Technology · Semiconductors · Synthos Deep Dive · 2026-07-03
| Verdict | Hold — systematic Synthos tier |
| Price (2026-07-02) | $369.09 · market cap ~$57.8B · note: −13.6% on the quote-day print |
| Synthos scores (0–10) | Downside Risk 7 · Growth Quality 7 · Exponential Potential 7 |
| Synthos fair value (base case) | ~$400 → +8% · full range $250 (bear) – $560 (bull) |
| Street consensus | $385.73 (high $550 / low $220; 20 Buy · 11 Hold · 0 Sell) — context, not our anchor |
| Valuation | ~68× trailing EPS · ~49× FY26E · ~36× FY27E · ~27× FY28E · EV/S 15.2× · EV/EBITDA 52× TTM |
| Exponential Potential | 7/10 · High — FY26E revenue +42% and EPS +114%, second derivative positive on AI-compute test; but this is a semi-cap cycle, not a straight secular line |
| Technicals | Mixed — $369, −24% off the 52-wk high, above 200-DMA but below 50-DMA, RSI 48, +302% 12-mo (SPY +21%) |
| Conviction | Low — 1 net-bullish voice (Jordi Visser, skill 2.0, conviction 65), 1 reconciled claim; not a breadth call |
| Position sizing | Satellite-only if at all, ≤1–2%, scale on cyclical pullbacks — not a core holding |
| Next catalyst | 2026-07-29 Q2'26 earnings (Street EPS $2.04, revenue ~$1.22B) |
| Single biggest risk | Semi-cap cyclicality + customer/geo concentration — a demand air-pocket after the AI-test surge |
One-line thesis. Teradyne is the #2 name in the automated-test-equipment duopoly, riding a genuine AI-compute and high-bandwidth-memory test surge (FY26E revenue +42%, EPS +114% estimated) with a net-cash balance sheet — but the stock has already tripled in twelve months, trades at ~49× FY26E earnings with a 1.79 beta, and the whole call rests on an inherently cyclical, concentrated end-market, so we rate it Watch rather than chase.
Teradyne makes the machines that test computer chips before those chips ship. Every advanced AI processor and memory stack has to be tested, and as AI chips get more complex, they need more (and more expensive) testing — that is the tailwind driving the business right now. Teradyne also owns a robotics arm (collaborative robots and warehouse robots).
The business is genuinely good: it is very profitable (keeps about 59 cents of gross profit per sales dollar), has more cash than debt, and its earnings are expected to roughly double next year. The catch: the stock has already tripled in the past year, so a lot of good news is baked in. At today's price you are paying about 49 times next year's expected profit — expensive — and this industry runs in boom-and-bust cycles, so the next down-leg can be brutal.
Our verdict is Watch: put it on the list, but the price today does not leave much cushion if the AI-test boom cools. Here is what our three scores mean in everyday terms:
The one big worry: this is a cyclical, customer-concentrated business (a lot of revenue flows through a handful of big Asian chipmakers). If AI-chip test demand cools even for a couple of quarters, both the earnings and the rich multiple can fall at the same time.
Solid = price · dashed = 50-day average · dotted = 200-day average · amber = 52-week high/low. Price above both averages is an uptrend.
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Above 70 (red band) = overbought, below 30 (green band) = oversold. Currently 44.
Blue crossing above amber (bars flip green) = momentum turning up; below (bars red) = turning down. Bar height = the size of that gap.
Solid = TER · dashed = S&P 500 · dotted = XLK (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
“Analog-based semiconductor names like Teradyne and Lattice haven't participated yet and have a long road ahead.”
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
Teradyne (NASDAQ: TER), founded 1960, headquartered in North Reading, MA, is a global leader in automated test equipment (ATE) — the systems that test semiconductors at wafer and packaged-device stages — plus an industrial-automation (robotics) arm. It reports across four lines:
1. Semiconductor Test — the core: FLEX, J750, Magnum (memory), ETS (analog/mixed-signal) platforms serving IDMs, fabless designers, foundries, and OSATs (outsourced assembly/test).
2. System Test — defense/aerospace instrumentation, storage test, PCB inspection.
3. Industrial Automation — Universal Robots (collaborative arms) and MiR (autonomous mobile robots).
4. Wireless Test — the LitePoint brand (Wi-Fi, 5G, connectivity test).
Fiscal year ends late December. CEO Gregory Smith. ~6,500 employees.
Revenue mix (FY2025, from FMP segmentation):
The forward story the numbers tell: after a soft 2023–24 (revenue troughed vs the 2021 peak of $3.70B), Teradyne is inflecting hard on AI-compute SoC and high-bandwidth-memory (HBM) test intensity — the FY26E revenue jump to ~$4.54B (+42% estimated) is the whole bull case.
Honest disclosure: expert coverage in the Synthos KB is thin. There is 1 traceable claim (total_claims = 1), so this is not a breadth-conviction name — the verdict below is fundamentals- and quant-driven, not panel-driven.
jordi_visser_m-jJvVd29aY-4:bd3baee54f, 2026-02-05): "Analog-based semiconductor names like Teradyne and Lattice haven't participated yet and have a long road ahead." This is a catch-up / long-runway call, categorized under analog and semiconductors. It is dated early February 2026 — before the subsequent 12-month surge in the stock, so the "haven't participated yet" framing is now partly stale (TER is +302% over twelve months). Treat it as directional support, not a live price view.Net weighted conviction is +0.65 (single voice, skill- and recency-weighted) — enough to note, nowhere near enough to headline. Anyone underwriting TER should do so on the numbers and the cycle, which is exactly how we score it below.
The one-glance judgment — three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):
| Score | 0–10 | The read |
|---|---|---|
| Downside Risk (lower = safer) | 7 · Elevated | Net-cash balance sheet (net debt $53M, net-debt/EBITDA −0.14×) is a genuine cushion, but beta 1.79, a cyclical semi-cap end-market, Taiwan/OSAT concentration, and ~49× FY26E after a 3× 12-month run stack the downside. |
| Growth Quality | 7 · Strong | FY26E revenue +42% and EPS +114% (estimated), 59% gross margin, ROE 30%, ROIC 26% — but the revenue line has been lumpy and cyclical (peak $3.70B FY21 → trough → recovery), so it is not a smooth compounder. |
| Exponential Potential | 7 · High | The second derivative is positive — growth is accelerating on AI-compute/HBM test intensity — and at $58B cap there is room vs a large test + robotics TAM. Docked from higher because this is a cycle, not a secular straight line. |
The three cases (our own scenario model — assumptions shown; each target is a ~12–18-month fair value). We deliberately do not attach probabilities: the base case is by definition the expected path, so a weighted blend would just restate it with false precision. Instead the cases bound the range, and the scores above summarize them.
| Case | Key assumptions | Fair value |
|---|---|---|
| Bull | AI-compute + HBM test demand runs harder and longer; robotics inflects; FY27E EPS beats to ~$12 (vs $10.17 cons) and the market keeps paying a peak-cycle ~46×. | ~$560 (+52%) |
| Base (our anchor) | Estimates roughly hit — FY26E EPS $7.46, FY27E $10.17; a cyclical grower earns a ~39× FY27E multiple as the cycle is mid-stream. | ~$400 (+8%) |
| Bear | AI-test air-pocket / digestion; a semi-cap down-leg pulls FY27E EPS toward ~$8 and the multiple de-rates to a mid-cycle ~30×. Earnings and multiple fall together. | ~$250 (−32%) |
Synthos fair value = the base case, ~$400 (+8%), with the full $250–$560 span as the honest range. Note how tight the base upside is: at today's price the reward is modest and the cyclical bear is a −32% drawdown — an asymmetry that argues for patience, not chasing. Our base sits essentially on top of the Street's $385.73 consensus (the Street high is $550, low $220 — a very wide band that reflects exactly this cyclical uncertainty). This is a tracked call — the Forecaster Scorecard grades it once it matures.
Synthos separates compounders (durable high returns on capital) from exponentials (accelerating, multi-baggers-from-here). TER is a cyclical grower in an accelerating phase — the classic "exponential-looking" mid-cycle:
Exponential Potential: High (7/10). The rare combination of accelerating growth + real room + net-cash funding earns a 7 — meaningfully above a decelerating megacap. It is not an 8–9 because the acceleration is cyclical, not secular: own it for the up-leg with clear eyes that the same second derivative reverses on the way down.
There is no way to call TER cheap: ~68× trailing EPS (flattered by a strong Q1'26), ~49× FY26E, ~36× FY27E, ~27× FY28E, EV/sales 15.2×, EV/EBITDA 52× TTM. The letter rating agrees — FMP's model scores price-to-earnings and price-to-book at 1/5 (richest possible) even while ROE/ROA score 5/5. The bull's defense is the same as every accelerating cyclical: earnings grow into the multiple — if FY26E–28E estimates hit, the forward P/E compresses from ~49× to ~27× at a flat price. The risk is that you are paying a peak-ish multiple on peak-ish (cyclical) earnings, and if the cycle rolls both compress together. A reverse read: at $369 the market is pricing continued high-30s%/20s% growth with little room for a demand stumble. Street targets (context): consensus $385.73, high $550, low $220 — the ~2.5× spread from low to high is itself the tell that this is a wide-outcome, cyclical name. Not a value buy; an accelerating-cyclical-at-a-full-price name where entry timing matters.
Teradyne's moat is its position in the automated-test-equipment duopoly — it and Japan's Advantest split the leading-edge semiconductor-test market, with high switching costs (test programs, correlation data, and customer qualification lock in incumbents for years) and a large installed base pulling recurring service revenue (17% of sales). The AI-compute and HBM wave raises test intensity per device, which structurally favors the two incumbents. The robotics arm (Universal Robots, MiR) is a genuine optionality leg but has been a slower, more competitive grind. Threats: the cyclicality of chip capex, customer concentration (a handful of large foundries/OSATs, Taiwan-heavy), and Advantest's competitive push in SoC test.
Peer set (from FMP, market cap): the file's "peers" are a loose semiconductor/test basket rather than direct ATE comps — Astera Labs $70B, ASE Technology (OSAT) $92B, Keysight $54B (the closest test-and-measurement comp), Microchip $46B, ON Semi $36B, STMicro $61B, United Microelectronics (foundry) $61B, Sandisk $258B, Super Micro $18B, Wipro $20B. TER's truest competitor — Advantest — is not in this list; note that when reading the basket. Against these, TER commands a rich multiple justified only if the AI-test growth persists.
Thesis tripwires (what would change the call): a book-to-bill dropping below 1.0 for two consecutive quarters; AI-test order digestion language on the call; a break and hold below the 200-DMA (~$268); or FY27E estimates being cut — any of which would move this from Watch toward Avoid on valuation, or toward Buy — Tactical only after a meaningful de-rating.
Watch. Teradyne is a genuinely good business at the right point of a real AI-test acceleration — net-cash balance sheet, 59% gross margin, FY26E revenue +42% and EPS +114% (estimated), and support (if dated) from our highest-skill KB voice. But the stock has tripled in twelve months, trades at ~49× FY26E with a 1.79 beta in an inherently cyclical, customer-concentrated market, and our base-case fair value (~$400) is only +8% while the cyclical bear is −32%. That asymmetry, plus thin expert breadth (1 claim), argues for the watchlist, not the buy button, at today's price.
claim_id (cited inline). Fabricated conviction is structurally impossible (claim-ID reconciliation). Thin coverage is disclosed, not papered over; the verdict is fundamentals/quant-driven.