Texas Instruments designs and makes semiconductors, the tiny chips inside almost every piece of electronic equipment on earth. It earns money each time a customer purchases one of its more than 80,000 chip products, which range from power management chips that control how a battery discharges to the microcontrollers that act as the brain inside industrial machines and cars. Its two main businesses are Analog, which converts real-world signals like sound and temperature into digital data, and Embedded Processing, which handles specific computing tasks. Together those two segments generated $16.7 billion of the company's $17.68 billion in 2025 revenue. The diagram below traces where the money goes.
Five years of financial data tell a story with three distinct chapters. First came a peak. Then came a painful slide. Now there are early signs of recovery, but the recovery is not yet complete and some important measures are still moving in the wrong direction.
Revenue hit $20.0 billion in 2022. Then the semiconductor cycle turned downward. Customers had stockpiled too many chips during the supply shortages of the pandemic era, and when they started drawing down those stockpiles, they stopped ordering new ones. By 2024, revenue had fallen to $15.6 billion. The 2025 number of $17.68 billion shows a real bounce back, but the company's own filing notes the recovery has been slower than past upturns, likely because of broader economic uncertainty.
Gross margin is the share of each dollar of revenue the company keeps after paying to make its chips. That number has been falling steadily. In 2022, Texas Instruments kept about 69 cents of every dollar of revenue as gross profit. By 2025, that figure had dropped to about 57 cents. The company points to high manufacturing costs tied to its large expansion of new chip factories as the main reason. Those factories are not yet running at full speed, so their fixed costs weigh heavily on each chip produced.
Free cash flow tells a similar story. This is the cash left over after the company pays for running its factories and building new ones. It was $6.3 billion in 2021. It fell to $1.3 billion in 2023, recovered slightly to $1.5 billion in 2024, and reached $2.94 billion in 2025. The company's own documents state it is nearing the end of a six-year cycle of elevated spending on new factories, and it expects capital spending to drop to roughly $2 billion to $3 billion in 2026, compared with $4.55 billion in 2025. If that happens, free cash flow could rise sharply even without much revenue growth.
That rising debt load is worth watching. Net debt grew from $3.1 billion in 2021 to $10.8 billion by the end of 2025. The company funded its factory buildout partly by issuing more long-term bonds. On top of that, Texas Instruments announced in February 2026 that it agreed to acquire Silicon Labs for approximately $7.5 billion in cash, funded by a combination of cash on hand and new debt. That deal has not yet closed and is expected to do so in the first half of 2027, pending regulatory approval.
The risks facing the company are specific and documented, not just generic warnings. The most serious involves China. About 50% of Texas Instruments' products flow to China, and about 20% of revenue comes directly from Chinese customers. The United States and China have imposed tariffs and export controls on semiconductors. If those restrictions tighten further, Texas Instruments could find itself unable to sell into its largest market or unable to get materials it needs from suppliers there. A second major risk is that the company owns most of its own manufacturing, which means most of its costs are fixed. If customer demand falls again before the new factories are fully loaded, those fixed costs will crush margins just as they did in 2023 and 2024.
There are also supply chain risks the company itself flags. It depends on a small number of global suppliers for specialized manufacturing materials and equipment. Geopolitical tensions have already caused some suppliers to delay shipments and raise prices. Natural disasters at any of its key manufacturing sites could slow production with little ability to quickly shift output elsewhere.
The whole story of the next several years hinges on one thing: whether the new factories fill up with orders fast enough to justify their cost. The company itself measures its success by growth in free cash flow per share over the long term. That metric went from healthy to compressed during the buildout. The question now is whether the ramp-up of new capacity arrives before another market downturn does.