revenue or cost savings, subtract the contributions of other assets needed to support those cash flows, apply appropriate tax rates, and discount at a rate reflecting the risk. A common variation is the relief-from-royalty method, used often for trademarks and patents, which estimates what the business would pay in licensing fees if it didn’t own the asset. Another, the multi-period excess earnings method, works best when the intangible is the primary asset driving the business.
Professional appraisers typically combine methods and weight the results based on data quality and asset type. The conclusions have to hold up under scrutiny from buyers, auditors, or the IRS. Getting it wrong in either direction creates problems: overvaluation triggers tax disputes, and undervaluation leaves money on the table during a sale.