Quick answer: what is Teradyne?
Teradyne builds automated test equipment (ATE) for semiconductors, printed circuit boards and electronic systems. Its test systems verify that chips work correctly after manufacturing, before they are shipped to customers. Teradyne holds leading market share in semiconductor logic test (used to test compute, AI and communications chips) and in memory test (used to test DRAM and NAND flash). It also operates an industrial automation division through Universal Robots (collaborative robots) and MiR (autonomous mobile robots).
Teradyne is a member of the PHLX Semiconductor Sector Index (SOX). The central investor question is whether AI chip complexity can drive a new test-intensity cycle for semiconductor test equipment while industrial automation recovers.
Investor takeaway: Teradyne's AI test opportunity is real: larger, more complex chips take more time to test per unit, and AI accelerators are among the most complex chips ever produced. The challenge is the inherent cyclicality of semiconductor capex spending, mobile test concentration risk, and the need for industrial robotics to recover from its own investment cycle downturn.
Company at a glance
| Item | Overview |
|---|---|
| Company | Teradyne, Inc. |
| Ticker | TER |
| Sector | Information Technology |
| Industry | Semiconductor equipment (automated test) |
| Index membership | SOX (PHLX Semiconductor Sector Index), S&P 500 |
| Core customers | Semiconductor manufacturers (IDMs and OSATs), fabless chip designers purchasing test at OSATs, industrial manufacturers (robotics) |
| Primary economic drivers | Semiconductor capex cycles, AI chip complexity (test time per unit), smartphone production (mobile SoC test), DRAM/NAND capex (memory test), industrial automation investment |
| Key investor metrics | Revenue by segment (semiconductor test, system test, industrial automation), gross margin, order backlog, AI chip test as percent of semiconductor test revenue |
| Major peers | Advantest (semiconductor test), KLA Corporation (process control, not direct test), Cohu (smaller test), ABB, Fanuc, KUKA (robotics) |
What Teradyne actually sells
Semiconductor test (largest segment)
The UltraFLEX platform is Teradyne's primary system for testing logic and compute chips: CPUs, GPUs, AI ASICs, mobile SoCs, RF chips and networking ICs. The J750 is an older, lower-cost platform that continues to serve consumer and industrial markets where ultra-high performance is not required. The Magnum platform tests NAND flash and DRAM memory. Semiconductor test equipment is capital equipment with long useful lives (often 5 to 10 years for test systems), meaning demand is lumpy: customers buy in bursts when chip complexity or volume increases require more test capacity, then use existing systems for extended periods before buying again.
Why test time matters for AI chips
The economics of semiconductor test are driven primarily by test time per chip. A more complex chip requires more test patterns to be applied, more functional states to be verified, and more time on a tester before it can be released as a known-good die or tested package. AI accelerators have grown dramatically in die area and functional complexity. NVIDIA's H100 is significantly more complex than A100, with more CUDA cores, HBM memory stacks, and interconnect complexity. Each H100 unit takes materially longer to test than an A100 unit. This means test equipment demand grows not just with unit volumes but with test intensity per unit, making AI chip complexity a genuinely incremental driver of ATE demand.
System test and circuit board test
Teradyne's system test division produces instruments for testing finished electronic systems (notebook computers, network equipment) and circuit boards (in-circuit test). This segment is smaller and less strategically significant than semiconductor test, but it provides some diversification. Demand correlates with consumer electronics and communications infrastructure production cycles.
Industrial automation (Universal Robots and MiR)
Universal Robots, acquired in 2015, makes collaborative robots (cobots) designed to work safely alongside human workers without requiring safety cages. Cobots are targeted at small and medium-sized manufacturers who cannot afford traditional industrial robots' installation and programming costs. MiR, acquired in 2018, makes autonomous mobile robots (AMRs) for material transport on factory floors. Together these businesses represent roughly 15 to 20 percent of Teradyne revenue. They experienced significant demand softness in 2022 through 2024 as industrial companies paused automation investments after over-ordering during the post-pandemic period.
AI chip test as a growth driver
The AI semiconductor test opportunity for Teradyne operates on two levels. First, AI chip unit volumes have grown as hyperscalers build AI training clusters, requiring more chips to be tested in total. Second, each AI chip is more complex and takes longer to test per unit, so even stable unit volumes require more test system capacity.
Hyperscaler custom AI ASICs (Google TPUs, Amazon Trainium and Inferentia, Meta MTIA) add another dimension: each custom AI chip requires its own test program development and potentially unique test hardware configurations. This creates design win opportunities for Teradyne on each new custom silicon program. Winning the test platform for a major hyperscaler custom chip provides multi-year volume visibility as that chip goes into production.
Investors should assess the AI test opportunity against the mobile test cycle. Teradyne's semiconductor test segment has historically been heavily influenced by iPhone production ramps, as Apple's A-series SoCs are among the highest-volume, highest-performance chips tested on Teradyne platforms. A slowdown in smartphone unit volumes or a shift to a competing test platform could offset gains from AI chip test. Understanding the mix between mobile SoC test, AI/HPC test, and memory test within Teradyne's semiconductor test segment is essential to judging the durability of growth.
Key risks
- Semiconductor capex cyclicality: ATE demand tracks semiconductor industry capital spending, which cycles sharply. During semiconductor downturns, chip makers defer test equipment purchases, and Teradyne's revenue declines meaningfully. The ATE cycle can lag the chip production cycle by one to two quarters.
- Smartphone and mobile test concentration: A significant portion of Teradyne's logic test revenue has historically come from high-volume smartphone SoC production. Apple iPhone production ramps drive orders; shifts in Apple's test platform decisions would have material impact.
- Robotics demand weakness: The industrial automation segment has faced cyclical demand softness. Universal Robots cobots are sold to small manufacturers whose investment cycles follow SME capital confidence. A sustained weak industrial environment further defers the robotics recovery.
- Customer concentration: NVIDIA, Apple and major OSAT (outsourced semiconductor assembly and test) companies represent important shares of Teradyne's semiconductor test revenue. Any of these customers reducing test purchases would have outsized impact.
- Technology platform risk: Shifts in chip architectures (such as greater use of chiplets, wafer-level test, or die-to-wafer bonding) could change test requirements in ways that favor different test system architectures or competitors.
How to think about valuation
Teradyne's semiconductor test and industrial automation segments have different growth profiles, margin structures and multiples, and should be analyzed separately before blending. Semiconductor test is the higher-growth, higher-multiple business given AI tailwinds. Industrial automation is more predictable but slower. Normalizing both segments against their respective mid-cycle demand assumptions yields a more defensible blended valuation than using peak or trough multiples.
P/E, price-to-free-cash-flow and EV/EBITDA through the cycle are all useful. Watch for the inflection point at which AI chip test revenue becomes large enough relative to mobile test that the overall semiconductor test revenue cycle smooths. If AI test demand is less correlated with smartphone production cycles than traditional SoC test, Teradyne's revenue volatility could structurally decrease over time.
What to monitor
| Area | What to watch | Why it matters |
|---|---|---|
| AI chip test | AI/HPC as percent of semiconductor test revenue; NVIDIA GPU unit volumes | Measures whether the AI test thesis is materializing in results. |
| Mobile test | iPhone production forecasts; Apple A-series chip complexity comments | Largest historical driver of logic test revenue swings. |
| Memory test | DRAM/NAND capex announcements from Samsung, SK Hynix, Micron | Determines memory test equipment order timing. |
| Robotics recovery | Universal Robots and MiR revenue growth; cobot order intake | Tests whether the industrial automation business has found its floor. |
| Margins and backlog | Gross margin by segment; order backlog visibility | Indicates cycle position and near-term revenue visibility. |
Frequently asked questions
What does Teradyne make?
Teradyne builds automated test equipment (ATE) for semiconductors, printed circuit boards and electronic systems. Its UltraFLEX and J750 platforms test logic and compute chips (CPUs, GPUs, AI ASICs, SoCs), while the Magnum platform tests NAND and DRAM memory. Teradyne also operates an industrial automation division through Universal Robots (collaborative robots) and MiR (autonomous mobile robots).
Is Teradyne in the SOX index?
Yes. Teradyne (TER) is a member of the PHLX Semiconductor Sector Index (SOX). While Teradyne does not design chips, its semiconductor test equipment is a critical part of the semiconductor manufacturing process, and SOX includes companies across the semiconductor value chain including equipment makers.
Why is automated test equipment important for AI chips?
AI accelerators like NVIDIA GPUs are among the largest and most complex chips ever produced. A chip must be electrically verified to function correctly before it ships. Larger, more complex chips take longer to test and require more test insertions, increasing the test time per unit. When NVIDIA introduced H100 (relative to A100), test time per unit increased significantly due to greater die area, HBM memory integration and functional complexity. This drives ATE demand even without unit volume increases, because each chip simply takes longer to test on each test system.
What is Teradyne's robotics business?
Teradyne acquired Universal Robots in 2015 and MiR (Mobile Industrial Robots) in 2018. Universal Robots makes collaborative robots (cobots) designed to work safely alongside humans without safety cages, enabling automation in small and medium-sized manufacturing operations. MiR makes autonomous mobile robots (AMRs) for factory floor material transport. Together these businesses represent roughly 15 to 20 percent of Teradyne revenue and are sold to manufacturers globally. The robotics segment has experienced cyclical demand softness as industrial companies paused automation investments in 2022 through 2024.
How does Teradyne compete with Advantest?
Teradyne and Advantest are the two dominant semiconductor test equipment companies globally. Teradyne holds stronger market share in logic and SoC testing (compute chips, AI ASICs, mobile processors), while Advantest is very strong in memory test (DRAM and NAND) and has also gained logic test share. Both compete for new AI chip test sockets. In mobile SoC test (a major revenue driver tied to smartphone production), Teradyne has traditionally been the leading platform, with Qualcomm Snapdragon and Apple A-series chip test as key reference points.